Focus Issue on Roots: The Optimal Lateral Root Branching Density for Maize Depends on Nitrogen and Phosphorus Availability

Observed phenotypic variation in the lateral root branching density (LRBD) in maize (Zea mays) is large (1–41 cm⁻¹ major axis i.e. brace, crown, seminal, and primary roots]), suggesting that LRBD has varying utility and tradeoffs in specific environments. Using the functional-structural plant model SimRoot, we simulated the three-dimensional development of maize root architectures with varying LRBD and quantified nitrate and phosphorus uptake, root competition, and whole-plant carbon balances in soils varying in the availability of these nutrients. Sparsely spaced (less than 7 branches cm⁻¹), long laterals were optimal for nitrate acquisition, while densely spaced (more than 9 branches cm⁻¹), short laterals were optimal for phosphorus acquisition. The nitrate results are mostly explained by the strong competition between lateral roots for nitrate, which causes increasing LRBD to decrease the uptake per unit root length, while the carbon budgets of the plant do not permit greater total root length (i.e. individual roots in the high-LRBD plants stay shorter). Competition and carbon limitations for growth play less of a role for phosphorus uptake, and consequently increasing LRBD results in greater root length and uptake. We conclude that the optimal LRBD depends on the relative availability of nitrate (a mobile soil resource) and phosphorus (an immobile soil resource) and is greater in environments with greater carbon fixation. The median LRBD reported in several field screens was 6 branches cm⁻¹, suggesting that most genotypes have an LRBD that balances the acquisition of both nutrients. LRBD merits additional investigation as a potential breeding target for greater nutrient acquisition.At least four major classes of plant roots can be distinguished based on the organ from which they originate: namely the seed, the shoot, the hypocotyl/mesocotyl, and other roots (Zobel and Waisel, 2010). The last class is lateral roots, which form in most plants the majority of the root length, but not necessarily of the root weight, as lateral roots have smaller diameter. Lateral roots start with the formation of lateral root primordia, closely behind the root tip of the parent root. These primordia undergo nine distinguishable steps, of which the last step is the emergence from the cortex of the parent root just behind the zone of elongation, usually only a few days after the first cell divisions that lead to their formation (Malamy and Benfey, 1997). However, not all primordia develop into lateral roots; some stay dormant (Dubrovsky et al., 2006), although dormancy of primordia may not occur in maize (Zea mays; Jordan et al., 1993; Ploshchinskaia et al., 2002). The final number of lateral roots is thereby dependent on the rate of primordia formation as well as the percentage of primordia that develop into lateral roots. This process of primordia formation and lateral root emergence is being studied intensively, including the genes that are activated during the different steps and the hormones regulating the process (López-Bucio et al., 2003; Dubrovsky et al., 2006; Osmont et al., 2007; Péret et al., 2009; Lavenus et al., 2013). Significant genotypic variation in the density of lateral roots has been observed, ranging from no lateral roots to 41 roots cm⁻¹ in maize (Trachsel et al., 2010; Lynch, 2013). This suggests that clear tradeoffs exist for the development of lateral roots and that these genotypes have preprogrammed growth patterns that are adaptive to specific environments. While some of the variation for lateral root branching density (LRBD) that has been observed across environments, for example by Trachsel et al. (2010), is constitutive, many genotypes have strong plasticity responses of LRBD to variations in soil fertility (Zhu et al., 2005a; Osmont et al., 2007). Both the nutrient and carbon status of the plant and the local nutrient environment of the (parent) root tip influence LRBD. Many studies have documented these plasticity responses, and others have tried to unravel parts of the sensing and signaling pathways that regulate LRBD. The utility of root proliferation into a nutrient patch has been studied and debated (Robinson et al., 1999; Hodge, 2004), but much less so the utility of having fewer or more branches across the whole root system. Our understanding of the adaptive significance of variation in LRBD among genotypes is thereby limited, with many studies not accounting for relevant tradeoffs. In this study, we integrate several functional aspects of LRBD with respect to nutrient acquisition, root competition, and internal resource costs and quantify these functional aspects using the functional structural plant model SimRoot. SimRoot simulates plant growth with explicit representation of root architecture in three dimensions (Fig. 1; Supplemental Movie S1). The model focuses on the resource acquisition by the root system and carbon fixation by the shoot while estimating the resource utilization and requirements by all the different organs.

Table I.

Minimum, maximum, and median LRBD in different populations phenotyped by various researchers at several locations in the worldLocations are as follows: D, Juelich, Germany; PA, State College, PA; and SA, Alma, South Africa. Some of the experiments included nutrient treatments: LN, low nitrogen availability; and LP, low phosphorus availability. Data were collected by counting the number of roots along a nodal root segment. Data were supplied by the person indicated under source: H.S., H. Schneider; L.Y., L. York; A.Z., A. Zhan; and J.P., J.A. Postma. WiDiv, Wisconsin Diversity panel; IBM, intermated B73 × Mo17; NAM, nested association mapping.

Population	No. of Genotypesa	Experiment	Location	Date	Nutrient Treatments	Source	LRBD
							Minimum	Maximum	Median
							cm−1
WiDiv	527	Field	SA	2010	–	H.S.	1	15	9
	400	Field	SA	2011, 2012	–	H.S.	0	18	6
	400	Field	SA	2013	LN	H.S.	0	13	6
IBM	30	Field	SA, PA	2012, 2013, 2014	LN	L.Y.	0	41	6
	18	Mesocosms	PA	2013	LN	A.Z.	1	10	4
NAM	1,235	Field	SA	2010, 2011, 2012	–	H.S.	0	14	6
	6	Rhizotrons	D	2011	LN, LP	J.P.	1	14	4

Open in a separate window^aMeans for the individual treatments are presented in Supplemental Appendix S4, Figure S5.Open in a separate windowFigure 1.Rendering of two simulated maize root systems. The model presents 40-d-old maize root systems with 2 (left) and 20 (right) branches cm⁻¹ major root axes. The simulations depicted here assumed that there were no nutrient deficiencies affecting growth. Carbon limitations do cause the laterals in the right root system to stay somewhat shorter. Different major axes, with their respective laterals, have different pseudocolors: light blue, primary root; green, seminal roots; red, crown roots; and yellow, brace roots. For animation of these root systems over time, see Supplemental Movie S1.The formation of lateral roots presumably increases the sink strength of the root system, promoting the development of greater root length and thereby greater nutrient and water acquisition. However, greater LRBD also places roots closer together, which may increase competition for nutrients and water among roots of the same plant, effectively reducing the uptake efficiency per unit of root length. This decrease in efficiency when the root system increases in size was nicely modeled by Berntson (1994). Furthermore, the metabolic costs of the construction and maintenance of the additional root length, either calculated in units of carbon or in terms of other limiting resources, may reduce the growth of other roots or the shoot (Lynch, 2007b). We can thereby logically derive that there will be an optimum number of lateral roots depending on the balance of the marginal cost of root production and the marginal utility of soil resource acquisition. Therefore, the optimal LRBD will depend on environmental conditions. It is not clear in the literature what the optimal branching density might be, and how different environmental factors shift this optimum to fewer or more lateral branches per centimeter of parent root. Considering the primacy of soil resources as pervasive limitations to plant growth, understanding the utility and tradeoffs of lateral root branching density is important in understanding the evolution of root architecture and plant environmental adaptation in general. In addition, such information would be useful for trait-based selection to develop cultivars with increased productivity on soils with suboptimal availability of nutrients. The necessity and prospects of developing such cultivars are outlined by Lynch (2007a, 2011).Here, we present results from root architectural simulations with which we estimated the optimal lateral branching density in maize in soils with variable availability of nitrogen and phosphorus. The model simulated the uptake benefits from having additional lateral roots, root competition as affected by the three-dimensional placement of roots over time, metabolic costs of lateral roots, and effects on whole-plant root architecture, notably with respect to rooting depth.