Multiscale systems analysis of root growth and development: modeling beyond the network and cellular scales

Abstract

Over recent decades, we have gained detailed knowledge of many processes involved in root growth and development. However, with this knowledge come increasing complexity and an increasing need for mechanistic modeling to understand how those individual processes interact. One major challenge is in relating genotypes to phenotypes, requiring us to move beyond the network and cellular scales, to use multiscale modeling to predict emergent dynamics at the tissue and organ levels. In this review, we highlight recent developments in multiscale modeling, illustrating how these are generating new mechanistic insights into the regulation of root growth and development. We consider how these models are motivating new biological data analysis and explore directions for future research. This modeling progress will be crucial as we move from a qualitative to an increasingly quantitative understanding of root biology, generating predictive tools that accelerate the development of improved crop varieties.

Figure 1.

Key Interacting Network/Cell/Tissue-Scale Processes during Root Growth and Development. Predicting root growth and development (F) requires understanding of gene regulation and protein interactions (A) and how these affect the cell wall biomechanics (B) and hydraulics (D) that determine cell growth. Cell growth in turn causes growth and patterning on the tissue scale (E), which feeds back on the subcellular hormone levels (C) and, hence, gene regulation. QC, quiescent center. [Panel A reprinted from Middleton et al. (2012), Figure 1].

Figure 2.

Recent Multiscale Auxin Transport Models. (A) A three-dimensional model from Swarup et al. (2005) predicted the auxin distribution in root elongation zone tissues following a gravitropic stimulus. [Adapted by permission from Macmillan Publishers Ltd: Nature Cell Biol. Swarup et al. (2005), Figure 2.] (B) A two-dimensional model from Grieneisen et al. (2007) predicted the auxin distribution in the growing root tip, using idealized cell geometries. [Adapted by permission from Macmillan Publishers Ltd: Nature. Grieneisen et al. (2007), Figure 1.] (C) A two-dimensional model from Stoma et al. (2008) predicted the auxin distribution in the root tip using cell geometries extracted from confocal images. EZ, elongation zone; MZ, meristematic zone. [Adapted from Stoma et al. (2008), Figure 10.]

Figure 3.

Extracting Quantitative Data from Digital Images at Multiple Scales. (A) Color images of plated Arabidopsis seedlings provide length, curvature, and growth data (French et al., 2009). [Reprinted from French et al. (2012), Figure 3.] (B) Segmentation of confocal laser scanning microscopy images provides realistic tissue and cell geometries (Pound et al., 2012). [Reprinted from Pound et al. (2012), Figure 1.] (C) Protein concentration and localization data can be recovered from confocal microscope images (red channel, propidium iodide; green channel, PIN2–green fluorescent protein) by first extracting cell-scale descriptions (Pound et al., 2012). [Reprinted from Pound et al. (2012), Figure 3.]