Lines and growth conditions

Arabidopsis thaliana lines used in this study are listed in S1 Table. Mutant lines were selected via the TAIR and NASC web sites [16]. EMS mutagenesis was carried out on Landsberg erecta (Ler) seed as described [17]. Seed were surface sterilized, stratified at 4°C, and sown on Murashige and Skoog (MS) medium supplemented with B5 Vitamins (Merck group; https://www.sigmaaldrich.com). For nutrient stress conditions, NPK media was used. Plates were incubated under controlled growth chamber conditions (22°C, 80 μmol m^-2s^-1). 10-day-old plate grown seedlings were used for organ length measurements and for scanning electron microscopy, and six or seven-day-old root tips were used for light microscopy. See S1 and S2 Methods.

Screen conditions

For all screen conditions, seed were germinated on nutrient or nutrient-deficient medium without a carbon source (see S1 Fig). Water deficit plates were prepared with PEG-6000 (Merck group; https://www.sigmaaldrich.com) as described in S3 Method. Plants were grown under optimal conditions (16: 8 hr light: dark photoperiod, 22°C, 180 μmol m^-2s^-1) at the TUMmesa ecotron [18] and the sterilized seed were imbibed at 4°C in the dark for 7 days (prior to plating) to break dormancy. Initial screen conditions compared growth on full-strength MS medium, with or without PEG-6000 for an additional -0.4 MPa. As the nutrients in MS medium generate a pressure of -0.2 MPa, initial screen conditions thus compared -0.2 MPa and -0.6 MPa. For optimized screen conditions, we later replaced full-strength MS by half-strength (½) MS medium, which corresponds to 0 MPa; thus, optimized screen conditions compare 0 MPa to -0.4 MPa. Under both initial and optimized screen conditions only the roots were exposed to the medium and, thus, to water stress; this was initially achieved via cutting a window in the agar and later via a plastic strip between the medium and the shoots. A final optimization was to incline the plates to promote root growth on the surface of the agar as opposed to in the agar. In the tables and graphs, we refer to initial versus optimized screen conditions to most accurately describe how the measurements were carried out. For growth under different light qualities blue, red or far-red LEDs (Quantum Devices) were employed. See S3–S5 Methods.

Molecular methods

Mapping, positional cloning and allele sequencing were carried out as described in Jaber et al., 2010 [19] using primers tabulated in S6 Method. qPCR was carried out as described in S7 Method.

Light and electron microscopy

For scanning electron microscopy, a Zeiss (LEO) VP 438 microscope was operated at 15 kV. Fresh seedlings were placed onto stubs and examined immediately in low vacuum. Confocal microscopes used for imaging were an Olympus (www.olympus-ims.com) Fluoview 1000 confocal laser scanning microscope (CSLM) and a Leica (www.leica-microsystems.com) SP8 Hyvolution CSLM. 40x and 60x water immersion 0.9 numerical aperture objectives (Olympus) were used. FM4-64 staining was as described in Ravikumar et al. [20], with emission at 640 nm. Root apical meristem properties are described in S8 Method.

For GUS staining, seed were plated under initial screen conditions (MS salts) and 7- or 10-day-old seedlings were used. Seedlings were fixed for 20 min in 90% acetone, washed with staining buffer (10 mM EDTA, 0.1% Triton-X 100, 2 mM potassium ferrocyanide, 2mM potassium ferricyanide, 50 mM sodium-phosphate buffer (pH 7)) on ice and transferred into staining buffer containing additionally X-Gluc (2.31 mM) and chloramphenicol (100 mg/ml). Seedlings were incubated at 400 mbar for 15 min on ice before being incubated for 3–5 h at 37°C. Seedlings were dehydrated in an ethanol series (20%, 35%, 50%, 30 min per concentration) and finally fixed for 30 min at room temperature in FAA (50% ethanol, 3.7% formaldehyde, 5% acetic acid). Micrographs were recorded with an Olympus BX61 microscope. A 40x objective (water) was used.

Image processing, data and statistical analysis

Shoot (hypocotyl) and root lengths were scored with the Image J free-hand tool (https://imagej.nih.gov). Images were processed with ImageJ, Adobe photoshop (www.adobe.com) and assembled with Adobe Illustrator. The hypocotyl volume was computed by assuming a cylindrical organ shape as follows: V = πWL, where W is the width and L the length of the hypocotyl.

Root and hypocotyl responses to water deficit in the dark (abbreviated as darkW) were computed as

And for the ratio:

Due to variability between PEG lots and PEG plates (see text), we normalized each mutant to the corresponding wild-type ecotype (see S1 Table) on the same plate. Thus, the normalized ratio adjustment to water stress in the dark (darkW) was computed as follows:

RQ_hypocotyl and RQ_root as well as light versus dark responses were computed in a similar fashion. The mean RQ_ratio of at least three biological replicates (i.e. the seed stocks from different mother plants) is shown. Responses were considered to be attenuated for RQ_ratio < 0.8, normal in the 0.8–1.2 range, and exaggerated for RQ_ratio > 1.2. For volcano plots we plotted the mean RQ_ratio on the X-axis and the median P_ratio-value on the Y-axis. P-values were computed with the Student’s T-test when the two populations had equal variances and with the Welch’s T-test when variances were unequal. That distributions were normal was verified with the Shapiro-Wilk test for selected samples, and in the rare cases where normality was not clear, we applied the non-parametric Mann-Whitney-U tests. The computations were carried out in excel for graphic rendition and verified in R. For multiple testing, as carried out when different wavelengths and light intensities were compared, we applied a Benjamini–Hochberg correction. This is specified in the legend where applicable. Responses were considered to be insignificant for P-values ≥ 0.05 and attenuated for P-values ≥ 0.00001. The median P-value for at least three replicates is shown in the volcano plots.

Hypocotyl growth in search of light is prioritized over root growth in search of nutrients in the young seedling

To understand decision making processes in the Arabidopsis seedling, we set up a “conflict of interest” scenario between shoot and root growth. To this end, seed were germinated under growth conditions designed to place contradictory demands on hypocotyl (growth at low light intensities or in the dark) or root growth (nutrient deficiency or water stress). We first tested several nutrient media for their ability to promote trade-offs between hypocotyl versus primary root growth in dark-grown seedlings; these include -P, -N, and -K media lacking phosphate, nitrate, or potassium (S1 Fig). We also tested low levels of osmotic stress (100 mM mannitol; S2 Fig) as well as salt stress (50–100 mM NaCl; S2 Fig), which has been shown to impair hypocotyl elongation in response to far-red light [21]. None of these media, however, gave rise to considerable or reproducible trade-offs between root and hypocotyl growth, by which we refer to the growth of one organ at the expense of another (S1 and S2 Figs). There was a clear priority for hypocotyl growth in search of light over primary root growth in search of nutrients (S1D Fig). It is to be noted in this context that we are looking not at root architecture but exclusively at primary root growth and that the seed and embryo contain sufficient nutrients to support the initial phases of seedling growth [22].

Opposing gradients of water stress and light intensity enable us to visualize shoot versus root growth trade-offs in the seedling

We next attempted to identify what resources might be as important to a germinating seedling as the light. Here, we explored the relative importance of water. To this end, we germinated seedlings in the presence of water stress, using polyethylene glycol (PEG) to withdraw available water from the medium in a standardized tissue culture setting [23,24]. In the light, PEG significantly decreased hypocotyl growth (P_hypocotyl = 3E^-48), but this was not mirrored by an increase in root growth; root growth was, in fact, slightly reduced (P_root = 1.3E^-02; S3A Fig). In the dark, increasing degrees of water stress considerably and reproducibly increased root length and decreased hypocotyl length (three-fold change for the hypocotyl and almost four-fold change for the root for -0.7 MPa compared to -0.2 MPa; Figs 1A–1E and S3B–S3D). In the light, water stress decreased the total length of the seedlings (P_total = 8.2E^-4; S3A Fig). In the dark, however, water stress did not impact the total length of the seedling (for ≤ -0.6 MPa compared to -0.2; Figs 1E cf. S3A). In comparison to water-stress in the light, we were clearly observing trade-offs between hypocotyl and root growth in response to a gradient of water stress in dark-grown seedlings. We then tested a gradient of decreasing light intensity. Under low light, the trade-off was more pronounced, with longer roots and shorter hypocotyls than in the dark, even when the light intensity was as low as 2 μmol m^-2 s^-1 (Fig 1F–1I). Similarly, for both blue and red light, hypocotyl length increased whereas root length decreased with decreasing light intensity and this trend was enhanced by water stress (-0.6 MPa; S4A and S4B Fig). Far-red light showed a comparable trend (S4C Fig). In conclusion, using opposite gradients of decreasing light intensity versus increasing water stress, we were able to fine tune root growth at the expense of hypocotyl growth.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 1. Hypocotyl versus root growth in response to light and water availability.

Wild type (Col-0). (A-E) germination in the dark on a gradient of water stress ranging from 0 to -0.7 MPa. Recordings were taken 10 days after germination. (A) scanning electron micrographs; arrows point to the hypocotyl/root junction and arrowheads to the end of the root. (B-E) panels are from the same experiment. Water stress applied in the dark increases root length (C) at the expense of hypocotyl length (B), giving rise to a decrease in the hypocotyl/root ratio (D). Note that the total seedling length does not vary at -0.2 MPa versus -0.6 or -0.7 MPa (E). Thus, there was a clear tradeoff, by which we refer to the growth of one organ at the expense of another. (F-I) germination on MS medium under white light at varying intensities ranging from 40 to 0 μmol m^-2 s^-1, with or without -0.6 MPa water stress. A decreasing light intensity gradient increases hypocotyl length (F) at the expense of root length (G). The hypocotyl/root ratio was calculated (H). The number (n) of seedlings measured per condition is in grey below the mean ±StDev bar graphs. P-values were computed with a two-tailed student’s T-test and are represented as follows: *: 0.05–0.01; **: 0.01–0.001; ***: 0.001–0.0001; ****: 0.0001–0.00001, *****: < 0.00001. See related S1, S2, S3, S4 and S5 Figs.

https://doi.org/10.1371/journal.pgen.1010541.g001

We next asked whether the reduction in hypocotyl length observed in response to water stress applied in the dark was accompanied by a change in organ width and, ultimately, volume. To this end, seedlings were imaged by scanning electron microscopy (Fig 1A) and the lengths and widths of the hypocotyls measured (S5A and S5B Fig). In a comparison between light to dark, the hypocotyl length and volume increased despite a decrease in hypocotyl width (S5A–S5C Fig). When water stress was applied to dark-germinated seedlings, the reduced length of the hypocotyl was not accompanied by an increase in width; in fact, a slight reduction in width was observed and overall the volume of the hypocotyl was two-fold reduced (S5A–S5C Fig). The change in organ volume suggests that resources such as water, which accounts for organ volume to a large extent, are being differentially allocated in response to light and water stress. In summary, hypocotyl and root lengths and the ratio thereof are exquisitely fine-tuned to the wavelength and intensity of the light source, and to the severity of water stress. In the case of the seedling, what we observe is a clear consistency in a continuum of preferred outcomes along a gradient in response to opposing gradients of light intensity and water stress.

A forward genetic screen for “decision” mutants identifies BRASSINOSTEROID INSENSITIVE 2 (BIN2)

The fine-tuning of hypocotyl/root ratios can conceptually be broken down into four steps: (i) sensing, (ii) downstream signalling, (iii) decision making processes, and (iv) the execution of these decisions (the action). The gradients of hypocotyl and root growth we describe in response to our screen conditions can help distinguish between mutants impaired in the process of decision making per se, versus mutants with primary defects in one of the other steps (sensing, signalling or execution). Perception mutants would fail to perceive light or water stress; a good example of this is the cry1 cry2 phyA phyB quadruple photoreceptor mutant [25], which had a severely impaired light response (S4F Fig), but a “normal” response to water stress in the dark (S4G Fig). In contrast, execution mutants may have aberrantly short hypocotyls or roots that are nonetheless capable of differentially increasing in length depending on the stress conditions. Decision mutants would differ from perception or execution mutants as they would clearly perceive the stress factors yet fail to adequately adjust their hypocotyl/root ratios in response to a gradient of multiple stress conditions. Failure to adjust organ lengths would be seen as a non-significant response, or as a significant response but in the wrong direction as compared to the wild type. We thus used organ lengths, the hypocotyl/root ratio and the significance of the responses as decision read outs. We specifically looked for mutants in which at least one organ exceeded wild-type length under darkW. To operate under limiting conditions, we germinated seedlings in the dark on media lacking a carbon source (see S1 Fig), avoiding even low levels of light.

To identify genes implicated in decision making, we performed a genetic screen in two consecutive steps (Fig 2A). In the first step, we germinated seed in the dark. In the second round, viable mutants with aberrant hypocotyl to root ratios in round 1 were rescreened for their ability to adjust their hypocotyl to root ratios in response to water stress in the dark. We initially screened in the dark because the high variance in root growth under water deficit in the dark in the wild type (see below) would obscure the distinction between putative mutants versus stochastically occurring wild-type seedlings with short roots under darkW. 83000 EMS-mutagenized M2 seed were subjected to the first screen and over 100 viable mutants rescreened in the dark in the presence or absence of water stress. 19 viable seedlings with aberrant hypocotyl to root ratios in our forward genetic screen had brassinosteroid-related dwarf phenotypes (Figs 2B, S6A, and S6C). Of these, one—named B1—had a very pronounced inability to adjust its hypocotyl/root ratio in response to multiple stress conditions: while the root response to water stress in the dark was only slightly attenuated (P_root = 6E^-05; Fig 2C), the hypocotyl response was aberrant (P_hypocotyl = 0.05; Fig 2C) and the ratio adjustment insignificant (P_ratio = 0.83; Fig 2C). Positional cloning and allele sequencing suggested that the B1 locus encoded BRASSINOSTEROID-INSENSTIVE 2 (BIN2; Fig 2D and 2E). Our B1 allele carries a TREE domain E263K mutation, identical to bin2-1 and shown to block the BR-induced ubiquitination and degradation of BIN2 [26]. Segregation analysis showed that our B1 bin2 allele was semi dominant, like all TREE domain mutations of bin2-1. We sequenced 19 F2 individuals of the segregating mapping population and found that the E263K point mutation absolutely segregated with the phenotype (S2 Table). For complementation analysis, we crossed our B1 mutant with a known bin2-1 allele [27]; F1 individuals had phenotypes on soil that were characteristic of homozygous bin2 plants and, upon sequencing, exhibited EMS-induced G to A transitions at position 989 for both the B1 mutation and the bin2-1 allele (2E Fig). We complemented B1 with bin2-1 and ucu1 alleles and compared it to bin2-1, ucu1 and dwarf12 [28,29] alleles at the BIN2 locus; these three published mutant lines exhibited the same behaviour as B1, including semi-dominance and partial etiolation. Indeed, the B1 phenotype was similar to bin2-1 under our multiple stress conditions (S6B Fig; compare B1 in Fig 2C to bin2-1 in Fig 3D). In conclusion, positional cloning, allele sequencing, segregation analysis, complementation analysis and phenotypic analyses show that the B1 locus encodes BIN2.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 2. Identification of BIN2 and its role in hypocotyl/root tradeoffs.

(A) A screen for decision mutants; two consecutive steps are depicted (see text). (B) Dwarf mutants with curled-in leaves, a BR-related phenotype, identified in our forward genetic screen on the basis of altered hypocotyl/root ratios in the dark. Scale bars = 1cm. (C) B1 mutants fail to adjust their hypocotyl/root ratios in response to water stress in the dark (darkW). Note that hypocotyls had an inverse response (significance marked with red asterisk). The number (n) of seedlings measured per condition is in grey below mean ±StDev bar graphs. P-values were computed with a two-tailed student’s T-test and are represented as follows: *: 0.05–0.01; ***: 0.001–0.0001; ****: 0.0001–0.00001, *****: < 0.00001. (D) Fine mapping of mutant B1 identifies a 100 kb interval on chromosome 4, which spans the BIN2 locus. Markers used for mapping are depicted in abbreviated form above the line and in full detail in S6 Method. Rec: recombinants. See S6 Method for mapping and S2 Table for segregation analyses. (E) Sequencing of the BIN2 TREE domain in F1 segregants of a complementation cross between bin2-1 and B1 shows that B1 harbors a G to A transition at position 989 (yellow highlight), giving rise to an E263K mutation identical to that of bin2-1. n = 10 F1 plants were sequenced. See related S6 Fig.

https://doi.org/10.1371/journal.pgen.1010541.g002

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 3. Role of BR signalling in hypocotyl/root tradeoffs.

Seedlings were germinated on ½ MS in the dark (dark) or in dark with -0.4MPa water stress (darkW). (A) Col-0 (wild type). (B) BR biosynthesis mutant cpd. (C) BR perception mutant bri1brl1brl3 (a segregating triple null). (D) BR signalling mutant bin2-1 (a semidominant gain of function allele). (E) bin2-3bil1bil2 tripple knockout; note that, in stark contrast to the wild type, the bin2-3bil1bil2 tripple knock-out has shorter roots under darkW than in the dark (red asterisks). (F) Transcription factor mutant bzr1-1D, a dominant allele. (G) BIN2 bypass mutant bri-116 bzr1-1D. Note that bin2-1 mutants have a severely attenuated hypocotyl response (D), bin2-3bil1bil2 mutants have an inverse root response (red asterisks denote a significant response in the opposite direction to the wild type) and no ratio response (E), and bri-116 bzr1-1D severely attenuated hypocotyl, root and ratio responses (G). Null alleles are depicted in regular font, semi-dominant or dominant in bold and higher order mutants are underlined. At least 3 experiments were performed for each line, and a representative one is shown here on the basis of RQ and P values (see Fig 4E and 4F). The number (n) of seedlings measured per condition is in grey below the mean ±StDev bar graphs. P-values were computed with a two-tailed student’s T-test and are represented as follows: *: 0.05–0.01; **: 0.01–0.001; ***: 0.001–0.0001; ****: 0.0001–0.00001, *****: < 0.00001. For mean RQ values and median P-values see Figs 4E, 4F, and S8. Mutant alleles and the corresponding ecotypes are described in S1 Table. See related S6, S7, S8, S9, S10 and S11 Figs.

https://doi.org/10.1371/journal.pgen.1010541.g003

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 4. Responses of BR pathway mutants to water stress in the dark: violin plots, response quotients and volcano plots.

(A-C) Violin plots of the hypocotyl (A), root (B) and ratio responses (C) of the triple bin2-3bil1bil2 knock out line shown in Fig 3E, with the wild-type ecotype (Ws-2) as reference. The dot represents the mean and the line the 95% confidence interval. Note the high variance of the wild-type root response under darkW. bin2-3bil1bil2 mutants qualified as decision mutants on 3 counts: (i) failure to adjust the hypocotyl/root ratio to darkW (the ratio for darkW is the same as for dark in panel C), (ii) low or non-significant P-value (see panel f below) and (iii) one organ (here the hypocotyl in panel a) exceeded wild-type length under darkW. (D) Color coding of steps in the BR signalling pathway; 1: biosynthesis, 2: perception, 3: signalling, 4: transcription. See text for further detail. (E) RQ_ratio response quotient of the hypocotyl/root ratios under dark/darkW conditions, normalized to the wild-type ratio quotient; a value of 1 (vertical red line) indicates that the response to a shift from dark to darkW is similar to that of the respective wild-type ecotype. Each replicate is represented by a dot; purple dots are for initial and grey dots for optimized screen conditions; blue dots are for data from SEM measurements. Note that the triple bin2bil1bil2 knock out has the strongest phenotype, followed by bri1-116 bzr1-1D and bin2-1. (F) Volcano plot with the mean RQ_ratio depicted in (E) on the X-axis and the median P-Value of the response on the Y-axis (negative log scale; a median of all replicates was used). Mutants in the lower left quadrant are considered to have a “confused decision phenotype” (see text). Only bin2 alleles (magenta arrows) or bypass mutants map to this quadrant. The area shaded in grey on the red line (green arrow) is where wild-type ecotypes would theoretically map onto the plot. Mutant alleles and their ecotypes are described in S1 Table. Null alleles are depicted in regular font, semi-dominant or dominant in bold and higher order mutants are underlined. See related S6, S7, S8, S9, S10 and S11 Figs.

https://doi.org/10.1371/journal.pgen.1010541.g004

The BR pathway is implicated in hypocotyl versus root trade-offs in the Arabidopsis seedling

Brassinosteroids are known to be involved in the response to abiotic stress cues such as drought and salinity [21,30,31]. To analyse the role of BR signalling in decision making processes, we studied a set of known mutants impaired in BR biosynthesis, perception, signalling or in BR-responsive gene expression (S1 Table). In addition to bar graphs representing hypocotyl and root lengths (Fig 3), the distribution of datapoints was represented by violin plots (Figs 4A–4C and S7). The violin plots compare organ length distributions in mutants versus the corresponding wild-type ecotype, which depicts dwarfism in some brassinosteroid mutants. It is also apparent that wild-type (Col-0) root length varies under water-deficit in the dark (S7 Fig). Although we have optimized protocols for PEG plates to the best of our ability, there is still a lot-to-lot and plate-to-plate variation. This emphasizes the need for normalizing each mutant line to its corresponding wild-type ecotype on the same (PEG) plate in the same experiment. To this end, the response to water stress in the dark was represented as a normalized response quotient (RQ), which is an indication of how much the mutant deviates from the corresponding wild type (Fig 4E; see methods). We used the normalized ratio response RQ_ratio as our main “decision” readout (Fig 4E) and show RQ_hypocotyl and RQ_root in the supplement (S8 Fig). This is because water stress in the dark is a “conflict of interest” scenario in which hypocotyl and root growth have competing interests (see Fig 1). We reason that mutants unable to integrate environmental cues might have a “confused phenotype” under our multiple stress conditions. The read out for a “confused phenotype” would translate into erratic (i.e. highly variable) hypocotyl versus root lengths. This high variance would, in turn, translate into a low signal to noise ratio, and this can be seen as a high P-value. We, therefore, plotted the median P-values against the normalized response quotients (referred to as volcano plots; mean RQ_ratio in Fig 4F). Wild-type ecotypes invariably had low P values < 10⁻¹⁰ (grey shading on the red line in Fig 4F, green arrow). Mutants with high P-values and low response quotients would be considered “confused” and these would map in the lower left quadrant of the RQ_ratio volcano plot (response to water stress in the dark, Fig 4F).

In the BR pathway (Fig 4D), we first looked at cpd, a BR biosynthesis mutant. CPD (CONSTITUTIVE PHOTOMORPHOGENIC DWARF) encodes a cytochrome P450 monooxygenase involved in the C6 oxidation pathway of brassinolide biosynthesis [32]. cpd mutants are dwarfs with attenuated but significant responses to water stress in the dark (Figs 3B, S7, 4E, 4F, and S8). We then turned to higher order mutants of the BR receptor, BRI1, and its homologues, bri1-116 brl1 brl3 [33]. bri1-116 brl1 brl3 mutants had an attenuated response to water stress in the dark; however, as for cpd, all trends were similar to the wild type and all responses significant (Figs 3C, S7, 4E, 4F, and S8). An allele of the BAK1 coreceptor, bak1-1 [34], showed fairly similar responses to a hypomorphic BRI1 allele, bri1-6 (Fig 4F).

When BR binds to the extracellular domain of the BRI1 receptor, the BIN2 BRASSINOSTEROID INSENSISTIVE 2 shaggy-like kinase is inactivated (Fig 4D). We assessed the phenotype of the semi-dominant bin2-1 allele, which has an identical amino acid change (TREE to TRKE) in the tree domain as our bin2 B1 allele. For bin2-1, the response to water stress in the dark was severely impaired (Figs 3D and S7), with considerably attenuated hypocotyl and hypocotyl/root ratio responses (mean RQ_hypocotyl = 0.65—see magenta arrow in S8A Fig; mean RQ_ratio = 0.57 in Fig 4E; median P_ratio = 0.02—see magenta arrow in Fig 4F). bin2-1 seedlings are dwarfs with severe and pleiotropic phenotypes [27], including a reduction in seed size. We, therefore, turned to higher order null alleles. The Arabidopsis genome encodes ten shaggy-like kinases, of which three in the BIN2 clade have been shown to function redundantly [35]. A triple knock-out of all three clade 2 shaggy like kinases, bin2-3bil1bil2, has a large stature, short roots and an overall phenotype characteristic of plants with enhanced BR signalling outputs [35]. Seed weight in bin2-3bil1bil2 did not differ from the wild type (S9A Fig). This mutant was severely impacted, with a non-significant (median P_root = 0.44; mean RQ_root = 2.42 (magenta arrow in S8B Fig) or aberrant root response to water stress applied in the dark (Figs 3E, 4E, and 4F). bin2-3bil1bil2 mutants fit the above definition of decision mutants as they have a significant root response but in the wrong direction as compared to the wild type, as denoted by red asterisks (Fig 3E). The hypocotyl/root ratio response was also non-significant (median P_ratio = 0.19; magenta arrow in Fig 4F) and severely impaired (mean RQ_ratio = 0.39; Fig 4E). Whereas the hypocotyl response was severely impaired in the semi-dominant bin2-1 allele (magenta arrow in S8A Fig; cf. green arrow for bin2-3bil1bil2), the root response was severely impaired in the loss of function bin2-3bil1bil2 line (magenta arrow in S8B Fig; cf. green arrow for bin2-1). This shows opposite phenotypes in gain- versus loss- of function bin2 alleles. In brief, both semi-dominant and recessive mutations in clade 2 shaggy-like kinases impair differential growth responses to light and water stress in the seedling.

The active BIN2 kinase phosphorylates BZR and BES transcription factors, which reduces their DNA-binding activity, excludes them from the nucleus and targets them for degradation [36]. By inactivating BIN2, BR signalling results in the accumulation of unphosphorylated BZR1/2 in the nucleus and a concomitant expression of BR-target genes. We assessed the behaviour of a dominant BZR1 allele, bzr1-1D [37], which is, like bin2-3bil1bil2, characterized by enhanced BR signalling outputs [35]. Seed weight was not impacted in bzr1-1D (S9B Fig). With respect to water stress in the dark, the root and hypocotyl/root ratio responses were significant and fairly similar to the wild type (mean RQ_root = 1.17; mean RQ_ratio = 0.89: Figs 3F, 4E, 4F, and S8B). A BES1 dominant allele, bes1-1D [38], had a weaker phenotype than bzr1-1D in that it did not exhibit a considerable difference from the wild type in terms of its response to water stress in the dark (mean RQ_ratio = 0.97; median P_ratio = 2E^-11; Fig 4E and 4F).

Looking at BR pathway mutants suggests that BR signalling is implicated in hypocotyl versus root growth trade-offs in the Arabidopsis seedling. To address this hypothesis, we looked at the bri1-116 bzr1-1D double mutant [37], in which a null non-viable BRI1 receptor mutant is partially rescued by a dominant BZR1 transcription factor allele. In this line and downstream of BIN2, BZR1 is constitutively active regardless of the BR signal, which is effectively bypassed. bri1-116 bzr1-1D double mutants had a severely attenuated hypocotyl response to water stress in the dark (mean RQ_hypocotyl = 0.58; median P_hypocotyl = 0.01; Fig 3G; magenta arrow in S8A Fig). Similarly, the double mutants had very short roots in the dark that elongated erratically (RQ_root = 1.43; non-significant median P_root = 0.07; S8B Fig; note the large variance in Fig 3G) when water stress was applied in the dark. The hypocotyl to root ratio adjustment was severely impaired (mean RQ_ratio = 0.46; median P_ratio = 0.003; Figs 3G, 4E, and 4F). In summary, BR bypass mutants mapped together with bin2 gain of function and loss of function mutants in the same quadrant of the volcano plots, showing a “confused” decision phenotype (high P-value, low ratio quotient; Fig 4F). We conclude that BR signalling and/or the BIN2 clade of shaggy-like kinases are required for differential growth decisions in the Arabidopsis seedling.

BR pathway mutants perceive light and water withdrawal

We next asked whether BR mutants are capable of perceiving and responding to light or water-stress. We first compared light versus dark conditions. Some BR pathway mutants exhibited etiolated phenotypes (short hypocotyl, long root, at least partially open cotyledons in the dark), as described in the literature [32]. Nonetheless, and even though responses were attenuated in some mutants, all BR pathway mutants had significant responses to light (shorter hypocotyls, longer roots as compared to dark-grown seedlings; S10 and S11 Figs). bin2-1 had a dwarf phenotype and an aberrant hypocotyl/root ratio, with a short hypocotyl and a relatively long root even in the dark (S10D and S11 Figs). Interestingly, the BR mutant lines with the strongest etiolation phenotypes (cpd and bri1-116brl1brl3; S11A and S11B Fig) in the dark were not the ones with the strongest deviation from the wild type under water deficit in the dark (S8 Fig). We also looked at the expression levels of the light responsive gene LHCB1.2 via qPCR in wild-type Ws-2 versus bin2-3bil1bil2. The data show that LHCB1.2 gene expression is light-regulated in bin2-3bil1bil2 seedlings (S12 Fig). We further looked at the response of selected BR mutants to water stress (-0.4MPa) in the light. Here, we focused on the hypocotyl response as this was clear and consistent in the wild type (S3A Fig). We found that both bin2-3bil1bil2 and bzr1-1D mutants were unimpaired in their hypocotyl responses to water stress in the light (S13 Fig). We conclude that the investigated BR mutants are not primarily impaired in their ability to perceive light or water stress.

BIN2 is required for a differential regulation of cell anisotropy in the hypocotyl

Hypocotyl elongation in the dark is known to occur via cellular elongation, with no significant contribution of cell division in the epidermis or cortex [39]. An assessment of organ and cell length in this study corroborated these findings, also for water stress in the dark. Indeed, we show that cellular parameters (cell length and width, assessed in scanning electron micrographs of hypocotyl cells) accounted for more than 73% of the observed differences in organ length and >89% of the differences in organ width (see Fig 5A and 5B for cellular parameters; S5A and S5B Fig for organ length and width and Fig 5G and 5H for a computation of fold-changes). We, therefore, conclude that it is predominantly a cellular response that controls hypocotyl growth under our conflict-of-interest scenario. To explore whether general BR-related cell elongation defects led to the confused phenotypes of some BR pathway mutants, we analysed bin2-1 mutants, which were among the most severely impaired hypocotyl response to water stress in the dark (S8A Fig). The data show a most striking impact of bin2-1 on growth anisotropy, assessed in 2D as length/width (Fig 5F). Indeed, in a comparison between dark and dark with water stress (darkW), the anisotropy of hypocotyl cells decreased considerably in the wild type (Fig 5C), but showed no adjustment in bin2-1 (Fig 5F). Cell length alone showed the elongation defect typical of bin2-1 mutants, with a much greater deviation from the wild type under darkW than under dark or light conditions; nonetheless, there was a significant length adjustment to water stress in the dark, even in bin2-1 (Fig 5E). These observations suggest that the impaired bin2-1 hypocotyl response can be attributed to an inability to differentially regulate cell anisotropy in response to the simultaneous withdrawal of light and water.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 5. Hypocotyl properties under single versus multiple stress conditions in wildtype and bin2-1 mutants.

Seed were germinated on 1/2MS medium and incubated for ten days under light, dark or darkW conditions. (A-C) Col-0 wild type; (D-F) bin2-1 semi-dominant allele. Scanning electron micrographs (A, D) were used to assess cellular parameters. (B, E) hypocotyl cell length. (C, F) hypocotyl cell anisotropy measured in 2D as length/width. Notice the highly significant decrease in anisotropy in the wild type (C) but the lack of response in bin2-1 (F) for darkW as compared to dark. (G, H) A comparison of organ versus cell parameters for the Col-0 wild type under the different environmental conditions, with a computation of fold-changes (FC) with respect to the light condition. Mean cell width significantly (P < 0.00001; not shown) decreased between light and dark and again between dark and darkW (H). Note that, as depicted in the last columns labelled “contributions”, the cellular parameters can account for 73–74% of organ length and 89–106% of organ width. The sample size (n) is given as the number of cells/ number of seedlings that were analysed. P-values were computed with a two-tailed student’s T-test and are represented as follows: ***: 0.001–0.0001; *****: < 0.00001. See related Fig 7.

https://doi.org/10.1371/journal.pgen.1010541.g005

The BIN2 clade of shaggy-like kinases differentially regulate the timing and extent of cell elongation in the root apical meristem in response to additive stress

It is generally accepted that root growth correlates with the size of the root apical meristem (RAM; [40]). Meristem size was assessed by computing the number of isodiametric and transition cells ([41,42]; S8 Method). In addition, we applied a Gaussian mixed model of cell length to distinguish between short meristematic cells and longer cells in the elongation zone (S14 Fig and S8 Method; [43]). Meristem size was shortest under water deficit in the dark (Figs 6A, S15A, and S15B) and, surprisingly, did not correlate well with final organ length (Figs 1C and 6G). We, therefore, measured mature cell length. This was highest in the dark, the condition with the shortest roots (Fig 6B). Thus, neither meristem size nor mature cell length account for the fold-change in final organ length (Fig 6G). To address this counterintuitive result, we deployed the CycB1,1:GUS [44] marker, expressed exclusively in cells undergoing M phase (S15C and S15D Fig). We observed an eight-fold higher number of mitotic cells in darkW as compared to dark conditions (S15C Fig). This difference was enhanced at an earlier time point (Figs 6C and S15C) and could explain the longer organ length but not the smaller meristem size of roots under darkW versus dark (Fig 6G).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 6. Root meristem properties under single versus multiple stress conditions.

Seedlings were grown in the light (orange), dark (black) or dark with -0.4MPa water stress (blue). (A-F) Wild type, with ecotype specified in the panel. (A) Meristem size determined via mixed Gaussian models, as described ([43]; S14 Fig). (B) Mature cell length, based on the ten most elongated cells for each condition; similar conclusions were reached when the 50 longest cells were used. (C) mitotic index at day 7 based on CycB1,1:GUS ([44]; S15C and S15D Fig). (D-F and H-J) 10 days after incubation, single epidermal cell files were measured, starting at the epidermal/ lateral root cap initials. The fitted lines were generated with Local Polynomial Regression Fitting with the ‘loess’ method in R; grey shading designates the 95 percent confidence interval. (D-F) cell lengths (D), width (E) and anisotropy (in 2D as length/width; F) of consecutive cells as a function of cell number from the quiescent centre (QC); the green arrows point to the steep slope for length (D) and anisotropy (F) under the darkW condition and the arrowheads to the kinks in the curves–the initiation of elongation—under all three conditions. (G) A tabulation of fold-changes (FCs) of measured parameters between different environmental conditions in the wild type (Ws-2, Col-0), with the smallest number highlighted in pink and the largest in green; note that the only FCs that go in same direction as root organ length are for the mitotic index (bold; data depicted in panel C). (H-J) A direct comparison of cell anisotropy under different environmental conditions between the bin2-3bil1bil2 triple mutant (purple) and the corresponding Ws-2 wild type (grey); notice that the mutant most markedly deviates from the wild type (compare purple versus green arrows in J) in the darkW condition (J), where the steep slope characteristic of the wild type is replaced by a flatter, undulating curve. The sample size (n) is given as the number of seedlings in panels a-c and as the number of cells/ number of seedlings that were analysed in panels H-J. P-values were computed with a two-tailed student’s T-test and are represented as follows: *: 0.05–0.01; **: 0.01–0.001; ****: 0.0001–0.00001, *****: < 0.00001. See related Figs 7, S14, S15, and S16.

https://doi.org/10.1371/journal.pgen.1010541.g006

We subsequently looked at cell length, width and anisotropy (computed in 2D as length/width) along single epidermal cell files as a function of distance from the quiescent centre. Cell elongation occurred in cells closer to the QC under darkW (blue arrowhead in Fig 6D) as compared to dark or light conditions (black and orange arrowheads in Fig 6D). Furthermore, the slopes of both the length and anisotropy curves were steepest for darkW (green arrows in Fig 6D and 6F). This was also apparent when we looked at the length of the first elongated cell, which was highest in darkW (S15E Fig). Cell elongation in cells close to the QC translates into an early exit from the root meristem. We conclude that root growth under water deficit in the dark is due not only to increased cell division but also to an early exit from the meristem (Fig 7B).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 7. Growth regulation and information flow in response to the availability of water and light.

(A) Scanning electron micrographs of wild-type (Col-0) hypocotyls under different environmental conditions. (B) Confocal micrographs of root tips under different environmental conditions; white arrows mark the junction between the meristematic and elongation zones (depicted as mean meristem size in Fig 6A) as assessed by mixed Gaussian model analysis ([43]: S14 Fig) and orange asterisks mark the highly elongated cells at the very beginning of the elongation zone under darkW. Wild type (Col-0) for light and dark; P_BRI1::BRI1-GFP for darkW; FM4-64 signal in all cases. Scalebar: 50 μm. (C) Information-flow diagram as a translation of our empirical findings. The green rectangles in the upper part of the diagram represent hypocotyl cells, and the arrows within each cell depict growth as a decentralized cellular response. The lower part of the diagram (brown-beige) represents the root system, with growth responses (arrows) driven in part by the root apical meristem (RAM). The green arrow depicts a hypothetical hypocotyl to root (basipetal) signal that coordinates trade-offs in organ growth in response to light; this does not take into account the possibility that photoreceptors in the root also perceive and respond to light [49]. Conversely, the blue arrow depicts a hypothetical root to hypocotyl (acropetal) signal that coordinates trade-offs in organ growth in response to water stress.

https://doi.org/10.1371/journal.pgen.1010541.g007

We then investigated root meristem properties in bin2-3bil1bil2, which had the most aberrant root response to water stress in the dark (Figs 3E and S8B magenta arrow). Meristem size and mature cell length followed the same trends in a comparison between bin2-3bil1bil2 (S16A and S16B Fig) and the wild type (Fig 6A and 6B), but the extent of elongation in cells proximal to the QC differed (S16C Fig). Indeed, bin2-3bil1bil2 length and anisotropy curves lacked the steep slopes characteristic for darkW in the wild type (compare the green arrows in Fig 6D, 6F and 6J to the purple arrows in Figs 6J and S16C). We conclude that bin2-3bil1bil2 mutants fail to adjust their root length due to an inability to differentially regulate the elongation of meristematic cells in the root in response to water stress in the dark.

Our observations suggest that root growth under our conflict-of-interest scenario requires root apical meristem function as well as the differential regulation of cell length in the elongation zone. To address this hypothesis, we turned to PLETHORA (PLT) AP2-domain transcription factors, which play a pivotal role in the read-out of the auxin gradient in root tips [45]. Indeed, high levels of PLT activity at the stem cell niche promote stem cell identity and maintenance; intermediate levels at the transition zone promote mitotic activity; and low levels in the elongation zone are required for cell differentiation [45]. Interestingly, plt1plt2 mutants [46] had an unimpaired hypocotyl response but failed to elongate their roots in response to water stress in the dark (S15F Fig). Taken together, the cell length and anisotropy curves (Fig 6) and genetic analyses (Figs 6, S15F and S16) suggest that root length under our different environmental conditions is regulated by (i) the mitotic index, (ii) the timing and extent of cell elongation—translating into the timing of meristematic exit—and (iii) cell geometry. We also conclude that these are differentially modulated to account for increased root length under different environmental conditions (Fig 6C–6E). In addition, an analysis of root meristem properties in bin2-3bil1bil2 suggests that the BIN2 clade of shaggy-like kinases is required for the Arabidopsis seedling’s ability to deploy different root growth strategies in response to abiotic stress cues under limiting conditions.