Most plants absorb water and nutrients from the soil via structures called roots. The shape, size, and structure of a plant's root system can change over its lifetime as the plant responds to changes in their local environment. For example, if water is scarce, a plant may develop a very deep root system that is more efficient at capturing water. Understanding how root systems respond to environmental cues may help us to identify the genes and processes involved.

In this study, Rellán-Álvarez et al. report a new live-imaging platform for analyzing root architecture and its regulation. This platform is called Growth and Luminescence Observatory for Roots (or GLO-Roots for short) and uses ‘luminescent’ markers that allow growing roots to be visualized when plants are grown in thin, soil-filled, transparent pots. GLO-Roots can track the growth of the plant roots as well as the activity of genes that respond to environmental stress. Rellán-Álvarez et al. developed a software tool called GLO-RIA (GLO-Roots Image Analysis) to analyze the resulting images. GLO-RIA performs several different types of image analysis, including one that detects the position, length, and direction of roots, as well as their shape and depth.

Rellán-Álvarez et al. tested the GLO-Roots techniques in various ways, for example, by analyzing the effects that different conditions have on the growth of the roots of the model plant known as Arabidopsis thaliana. Depriving the plants of a nutrient called phosphorous caused the roots to grow more horizontally than when phosphorus is plentiful, presumably to allow the plants to expand their search for phosphate in the upper layers of the soil, where this nutrient is usually more abundant. On the other hand, a shortage of water caused the roots to grow more vertically to access water stored deeper in the soil. GLO-Roots can also be used to measure the water content of soil at different depths and how this influences the architecture of the root.

Further experiments on tomato plants and a grass species called Brachypodium distachyon revealed the different architectures of their root systems. Rellán-Álvarez et al. propose that this system could be used to study the roots of other plant species in a variety of environmental conditions. This will provide a more detailed understanding of the ways that different plants adapt in response to changes in their environment.

Plant roots are three-dimensional assemblies of cells that coordinately monitor and acclimate to soil environmental change by altering physiological and developmental processes through cell-type and organ-specific regulatory mechanisms (Dinneny et al., 2008; Duan et al., 2013). Soil comprises a complex distribution of particles of different size, composition and physical properties, airspaces, variation in nutrient availability and microbial diversity (Brady and Weil, 2009; Lynch and Wojciechowski, 2015). These physical, chemical, and biological properties of soil can vary on spatial scales of meters to microns, and on temporal scales ranging from seasonal change to seconds. Root tips monitor this environment through locally and systemically acting sensory mechanisms (Bao et al., 2014; Tabata et al., 2014).

The architecture of the root system determines the volume of soil where resources can be accessed by the plant (rhizosphere) and is under both environmental and genetic control. Plasticity in growth parameters allows the plant to adjust its form to suit a particular soil. Lateral roots, which usually make up the majority of the total root system, often grow at an angle divergent from the gravity vector. This gravity set-point angle is controlled by auxin biosynthesis and signaling and can be regulated by developmental age and root type (Rosquete et al., 2013). Recent cloning of the DRO1 quantitative trait locus demonstrates that natural genetic variation is a powerful tool for uncovering such control mechanisms (Uga et al., 2013).

Specific root ideotypes (idealized phenotypes) have been proposed to be optimal for acquisition of water and nitrogen, which are distinct from ideotypes for low phosphorus. Based on computational modeling and field studies, the ‘steep, deep, and cheap’ ideotype proposed by Lynch and colleagues may provide advantages to the plant for capturing water and elements like nitrogen that are water soluble and therefore tend to move in the soil column with water. This ideotype consists of highly gravitropic, vertically oriented roots that grow deep in the soil column and develop large amounts of aerenchyma, which reduces the overall metabolic cost of the root system (Lynch and Wojciechowski, 2015). Other nutrients, like phosphorus, which have limited water solubility and are tightly bound to soil particles, usually accumulate in the top layers of soil and favor root systems that are more highly branched and shallow. The low-phosphorus ideotype effectively increases root exploration at the top layers of soil (Lynch and Wojciechowski, 2015). Modeling of root system variables shows that optimum architecture for nitrogen and phosphorus uptake is not the same (Postma and Lynch, 2014) and suggests tradeoffs that may affect the evolution of root architecture as a population adapts to a particular environmental niche (Laliberté et al., 2013).

Clearly, understanding the architecture of root systems and how environmental conditions alter root developmental programs is important for understanding adaptive mechanisms of plants and for identifying the molecular-genetic basis for different response programs. In addition, root systems have complexity beyond their architecture that needs to be incorporated into our understanding of plant–environment interactions. Primary and lateral roots exhibit different stress response programs in Arabidopsis (Duan et al., 2013; Tian et al., 2014) and may play specialized roles in water and nutrient uptake. Thus, it is important to develop methods that allow for a multidimensional characterization of the root system that includes growth, signaling, and interactions with other organisms. Furthermore, physiological parameters that affect whole-plant responses to the environment, such as transpiration, are likely integrated into such processes, thus, requiring a more holistic approach to studies of root function.

Based on these considerations, we have developed a new root imaging platform, Growth and Luminescence Observatory for Roots (GLO-Roots), which allows root architecture and gene expression to be studied in soil-grown plants. GLO-Roots is an integrated system composed of custom growth vessels, luminescent reporters, and imaging systems. We use rhizotrons that have soil volumes equivalent to small pots and support growth of Arabidopsis from germination to senescence. To visualize roots, we designed plant–codon-optimized luciferase reporters that emit light of different wavelengths. To visualize reporter expression, plants are watered with a dilute luciferin solution and imaged afterwards. We have built a custom luminescence-imaging system that automatically captures images of rhizotrons held vertically. The signal from each reporter is distinguished using band-pass filters held in a motorized filter wheel, which enables automated acquisition of images from plants expressing both structural and environmentally or developmentally responsive reporters. We have also developed GLO-RIA (Growth and Luminescence Observatory Root Image Analysis), an ImageJ (Schneider et al., 2012) plugin that allows for automated determination of (among other traits) root system area, convex hull, depth, width, and directionality, a metric which quantifies the angle of root segments with respect to gravity. GLO-RIA is also able to relate root system parameters to local root-associated variables such as reporter expression intensity and soil-moisture content.

Overall GLO-Roots has great utility in presenting environmental stimuli to roots in physiologically relevant ways and provides tools for characterizing responses to such stimuli at the molecular level in whole-adult root systems over broad time scales.

We have developed an integrated platform for growing, imaging, and analyzing root growth that provides advances in physiological relevance and retains the ability to visualize aspects of root biology beyond structure (Box 1).

The GLO-Roots platform

GLO-Roots is comprised of four parts: (i) growth vessels called rhizotrons that allow plant growth in soil and root imaging; (ii) luminescent reporters that allow various aspects of root biology to be tracked in living plants; (iii) GLO1 (Growth and Luminescence Observatory 1) luminescence-imaging system designed to automatically image rhizotrons; (iv) GLO-RIA, an image analysis suite designed to quantify root systems imaged using GLO-Roots.

Plant growth system

GLO-Roots utilizes custom-designed growth vessels classically known as rhizotrons, which hold a thin volume of soil between two sheets of polycarbonate plastic. Acrylic spacers provide a 2-mm space in which standard peat-based potting mix is added. Black vinyl sheets protect roots from light and rubber U-channels clamp the rhizotron materials together. Plastic racks hold the rhizotrons vertically and further protect the roots from light. Rhizotrons and rack are placed in a black tub and water is added, to a depth of about 2 cm, at the bottom to maintain moisture in the rhizotrons during plant growth. The volume of soil in the rhizotrons (100 cm³) is similar to small pots commonly used for Arabidopsis and supports growth throughout the entire life cycle (Figure 1A–C and Figure 1—figure supplement 1).

Figure 1 with 5 supplements see all

Download asset Open asset

Figure 1—source data 1

Two way ANOVA p-values comparing plants grown in MS media vs plants grown in soil (pots or rhizotrons) and plants collected at day or night.

We used p-value < 0.00065 threshold based on Bonferroni adjustment for multiple testing.

https://doi.org/10.7554/eLife.07597.005

Download elife-07597-fig1-data1-v2.xlsx

Figure 1—source data 2

Luminescence intensity values of the different luciferase isoforms across the emission spectrum.

https://doi.org/10.7554/eLife.07597.006

Download elife-07597-fig1-data2-v2.csv

Figure 1—source data 3

Gene expression values used to construct the PCA of root samples.

Shoot fresh weight (FW), shoot area, lateral root number and primary root length of plants grown in different containers and media.

https://doi.org/10.7554/eLife.07597.007

Download elife-07597-fig1-data3-v2.xls

Figure 1—source data 4

Gene expression values used to construct the PCA of shoot samples.

https://doi.org/10.7554/eLife.07597.008

Download elife-07597-fig1-data4-v2.csv

Figure 1—source data 5

Shoot Fresh Weight (FW) and primary root length of plants grown with or without luciferin.

https://doi.org/10.7554/eLife.07597.009

Download elife-07597-fig1-data5-v2.csv

Figure 1—source data 6

Ground truth and GLO-RIA measured values of directionality, depth and width use for validation.

https://doi.org/10.7554/eLife.07597.060

Download elife-07597-fig1-data6-v2.xls

To determine how the biology of plants grown in rhizotrons compares to other standard growth systems, we utilized high-throughput qRT-PCR to study how these conditions affect expression of 77 marker genes in root and shoot samples. These genes were curated from the literature and belong to a wide array of biological pathways including nutrient acquisition, hormone and light response and abiotic stress. Whole root and shoot samples were collected at the end of the light and dark periods (long-day conditions: 16-hr light, 8-hr dark) from plants grown in rhizotrons, pots, and petri dishes with two different media compositions: 1x Murashige and Skoog basal salts (ms) 1% sucrose or 0.25x ms, no sucrose (ms25). Principal component analysis (PCA) of the gene expression values showed a separation of soil- and gel-grown root systems in the first principal components (Figure 1—figure supplement 1A, Figure 1—source data 3). In roots grown on gel-based media, we observed enhanced expression of genes associated with light-regulated pathways (flavonoid biosynthesis: FLAVONOL SYNTHASE1, FLS1, CHALCONE SYNTHASE, CHS and photosynthesis: RUBISCO SUBUNIT 1A, RBCS1A, CYCLOPHILIN 38, CYP38), which is expected due to the exposure of gel-grown roots to light. In addition, genes associated with phosphorus nutrition (LOW PHOSPHATE RESPONSE1, LPR1, PHOSPHATE STARVATION RESPONSE1, PHR1) were less expressed in soil-grown roots (Figure 1—figure supplement 1), suggesting differences in nutrient availability between the different growth systems. Interestingly, shoot samples where not as clearly separated by growth media, and instead, time of day had a greater effect (Figure 1—figure supplement 2, Figure 1—source data 4). These data suggest root systems may be particularly sensitive to media conditions and indicate that rhizotron-grown root systems more closely approximate the biology of pot-grown plants than standard gel-based media. Shoot weight and primary root length were significantly reduced for gel-grown plants compared to rhizotron- or pot-grown plants suggesting significant differences in the biology of plants grown under these conditions (Figure 1—figure supplement 1B,C).

While the 2-mm depth of the soil sheet is 10–20 times the average diameter of an Arabidopsis root (between 100 and 200 microns [Meijon et al., 2013]), we evaluated whether rhizotron-grown plants exhibited any obvious stress as a consequence of physical constriction. We compared traits of plants growing in vessels that hold similar volumes of soil but in different volumetric shapes (Figure 1—figure supplement 1). The number of lateral roots was significantly lower in pot- and cylinder-grown plants compared to rhizotron-grown plants (Figure 1—figure supplement 1D), whereas primary root length of rhizotron and cylinder-grown plants was significantly greater than pot-grown plants (Figure 1—figure supplement 1E). No significant differences in shoot area were observed between the three systems (Figure 1—source data 1). Thus, these data do not support the hypothesis that rhizotron-grown plants experience physical constriction greater than other vessels holding the same volume of soil.

Generation of transgenic plants expressing different luciferases

Arabidopsis roots cannot easily be distinguished from soil using brightfield imaging due to their thinness and translucency (Figure 1—figure supplement 3); thus, reporter genes are needed to enhance the contrast between the root and its environment. Luciferase (LUC) is an ideal reporter to visualize roots: (1) unlike fluorescent reporters, luciferase does not require high-intensity excitation light, which could influence root growth, (2) peat-based soil (a type of histosol) exhibits no autoluminescence but does autofluoresce at certain excitation wavelengths similar to GFP (Figure 1—figure supplement 3), (3) while GFP is very stable, and thus not well suited for imaging dynamic transcriptional events, the luciferase enzyme is inactivated after catabolism of luciferin, making it ideal for studying dynamic processes such as environmental responses. A considerable number of luciferases have been developed that emit light spanning different regions of the visible spectrum, but their utilization has been limited to studies in animals (Table 1).

Table 1

Luciferase	Origin	Maximum wavelength	Substrate
PpyRE8	Firefly	618	D-luciferin
CBGRed	Click beetle	615	D-luciferin
Venus-LUC2	FP + firefly	580	D-luciferin
LUC(+)	Firefly	578	D-luciferin
CBG99	Click beetle	537	D-luciferin
Lux operon	A. fischeri	490	Biosynthesis pathway encoded within operon
NanoLUC	Deep sea shrimp	470	Furimazine

To determine the efficacy of using luciferase to visualize roots in soil, we codon-optimized sequences of PpyRE8, CBGRed, LUC2, and CBG99 for Arabidopsis expression. In addition, nanoLUC (Hall et al., 2012) and venus-LUC2 (Hara-Miyauchi et al., 2012) were utilized. Constitutive luciferase expression was driven in plants using the UBIQUITIN 10 (UBQ10) or ACTIN2 (ACT2) promoters using vectors assembled through a Golden Gate cloning system (Emami et al., 2013). Plants homozygous for a single locus T-DNA insertion were evaluated for in vivo emission spectra and luminescence intensity (Figure 1D, Figure 1—source data 2). All the evaluated luciferases use D-luciferin as a substrate facilitating the simultaneous imaging of different luciferases except nanoLUC, which uses a proprietary substrate furimazine (Hall et al., 2012). Luciferases with red-shifted emission spectra were less intense than the green-shifted luciferases (Figure 1D). LUC2o showed an emission maximum at 580 nm and a minor peak at 620 nm while CBG99o lacks the minor peak.

Continuous addition of luciferin did not have a significant effect on shoot weight or primary root length (Figure 1—figure supplement 4, Figure 1—source data 5). After luciferin addition, luminescence signal could be reliably detected in root systems for up to 10 days, depending on the developmental state of the plant.

GLO1: a semi-automated luminescence-imaging system for rhizotrons

Luminescence-imaging systems commercially available for biomedical research are usually optimized for imaging horizontally held specimens or samples in microtiter plates. Placing rhizotrons in this position would induce a gravitropic response in plants. Working with Bioimaging Solutions (San Diego, CA), we designed and built a luminescence-imaging system optimized for rhizotron-grown plants. GLO1 uses two PIXIS back-illuminated CCD cameras (Princeton Instruments, Trenton, NJ) to capture partially overlapping images of rhizotrons while a motorized stage automatically rotates the rhizotron to capture images of both sides (Figure 1E). A composite image is generated from the images captured of each side; Figure 1F shows that approximately half of the root system is revealed on each side with few roots being visible on both sides. Apparently, the soil sheet is thick enough to block light from portions of the root system but thin enough to ensure its continuous structure can be compiled from opposite face views. We tested the ability of GLO1-generated images to reveal complete root systems by manually quantifying the number of lateral roots in excavated root systems of eight different plants and testing these results against estimates of lateral root number from images of the same plants visually inspected by four different persons. These comparisons revealed good correlation (R² = 0.974) between actual lateral root counts and image-based estimation, indicating GLO1-generated root images provide an accurate representation of the in soil root system.

GLO-RIA: GLO-Roots Image Analysis

We developed a set of image analysis algorithms that were well suited for the complex root systems that GLO-Roots is able to capture. GLO-RIA is an ImageJ plugin divided in two modules.

The first module (RootSystem) performs four different types of analysis: (i) a local analysis that detects all root particles in the image and computes their position, length, and direction; (ii) the global analysis performs a root system level analysis and computes the total visible surface, convex hull, width, and depth; (iii) the shape analysis uses elliptic Fourier descriptors or pseudo-landmarks similarly to RootScape (Ristova et al., 2013) to perform a shape analysis on the root system; (iv) the directionality analysis computes the mean direction of root particles in a root system (either on the full image or by a user-defined region of interest in the image). These four analysis methods are fully automated by default, but can be manually adjusted if needed.

The second module of GLO-RIA (RootReporter) was specifically designed for the analysis of multi-layered images such as combinations of gene reporter, root structure, and soil moisture. Shortly, the plugin works as follows: (i) detection of the gene reporters and the structure reporters in their respective images; (ii) if needed, a manual correction can be performed to correct the automated detection; (iii) gene reporters are linked with the soil water content and the structure reporters, based on their proximity; (iv) gene reporter intensity (either absolute or normalized using the structural reporter) is computed; (v) all data are exported and saved to a Root System Markup Language datafile (Lobet et al., 2015). Gene and structure reporters can be followed across different time and space points. Using an object-oriented approach, great care has been taken to facilitate the user interactions on the different images to streamline the analysis process. Table 2 shows a list of root system features extracted using GLO-RIA. GLO-RIA does not currently have the ability to reconstruct the root architecture in itself (topological links between roots). This is a challenge for analyzing images captured by GLO-Roots since soil particles cause disruption of root segments.

Table 2

Variable	Unit
Projected area	cm2
Number of visible roots	–
Depth	cm
Width	cm
Convex hull area	cm2
Width	cm
Feret	cm
Feret angle	°
Circularity	–
Roundness	–
Solidity	–
Center of mass	cm
Directionality	°
Euclidean Fourier descriptors	–
Pseudo landmarks	–

We tested the accuracy of the measurements obtained from GLO-RIA using two different ground-truthed data sets. Manual measurement of root system width, depth, and average lateral root angle was determined by hand using ImageJ from an independent set of images corresponding to roots of several Arabidopsis accessions growing in control conditions. We also used ArchiSimple (Pagès et al., 2014) to generate 1240 images of root system models with contrasting sizes and lateral root angles. Since these images are computationally generated, exact determination of root system parameters was possible. For both ground truth data sets, GLO-RIA quantification provided measurements that were well correlated for all three measured parameters (Figure 1—figure supplement 5D–F, Figure 1—source data 6). Sample images of real and ArchiSimple generated root images are shown with GLO-RIA-defined directionality color-coding (Figure 1—figure supplement 5G–I).

Continuous imaging of root growth

The size of our rhizotrons enables undisturbed root system development (before roots reach the sides or the bottom of the rhizotron) for about 21–23 days for the Col-0 accession growing under long-day conditions (Figure 2, Figure 2—source data 1); however, root traits such as directionality can be observed through later stages of plant development. See 35 DAS root system and directionality in Figure 2A,B. An example of a time series spanning 11 to 21 days after sowing (DAS) of Col-0 roots expressing ProUBQ10:LUC2o is shown in Figure 2A and Video 1 with a color-coded time projection shown in Figure 2C. Directionality analysis (Figure 2B) shows a progressive change in root system angles from 0° (vertical) to 55° as lateral roots take over as the predominant root type. Figure 2D shows the evolution over time of several root traits that can be automatically captured by GLO-RIA (depth, width, area) and others that were manually quantified (primary root growth rate or number of lateral roots per primary root).

Figure 2

Download asset Open asset

Figure 2—source data 1

Directionality and whole root system architecture trait values from the time series.

https://doi.org/10.7554/eLife.07597.018

Download elife-07597-fig2-data1-v2.xlsx

Video 1

Download asset

Root system architecture of different Arabidopsis accessions

As a proof of concept to estimate the utility of our root-imaging system to phenotype adult root system traits, we transformed a small set of accessions (Bay-0, Col-0, and Sha) with the ProUBQ10:LUC2o reporter and quantified root system architecture at 22 DAS (Figure 3A–C, Figure 3—source data 1). GLO-RIA analysis of these root systems identified several root traits that distinguish Col-0, Bay-0, and Sha. Directionality analysis revealed an abundance of steep-angle regions in the root system of Bay-0, while Sha showed an abundance of shallow-angled regions and Col-0 was intermediate (Figure 3D). Bay-0 shows the deepest and narrowest root system leading to the highest depth/width ratio, while Sha has the widest root system (Figure 3E). Other root traits such as root system area and the vertical center of mass also showed significant differences (Figure 3—figure supplement 1B). Broad-sense heritability values for depth (96.3), area (92.0), depth/width (97.8), width (95.7), and vertical center of mass (95.0) were all higher than 90%.

Figure 3 with 1 supplement see all

Download asset Open asset

Figure 3—source data 1

Directionality, whole root system architectural trait values and shape predictors from Bay-0, Col-0 and Sha.

https://doi.org/10.7554/eLife.07597.021

Download elife-07597-fig3-data1-v2.xlsx

Figure 3—source data 2

Shape predictor values (TPS format) from Bay-0, Col-0 and Sha used to perform PCA.

https://doi.org/10.7554/eLife.07597.022

Download elife-07597-fig3-data2-v2.zip

Figure 3—source data 3

Whole root system architecture trait values from Bay-0, Col-0 and Sha.

https://doi.org/10.7554/eLife.07597.023

Download elife-07597-fig3-data3-v2.csv

To capture the richness of root architecture shape, we used GLO-RIA to extract pseudo-landmarks describing the shape of the root system (see ‘Materials and methods’ for more details) and performed PCA analysis (Figure 3—source data 2). The first principal component captures differences in the distribution of widths along the vertical axis and separates Col-0 and Sha from Bay-0 root systems (Figure 3F). Bay-0 shows a homogenous distribution of widths along the vertical axis, while Sha and Col-0 are much wider at the top than bottom. PC2 seems to be capturing a relationship between width at the top and total depth and separates Sha root systems, which are wide at the top and deep from Col-0 root systems, which are wide but not as deep as Sha. Shape information extracted from pseudo-landmarks can distinguish the three different accession using PCA analysis (Figure 3G, Figure 3—source data 3).

Spectrally distinct luciferases enable gene expression patterns, characterization of root system interactions, and microbial colonization

We tested whether spectrally distinct luciferase reporters would enable additional information besides root architecture to be captured from root systems. Luciferase reporters have been commonly used to study gene expression and these resources can potentially be utilized to study such regulatory events in soil-grown roots. We transformed ProACT2:PpyRE8o into two well-studied LUC reporter lines: the reactive oxygen species response reporter ProZAT12:LUC (Miller et al., 2009) (Figure 4A,B) and the auxin response reporter line ProDR5:LUC+ (Moreno-Risueno et al., 2010) (Figure 4C,D). We implemented in GLO-RIA an algorithm that semi-automatically identifies gene reporter signal and associates this object to the corresponding root structure segment. A graphical representation of the results obtained with RootReporter can be observed in Figure 4—figure supplement 1 (Figure 4—source data 1). Reporter intensity values along the first 5 mm of root tips can also be observed in Figure 4—figure supplement 2 (Figure 4—source data 2).

Figure 4 with 4 supplements see all

Download asset Open asset

Figure 4—source data 1

Data for ProZAT12:LUC reporter gene expression in root segments extracted from a whole root system.

https://doi.org/10.7554/eLife.07597.026

Download elife-07597-fig4-data1-v2.xlsx

Figure 4—source data 2

Luciferase intensity values from the root tip to maturation zone of ProUBQ10:LUC2o, ProZAT12:LUC and ProDR5:LUC+.

https://doi.org/10.7554/eLife.07597.027

Download elife-07597-fig4-data2-v2.csv

Figure 4—source data 3

Distances to boundary between plants.

Shoot FW, root system architecture and shoot area of single and pairs of plants grown in the same rhizotron.

https://doi.org/10.7554/eLife.07597.028

Download elife-07597-fig4-data3-v2.csv

We then took advantage of our ability to constitutively express two spectrally different luciferases and imaged the overlapping root systems (one expressing ProUBQ10:LUC2o and the other ProACT2:PPyRE8o). While two root systems were distinguishable using this system (Figure 4—figure supplement 3); measurements of root system area did not reveal a significant effect on root growth when two plants were grown in the same rhizotron, compared to one; however, further studies are warranted (Figure 4—figure supplement 3, Figure 4—source data 3).

The GLO-Roots system uses non-sterile growth conditions, which allows complex biotic interactions that may affect responses to the environment. Bacteria themselves can be engineered to express luminescent reporters through integration of the LUX operon, which results in luminescence in the blue region of the spectrum and is thus compatible with the plant-expressed luciferase isoforms we have tested. Pseudomonas fluorescens CH267 (Haney et al., 2015), a natural Arabidopsis root commensal, was transformed with the bacterial LUX operon and used to inoculate plants. 13 days after inoculation, we were able to observe bacterial luminescence colocalizing with plant roots. P. fluorescens did not show an obvious pattern of colonization at the root system scale level (Figure 4—figure supplement 4). As a proof-of-principle test of the multi-dimensional capabilities of the GLO-Roots system, we visualized both LUC2o and PPyRE8o reporters in plants and the LUX reporter in bacteria in the same rhizotron (Figure 4—figure supplement 4).

Adaptive changes in root system architecture under water deficit, phosphorus deficiency, and light

To test the utility of the GLO-Roots system to understand response of root systems to environmental stimuli, we tested the effects of light and conditions that mimic drought and nutritional deficiency. To examine the effects of light exposure on the root architecture, the black shields, which normally protect the soil and roots from light, were removed from the top half of the rhizotrons 10 DAS. Using directionality analysis, we detected a significant increase in the steepness of roots only in the light-exposed region of the rhizotron, while the lower shielded region showed no difference (Figure 6—figure supplement 3A,B and Figure 6—figure supplements 4, 5). Light can penetrate the top layers of soil (Mandoli et al., 1990) and it has been proposed to have a role in directing root growth especially in dry soils (Galen et al., 2007) through the blue light receptor phot1. Root directionality was not significantly different between light- and dark-treated roots of the phot1/2 double mutant (Figure 6—figure supplement 3B, lower panel, Figure 6—source data 3). suggesting that blue light perception is indeed necessary for this response (Galen et al., 2007; Moni et al., 2014). These data highlight the strong effects of light on root system architecture (Yokawa et al., 2013), which GLO-Roots rhizotrons are able to mitigate.

Plants grown in low-P soil showed a significant increase in the width–depth ratio of the root system compared to plants grown in P-replete soil, as determined using the automated root system area finder in GLO-RIA (Figure 6—figure supplement 2A,B, Figure 6—source data 2). Plants under P deficiency showed an increase in the ratio between root–shoot area (Figure 6—figure supplement 2C), which indicates a higher investment of resources in the development of the root system at the expense of shoot growth (Figure 6—figure supplement 2D). Root systems of control and P-deficient plants showed no significant differences in directionality at 22 DAS but at 27 DAS, roots were more horizontally oriented in P-deficient plants (Figure 6—figure supplement 2E). The observed changes in root architecture are consistent with root system ideotypes that improve phosphorus uptake efficiency.

GLO-Roots is especially well suited for studying water-deficit (WD) responses. First, shoots are exposed to the atmosphere and vapor pressure deficit is maintained at levels that allow for transpiration of water from the shoot. Second, soil in rhizotrons is exposed to air at the top and dries from the top-down; drying soil increases the volume occupied by air and reduces contact of root with liquid water, all of which are similar to changes in soil expected in the field during WD. Finally, as peat-based soil dries, its optical properties change, allowing moisture content to be approximated from brightfield images. We took advantage of the change in gray-scale pixel intensity to construct a calibration curve (Figure 5—figure supplement 1, Figure 5—source data 1) that quantitatively relates gray-scale pixel intensity to moisture content (Figure 5A); water content can be color coded in images with appropriate look-up tables (Figure 5B). Soil color was not affected by the presence or absence of roots (Figure 5—figure supplement 2). Using this approach, water content in a rhizotron can be mapped and visualized in 2D (Figure 5C,D). In the example shown, we can observe that a 22 DAS Bay-0 plant-depleted soil-moisture content locally around the root system (Figure 5E).

Figure 5 with 2 supplements see all

Download asset Open asset

Figure 5—source data 1

Pixel intensity and water content values used to construct calibration curve.

https://doi.org/10.7554/eLife.07597.034

Download elife-07597-fig5-data1-v2.csv

We performed several trials to simulate WD in our growth system. Plants were germinated, grown under control conditions then transferred to 29°C, and standing water was removed from the container holding the rhizotrons starting at 9 DAS or 13 DAS. Elevated temperature combined with water deficit is a common stress that modern crop varieties are poorly adapted to, thus, highlighting the importance of examining this combined treatment (Lobell et al., 2014; Ort and Long, 2014). Plants were maintained in this WD regime until 22 DAS when luciferin solution was added and the plants imaged. At 13 DAS, lateral roots near the soil surface are already emerged (Video 1, Figures 2A) and 9 days of subsequent WD treatment caused lateral roots to show an increase in gravitropism leading to the development of a root system that was deeper and more vertically oriented (Figure 6A). Roots of Bay-0 plants showed similar responses, though the extent of change was less pronounced since Bay-0 roots are normally more vertically oriented (Figure 6B). Plants transferred at 9 DAS and grown for 13 days under WD showed less lateral root development in the top layer of soil (Figure 6E). At this time point, lateral roots start to emerge (Video 1) and early drought may lead to growth quiescence or senescence. Careful examination of roots in these regions showed evidence of small lateral root primordia populating the primary root (Figure 6F). After 24 hr of re-watering (Figure 6G), these lateral root primordia reinitiated growth (Figure 6H).

Figure 6 with 5 supplements see all

Download asset Open asset

Figure 6—source data 1

Directionality values of Bay-0, Col-0, miz1, tir1-1 grown under WW, WD and high and control temperature conditions.

https://doi.org/10.7554/eLife.07597.038

Download elife-07597-fig6-data1-v2.csv

Figure 6—source data 2

Directionality, root system architecture traits and shoot area values of Col-0 plants grown under different phosphorus concentrations.

https://doi.org/10.7554/eLife.07597.039

Download elife-07597-fig6-data2-v2.xlsx

Figure 6—source data 3

Directionality values of Col-0 and phot1/2 plants grown with the root system in the dark or exposed to light in the top third of the rhizotron.

https://doi.org/10.7554/eLife.07597.040

Download elife-07597-fig6-data3-v2.csv

Figure 6—source data 4

Directionality values at different depths of the rhizotron for Col-0 plants exposed to light in the top third of the rhizotron.

https://doi.org/10.7554/eLife.07597.041

Download elife-07597-fig6-data4-v2.csv

Figure 6—source data 5

Relative water content of leaves from plants grown under WW and WD conditions and high or control temperatures.

https://doi.org/10.7554/eLife.07597.042

Download elife-07597-fig6-data5-v2.csv

Time-lapse imaging of the water-deficit response showed that changes in root growth direction occurred ahead of the dry soil front (Video 2). Using GLO-RIA, we were able to correlate local water-moisture contents with the orientation of root segments. With this approach, we observed that root segments in dryer areas of the rhizotron grew at steeper root angles (Figure 7, Figure 7—source data 1) than roots in wetter regions, though lateral root angle in these regions was also affected. These data suggest that both local and systemic signaling are likely involved in redirecting lateral roots deeper during the simulated drought treatments tested here.

Video 2

Download asset

Figure 7

Download asset Open asset

Figure 7—source data 1

Individual root segment traits of plants growing under WW and WD conditions.

https://doi.org/10.7554/eLife.07597.050

Download elife-07597-fig7-data1-v2.csv

We also grew plants under WD at control temperatures or under WW conditions at elevated temperature to test the effects of these individual stresses on root architecture. We observed that both conditions were sufficient to induce a change in root directionality indicating that the plant uses similar mechanisms to avoid heat and water-deficit-associated stresses (Figure 6—figure supplement 1). We next asked which regulatory pathways controlled the observed changes in lateral root directionality during simulated drought. Hydrotropism is a known environmental response that directs root growth towards wet regions of soil. MIZ1 is an essential regulator of hydrotropism (Kobayashi et al., 2007); however, miz1 mutants had no significant effect on WD-induced changes in root directionality, compared to wild type (Figure 6C, Figure 6—source data 1), indicating that this response was distinct from hydrotropism. Auxin is an important mediator of gravitropism and auxin treatment causes lateral roots to grow more vertically (Rosquete et al., 2013). Consistent with this role for auxin, mutant plants with loss of function in the auxin receptor TIR1 (Dharmasiri et al., 2005, Kepinski and Leyser, 2005) showed a near-random distribution of root angles and did not show changes in the root system directionality between WW and WD conditions (Figure 6D).

GLO-Roots for Brachypodium and tomato

To examine the general applicability of the GLO-Roots system for other species, we introduced LUC2o-expressing reporters into the model grass Brachypodium distachyon and the crop plant Lycopersicon esculentum (tomato). Brachypodium is well suited to the GLO-Root system because, like Arabidopsis, its small size allows mature root systems to be studied in relatively small soil volumes (Watt et al., 2009; Pacheco-Villalobos and Hardtke, 2012). LUC2o driven by the ZmUb1 promoter was introduced into Brachypodium using the pANIC vector (Mann et al., 2012). Brachypodium roots showed a distinct architecture from Arabidopsis marked by prolific development of secondary and tertiary lateral roots (Figure 8A). This is consistent with other studies that show that Brachypodium has a typical grass root system (Watt et al., 2009). Comparison of root system development in rhizotrons with gel-based media showed that root growth is higher in soil than in plates (Figure 8—figure supplement 1, Figure 8—source data 1). Previous work has suggested that auxin levels in Brachypodium roots are sub-optimal for growth (Pacheco-Villalobos et al., 2013). Pacheco-Villalobos and colleagues suggest that, in Brachypodium, and contrary to what happens in Arabidopsis, ethylene represses YUCCA reducing the synthesis of auxin. The reduced growth that we observe in plates and the high levels of ethylene that build up in sealed plates (Buer et al., 2003) would support this mechanism.

Figure 8 with 1 supplement see all

Download asset Open asset

Figure 8—source data 1

Depth of Brachypodium primary roots grown in petri plates and rhizotrons.

https://doi.org/10.7554/eLife.07597.052

Download elife-07597-fig8-data1-v2.csv

Tomato plants were transformed with Pro35S:PPyRE8o and ProeDR5rev:LUC2 reporters. The plants showed more rapid growth than Arabidopsis or Brachypodium and required fertilizer to prevent obvious signs of stress (reduced growth, anthocyanin accumulation). Root systems were imaged from 17 DAS plants. Roots showed presumptive lateral root primordia marked by DR5-expression (Figure 8C,D). These results show that the GLO-Roots method can be applied to study root systems of different plant species and will likely be useful for studying root systems of other small- to medium-sized model plants and for early stages of larger crop plants.

1. 1. Rellán-Álvarez R
   2. Lobet G
   3. Lindner H
   4. Pradier PL
   5. Sebastian J
   6. Yee M
   7. Geng Y
   8. Trontin C
   9. LaRue T
   10. Schrager-Lavelle A
   11. Haney CH
   12. Nieu R
   13. Maloof J
   14. Vogel JP
   15. Dinneny JR

   (2015) Data from: GLO-Roots: an imaging platform enabling multidimensional characterization of soil-grown roots systems

   Available at Dryad Digital Repository under a CC0 Public Domain Dedication.

   http://dx.doi.org/10.5061/dryad.7tk51

1. 1. Adams D
   2. Otarola-Castillo E

   (2013) Geomorph: an R package for the collection and analysis of geometric morphometric shape data

   Methods in Ecology and Evolution 4:393–399.

   https://doi.org/10.1111/2041-210X.12035

   • Google Scholar
2. 1. Auguie B

   (2012) GridExtra: functions in grid graphics

   GridExtra: functions in grid graphics, http://CRAN.R-project.org/package=gridExtra.

   http://CRAN.R-project.org/package=gridExtra

   • Google Scholar
3. 1. Bao Y
   2. Aggarwal P
   3. Robbins NE II
   4. Sturrock CJ
   5. Thompson MC
   6. Tan HQ
   7. Tham C
   8. Duan L
   9. Rodriguez PL
   10. Vernoux T
   11. Mooney SJ
   12. Bennett MJ
   13. Dinneny JR

   (2014) Plant roots use a patterning mechanism to position lateral root branches toward available water

   Proceedings of the National Academy of Sciences of USA 111:9319–9324.

   https://doi.org/10.1073/pnas.1400966111

   • Google Scholar
4. 1. Barr H
   2. Weatherley P

   (1962)

   A re-examination of the relative turgidity technique for estimating water deficit in leaves

   Australian Journal of Biological Sciences 15:413–428.

   • Google Scholar
5. 1. Blossfeld S
   2. Schreiber CM
   3. Liebsch G
   4. Kuhn AJ
   5. Hinsinger P

   (2013) Quantitative imaging of rhizosphere pH and CO₂ dynamics with planar optodes

   Annals of Botany 112:267–276.

   https://doi.org/10.1093/aob/mct047

   • Google Scholar
6. Book

   1. Brady NC
   2. Weil RR

   (2009)

   Elements of the nature and properties of soils

   New Jersey, United States: Prentice Hall.

   • Google Scholar
7. 1. Branchini BR
   2. Ablamsky DM
   3. Davis AL
   4. Southworth TL
   5. Butler B
   6. Fan F
   7. Jathoul AP
   8. Pule MA

   (2010) Red-emitting luciferases for bioluminescence reporter and imaging applications

   Analytical Biochemistry 396:290–297.

   https://doi.org/10.1016/j.ab.2009.09.009

   • Google Scholar
8. 1. Branchini BR
   2. Ablamsky DM
   3. Murtiashaw MH
   4. Uzasci L
   5. Fraga H
   6. Southworth TL

   (2007) Thermostable red and green light-producing firefly luciferase mutants for bioluminescent reporter applications

   Analytical Biochemistry 361:253–262.

   https://doi.org/10.1016/j.ab.2006.10.043

   • Google Scholar
9. 1. Buer CS
   2. Wasteneys GO
   3. Masle J

   (2003) Ethylene modulates root-wave responses in Arabidopsis

   Plant Physiology 132:1085–1096.

   https://doi.org/10.1104/pp.102.019182

   • Google Scholar
10. 1. Chitwood DH
    2. Ranjan A
    3. Martinez CC
    4. Headland LR
    5. Thiem T
    6. Kumar R
    7. Covington MF
    8. Hatcher T
    9. Naylor DT
    10. Zimmerman S
    11. Downs N
    12. Raymundo N
    13. Buckler ES
    14. Maloof JN
    15. Aradhya M
    16. Prins B
    17. Li L
    18. Myles S
    19. Sinha NR

    (2014) A modern ampelography: a genetic basis for leaf shape and venation patterning in grape

    Plant Physiology 164:259–272.

    https://doi.org/10.1104/pp.113.229708

    • Google Scholar
11. 1. Covington MF
    2. Harmer SL

    (2007) The circadian clock regulates auxin signaling and responses in Arabidopsis

    PLOS Biology 5:e222.

    https://doi.org/10.1371/journal.pbio.0050222

    • Google Scholar
12. 1. Dharmasiri N
    2. Dharmasiri
    3. Estelle M

    (2005) The F-box protein TIR1 is an auxin receptor

    Nature 435:441–445.

    https://doi.org/10.1038/nature03543

    • Google Scholar
13. 1. Dinneny JR
    2. Long TA
    3. Wang JY
    4. Jung JW
    5. Mace D
    6. Pointer S
    7. Barron C
    8. Brady SM
    9. Schiefelbein J
    10. Benfey PN

    (2008) Cell identity mediates the response of Arabidopsis roots to abiotic stress

    Science 320:942–945.

    https://doi.org/10.1126/science.1153795

    • Google Scholar
14. 1. Dmitry Grapov

    (2014)

    DeviumWeb: Dynamic Multivariate Data Analysis and Visualization Platform. v0.4

    DeviumWeb: Dynamic Multivariate Data Analysis and Visualization Platform. v0.4.

    • Google Scholar
15. 1. Dryden IL

    (2013) Shapes: statistical shape analysis

    Shapes: statistical shape analysis, http://CRAN.R-project.org/package=shapes.

    http://CRAN.R-project.org/package=shapes

    • Google Scholar
16. 1. Duan L
    2. Dietrich D
    3. Ng CH
    4. Chan PM
    5. Bhalerao R
    6. Bennett MJ
    7. Dinneny JR

    (2013) Endodermal ABA signaling promotes lateral root quiescence during salt stress in Arabidopsis seedlings

    The Plant Cell 25:324–341.

    https://doi.org/10.1105/tpc.112.107227

    • Google Scholar
17. 1. Emami S
    2. Yee MC
    3. Dinneny JR

    (2013) A robust family of golden gate Agrobacterium vectors for plant synthetic biology

    Frontiers in Plant Science 4:339.

    https://doi.org/10.3389/fpls.2013.00339

    • Google Scholar
18. 1. Galen C
    2. Rabenold JJ
    3. Liscum E

    (2007) Functional ecology of a blue light photoreceptor: effects of phototropin-1 on root growth enhance drought tolerance in Arabidopsis thaliana

    The New Phytologist 173:91–99.

    https://doi.org/10.1111/j.1469-8137.2006.01893.x

    • Google Scholar
19. 1. Hall MP
    2. Unch J
    3. Binkowski BF
    4. Valley MP
    5. Butler BL
    6. Wood MG
    7. Otto P
    8. Zimmerman K
    9. Vidugiris G
    10. Machleidt T
    11. Robers MB
    12. Benink HA
    13. Eggers CT
    14. Slater MR
    15. Meisenheimer PL
    16. Klaubert DH
    17. Fan F
    18. Encell LP
    19. Wood KV

    (2012) Engineered luciferase reporter from a deep sea shrimp utilizing a novel imidazopyrazinone substrate

    ACS Chemical Biology 7:1848–1857.

    https://doi.org/10.1021/cb3002478

    • Google Scholar
20. 1. Haney CH
    2. Samuel BS
    3. Bush J
    4. Ausubel FM

    (2015) Associations with rhizosphere bacteria can confer an adaptive advantage to plants

    Nature Plants 1:15051.

    https://doi.org/10.1038/nplants.2015.51

    • Google Scholar
21. 1. Hara-Miyauchi C
    2. Tsuji O
    3. Hanyu A
    4. Okada S
    5. Yasuda A
    6. Fukano T
    7. Akazawa C
    8. Nakamura M
    9. Imamura T
    10. Matsuzaki Y
    11. Okano HJ
    12. Miyawaki A
    13. Okano H

    (2012) Bioluminescent system for dynamic imaging of cell and animal behavior

    Biochemical and Biophysical Research Communications 419:188–193.

    https://doi.org/10.1016/j.bbrc.2012.01.141

    • Google Scholar
22. 1. Iwata H
    2. Ukai Y

    (2002) SHAPE: a computer program package for quantitative evaluation of biological shapes based on elliptic Fourier descriptors

    The Journal of Heredity 93:384–385.

    https://doi.org/10.1093/jhered/93.5.384

    • Google Scholar
23. 1. Kepinski S
    2. Leyser O

    (2005) The Arabidopsis F-box protein TIR1 is an auxin receptor

    Nature 435:446–451.

    https://doi.org/10.1038/nature03542

    • Google Scholar
24. 1. Kobayashi A
    2. Takahashi A
    3. Kakimoto Y
    4. Miyazawa Y
    5. Fujii N
    6. Higashitani A
    7. Takahashi H

    (2007) A gene essential for hydrotropism in roots

    Proceedings of the National Academy of Sciences of the United States of America 104:4724–4729.

    https://doi.org/10.1073/pnas.0609929104

    • Google Scholar
25. 1. Laliberté E
    2. Grace JB
    3. Huston MA
    4. Lambers H
    5. Teste FP
    6. Turner BL
    7. Wardle DA

    (2013) How does pedogenesis drive plant diversity?

    Trends in Ecology & Evolution 28:331–340.

    https://doi.org/10.1016/j.tree.2013.02.008

    • Google Scholar
26. 1. Lane MC
    2. Alteri CJ
    3. Smith SN
    4. Mobley HL

    (2007) Expression of flagella is coincident with uropathogenic Escherichia coli ascension to the upper urinary tract

    Proceedings of the National Academy of Sciences of USA 104:16669–16674.

    https://doi.org/10.1073/pnas.0607898104

    • Google Scholar
27. 1. Lindeboom JJ
    2. Nakamura M
    3. Hibbel A
    4. Shundyak K
    5. Gutierrez R
    6. Ketelaar T
    7. Emons AM
    8. Mulder BM
    9. Kirik V
    10. Ehrhardt DW

    (2013) A mechanism for reorientation of cortical microtubule arrays driven by microtubule severing

    Science 342:1245533.

    https://doi.org/10.1126/science.1245533

    • Google Scholar
28. 1. Lobell DB
    2. Roberts MJ
    3. Schlenker W
    4. Braun N
    5. Little BB
    6. Rejesus RM
    7. Hammer GL

    (2014) Greater sensitivity to drought accompanies maize yield increase in the U.S. Midwest

    Science 344:516–519.

    https://doi.org/10.1126/science.1251423

    • Google Scholar
29. 1. Lobet G
    2. Pound MP
    3. Diener J
    4. Pradal C
    5. Draye X
    6. Godin C
    7. Javaux M
    8. Leitner D
    9. Meunier F
    10. Nacry P
    11. Pridmore TP
    12. Schnepf A

    (2015) Root system markup language: toward a unified root architecture description language

    Plant Physiology 167:617–627.

    https://doi.org/10.1104/pp.114.253625

    • Google Scholar
30. 1. Lynch JP
    2. Wojciechowski T

    (2015) Opportunities and challenges in the subsoil: pathways to deeper rooted crops

    Journal of Experimental Botany 66:2199–2210.

    https://doi.org/10.1093/jxb/eru508

    • Google Scholar
31. 1. Mandoli DF
    2. Ford GA
    3. Waldron LJ
    4. Nemson JA
    5. Briggs WR

    (1990) Some spectral properties of several soil types: implications for photomorphogenesis*

    Plant, Cell & Environment 13:287–294.

    https://doi.org/10.1111/j.1365-3040.1990.tb01313.x

    • Google Scholar
32. 1. Mann DG
    2. Lafayette PR
    3. Abercrombie LL
    4. King ZR
    5. Mazarei M
    6. Halter MC
    7. Poovaiah CR
    8. Baxter H
    9. Shen H
    10. Dixon RA
    11. Parrott WA
    12. Neal Stewart C Jr

    (2012) Gateway-compatible vectors for high-throughput gene functional analysis in switchgrass (Panicum virgatum L.) and other monocot species

    Plant Biotechnology Journal 10:226–236.

    https://doi.org/10.1111/j.1467-7652.2011.00658.x

    • Google Scholar
33. 1. Meijon M
    2. Satbhai SB
    3. Tsuchimatsu T
    4. Busch W

    (2013) Genome-wide association study using cellular traits identifies a new regulator of root development in Arabidopsis

    Nature Genetics 46:77–81.

    https://doi.org/10.1038/ng.2824

    • Google Scholar
34. 1. Miller G
    2. Schlauch K
    3. Tam R
    4. Cortes D
    5. Torres MA
    6. Shulaev V
    7. Dangl JL
    8. Mittler R

    (2009) The plant NADPH oxidase RBOHD mediates rapid systemic signaling in response to diverse stimuli

    Science Signaling 2:ra45.

    https://doi.org/10.1126/scisignal.2000448

    • Google Scholar
35. 1. Moni A
    2. Lee AY
    3. Briggs WR
    4. Han IS

    (2014) The blue light receptor Phototropin 1 suppresses lateral root growth by controlling cell elongation

    Plant Biology pp. 34–40.

    https://doi.org/10.1111/plb.12187

    • Google Scholar
36. 1. Moreno-Risueno MA
    2. Van Norman JM
    3. Moreno A
    4. Zhang J
    5. Ahnert SE
    6. Benfey PN

    (2010) Oscillating gene expression determines competence for periodic Arabidopsis root branching

    Science 329:1306–1311.

    https://doi.org/10.1126/science.1191937

    • Google Scholar
37. 1. Moriwaki T
    2. Miyazawa Y
    3. Kobayashi A
    4. Uchida M
    5. Watanabe C
    6. Fujii N
    7. Takahashi H

    (2011) Hormonal regulation of lateral root development in Arabidopsis modulated by MIZ1 and requirement of GNOM activity for MIZ1 function

    Plant Physiology 157:1209–1220.

    https://doi.org/10.1104/pp.111.186270

    • Google Scholar
38. 1. Ort DR
    2. Long SP

    (2014) Limits on yields in the corn belt

    Science 344:484–485.

    https://doi.org/10.1126/science.1253884

    • Google Scholar
39. 1. Pacheco-Villalobos D
    2. Hardtke CS

    (2012) Natural genetic variation of root system architecture from Arabidopsis to Brachypodium: towards adaptive value

    Philosophical Transactions of the Royal Society of London B: Biological Sciences 367:1552–1558.

    https://doi.org/10.1098/rstb.2011.0237

    • Google Scholar
40. 1. Pacheco-Villalobos D
    2. Sankar M
    3. Ljung K
    4. Hardtke CS

    (2013) Disturbed local auxin homeostasis enhances cellular anisotropy and reveals alternative wiring of auxin-ethylene crosstalk in Brachypodium distachyon seminal roots

    PLOS Genetics 9:e1003564.

    https://doi.org/10.1371/journal.pgen.1003564

    • Google Scholar
41. 1. Pagès L
    2. Bècel C
    3. Boukcim H
    4. Moreau D
    5. Nguyen C
    6. Voisin AS

    (2014) Calibration and evaluation of ArchiSimple, a simple model of root system architecture

    Ecological Modelling 290:76–84.

    https://doi.org/10.1016/j.ecolmodel.2013.11.014

    • Google Scholar
42. 1. Postma JA
    2. Lynch JP

    (2014) The optimal lateral root branching density for maize depends on nitrogen and phosphorus availability

    Plant Physiology 166:590–602.

    https://doi.org/10.1104/pp.113.233916

    • Google Scholar
43. 1. R Developement Core Team

    (2014) R: a language and environment for statistical computing

    R: a language and environment for statistical computing, R Foundation for Statistical Computing, http://www.R-project.org/.

    http://www.R-project.org/

    • Google Scholar
44. 1. Rellán-Álvarez R
    2. Lobet G
    3. Lindner H
    4. Pradier PL
    5. Sebastian J
    6. Yee M
    7. Geng Y
    8. Trontin C
    9. LaRue T
    10. Schrager-Lavelle A
    11. Haney CH
    12. Nieu R
    13. Maloof J
    14. Vogel JP
    15. Dinneny JR

    (2015) Data from: GLO-Roots: an imaging platform enabling multidimensional characterization of soil-grown roots systems

    Data from: GLO-Roots: an imaging platform enabling multidimensional characterization of soil-grown roots systems, http://datadryad.org/review?doi=doi:10.5061/dryad.7tk51, Available at Dryad Digital Repository under a CC0 Public Domain Dedication.

    http://datadryad.org/review?doi=doi:10.5061/dryad.7tk51

    • Google Scholar
45. 1. Ristova D
    2. Rosas U
    3. Krouk G
    4. Ruffel S
    5. Birnbaum KD
    6. Coruzzi GM

    (2013) RootScape: a landmark-based system for rapid screening of root architecture in Arabidopsis

    Plant Physiology 161:1086–1096.

    https://doi.org/10.1104/pp.112.210872

    • Google Scholar
46. 1. Rosquete MR
    2. von Wangenheim D
    3. Marhavý P
    4. Barbez E
    5. Stelzer EH
    6. Benková E
    7. Maizel A
    8. Kleine-Vehn J

    (2013) An auxin transport mechanism restricts positive orthogravitropism in lateral roots

    Current Biology 23:817–822.

    https://doi.org/10.1016/j.cub.2013.03.064

    • Google Scholar
47. 1. Ruegger M
    2. Dewey E
    3. Gray WM
    4. Hobbie L
    5. Turner J
    6. Estelle M

    (1998) The TIR1 protein of Arabidopsis functions in auxin response and is related to human SKP2 and yeast grr1p

    Genes & Development 12:198–207.

    https://doi.org/10.1101/gad.12.2.198

    • Google Scholar
48. 1. Schneider CA
    2. Rasband WS
    3. Eliceiri KW

    (2012) NIH Image to ImageJ: 25 years of image analysis

    Nature Methods 9:671–675.

    https://doi.org/10.1038/nmeth.2089

    • Google Scholar
49. 1. Shaw SL
    2. Ehrhardt DW

    (2013) Smaller, faster, brighter: advances in optical imaging of living plant cells

    Annual Review of Plant Biology 64:351–375.

    https://doi.org/10.1146/annurev-arplant-042110-103843

    • Google Scholar
50. 1. Tabata R
    2. Sumida K
    3. Yoshii T
    4. Ohyama K
    5. Shinohara H
    6. Matsubayashi Y

    (2014) Perception of root-derived peptides by shoot LRR-RKs mediates systemic N-demand signaling

    Science 346:343–346.

    https://doi.org/10.1126/science.1257800

    • Google Scholar
51. 1. Tian H
    2. De Smet I
    3. Ding Z

    (2014) Shaping a root system: regulating lateral versus primary root growth

    Trends in Plant Science 19:426–431.

    https://doi.org/10.1016/j.tplants.2014.01.007

    • Google Scholar
52. 1. Uga Y
    2. Sugimoto K
    3. Ogawa S
    4. Rane J
    5. Ishitani M
    6. Hara N
    7. Kitomi Y
    8. Inukai Y
    9. Ono K
    10. Kanno N
    11. Inoue H
    12. Takehisa H
    13. Motoyama R
    14. Nagamura Y
    15. Wu J
    16. Matsumoto T
    17. Takai T
    18. Okuno K
    19. Yano M

    (2013) Control of root system architecture by DEEPER ROOTING 1 increases rice yield under drought conditions

    Nature Genetics 45:1097–1102.

    https://doi.org/10.1038/ng.2725

    • Google Scholar
53. 1. Vogel J
    2. Hill T

    (2008) High-efficiency Agrobacterium-mediated transformation of Brachypodium distachyon inbred line Bd21-3

    Plant Cell Reports 27:471–478.

    https://doi.org/10.1007/s00299-007-0472-y

    • Google Scholar
54. 1. Watt M
    2. Schneebeli K
    3. Dong P
    4. Wilson IW

    (2009) The shoot and root growth of Brachypodium and its potential as a model for wheat and other cereal crops

    Functional Plant Biology 36:960–969.

    https://doi.org/10.1071/FP09214

    • Google Scholar
55. 1. Wickham H

    (2009) Ggplot2: elegant graphics for data analysis

    Ggplot2: elegant graphics for data analysis, Springer New York, http://had.co.nz/ggplot2/book.

    http://had.co.nz/ggplot2/book

    • Google Scholar
56. 1. Wickham H

    (2014) Tidyr: easily tidy data with spread and gather functions

    Tidyr: easily tidy data with spread and gather functions, http://CRAN.R-project.org/package=tidyr.

    http://CRAN.R-project.org/package=tidyr

    • Google Scholar
57. 1. Wilke CO

    (2015) Cowplot: streamlined plot theme and plot annotations for ggplot2

    Cowplot: streamlined plot theme and plot annotations for ggplot2, http://cran.r-project.org/web/packages/cowplot/index.html.

    http://cran.r-project.org/web/packages/cowplot/index.html

    • Google Scholar
58. 1. Yokawa K
    2. Kagenishi T
    3. Baluška F

    (2013) Root photomorphogenesis in laboratory-maintained Arabidopsis seedlings

    Trends in Plant Science 18:117–119.

    https://doi.org/10.1016/j.tplants.2013.01.002

    • Google Scholar

Author details

1. Rubén Rellán-Álvarez

   Department of Plant Biology, Carnegie Institution for Science, Stanford, United States

   Present address

   Unidad de Genómica Avanzada, Laboratorio Nacional de Genómica para la Biodiversidad, Centro de Investigación y de Estudios Avanzados del Instituto Politécnico Nacional, Irapuato, Mexico

   Contribution

   RR-Á, Conception, design and development of the growth and imaging system and Arabidopsis transgenic lines; acquisition, analysis and interpretation of data; drafting and revising the article

   Competing interests

   The authors declare that no competing interests exist.

   "This ORCID iD identifies the author of this article:" 0000-0001-6843-3716

2. Guillaume Lobet

   PhytoSystems, University of Liège, Liège, Belgium

   Contribution

   GL, Development of the GLO-RIA image analysis plugin, analysis and interpretation of data, drafting and revising the article

   Competing interests

   The authors declare that no competing interests exist.

   "This ORCID iD identifies the author of this article:" 0000-0002-5883-4572

3. Heike Lindner

   Department of Plant Biology, Carnegie Institution for Science, Stanford, United States

   Contribution

   HL, Acquisition of data, development of the tomato growth and imaging setup

   Contributed equally with

   Pierre-Luc Pradier and Jose Sebastian

   Competing interests

   The authors declare that no competing interests exist.

   "This ORCID iD identifies the author of this article:" 0000-0002-5913-9803

4. Pierre-Luc Pradier

   Department of Plant Biology, Carnegie Institution for Science, Stanford, United States

   Present address

   Boyce Thompson Institute for Plant Research, United States Department of Agriculture, Ithaca, United States

   Contribution

   P-LP, Acquisition of data, analysis and interpretation of data

   Contributed equally with

   Heike Lindner and Jose Sebastian

   Competing interests

   The authors declare that no competing interests exist.

   "This ORCID iD identifies the author of this article:" 0000-0002-2426-0742

5. Jose Sebastian

   Department of Plant Biology, Carnegie Institution for Science, Stanford, United States

   Contribution

   JS, Development of Brachypodium transgenic lines, acquisition and analysis of Brachypodium, tomato and Arabidosis data

   Contributed equally with

   Heike Lindner and Pierre-Luc Pradier

   Competing interests

   The authors declare that no competing interests exist.

   "This ORCID iD identifies the author of this article:" 0000-0002-1826-0308

6. Muh-Ching Yee

   Department of Plant Biology, Carnegie Institution for Science, Stanford, United States

   Contribution

   M-CY, Development of Arabidopsis and Brachypodium transgenic lines

   Competing interests

   The authors declare that no competing interests exist.

   "This ORCID iD identifies the author of this article:" 0000-0002-0445-7927

7. Yu Geng

   1. Department of Plant Biology, Carnegie Institution for Science, Stanford, United States
   2. Department of Energy, Department of Energy Joint Genome Institute, Walnut Creek, United States

   Contribution

   YG, Development of Arabidopsis transgenic lines

   Competing interests

   The authors declare that no competing interests exist.

8. Charlotte Trontin

   Department of Plant Biology, Carnegie Institution for Science, Stanford, United States

   Contribution

   CT, Acquisition and analysis of the QPCR data

   Competing interests

   The authors declare that no competing interests exist.

   "This ORCID iD identifies the author of this article:" 0000-0002-2235-4390

9. Therese LaRue

   Department of Biology, Stanford University, Stanford, United States

   Contribution

   TLR, Acquisition and analysis of the QPCR data

   Competing interests

   The authors declare that no competing interests exist.

   "This ORCID iD identifies the author of this article:" 0000-0002-6805-5118

10. Amanda Schrager-Lavelle

    Department of Plant Biology, University of California, Davis, Davis, United States

    Contribution

    AS-L, Contributed the unpublished dual-color tomato line

    Competing interests

    The authors declare that no competing interests exist.

    "This ORCID iD identifies the author of this article:" 0000-0003-2435-9336

11. Cara H Haney

    Department of Genetics, Department of Molecular Biology, Massachusetts General Hospital, Harvard Medical School, Boston, United States

    Contribution

    CHH, Contributed the unpublished *Pseudomonas fluorescens* CH267-lux strain

    Competing interests

    The authors declare that no competing interests exist.

    "This ORCID iD identifies the author of this article:" 0000-0002-6099-6677

12. Rita Nieu

    Western Regional Research Center, United States Department of Agriculture, Albany, United States

    Contribution

    RN, Contribution to the development of the Brachypodium transgenic line

    Competing interests

    The authors declare that no competing interests exist.

13. Julin Maloof

    Department of Plant Biology, University of California, Davis, Davis, United States

    Contribution

    JM, Contributed the unpublished dual-color tomato line

    Competing interests

    The authors declare that no competing interests exist.

    "This ORCID iD identifies the author of this article:" 0000-0002-9623-2599

14. John P Vogel

    Department of Energy, Department of Energy Joint Genome Institute, Walnut Creek, United States

    Contribution

    JPV, Contribution to the development of the Brachypodium transgenic line

    Competing interests

    The authors declare that no competing interests exist.

    "This ORCID iD identifies the author of this article:" 0000-0003-1786-2689

15. José R Dinneny

    Department of Plant Biology, Carnegie Institution for Science, Stanford, United States

    Contribution

    JRD, Conception, design and development of the growth and imaging system and Arabidopsis transgenic lines; acquisition, analysis and interpretation of data; drafting and revising the article

    For correspondence

    jdinneny@carnegiescience.edu

    Competing interests

    The authors declare that no competing interests exist.

    "This ORCID iD identifies the author of this article:" 0000-0002-3998-724X

• 12,369

  views

• 2,324

  downloads

• 193

  citations

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