Boron and the evolutionary development of roots

ABSTRACT

Experimental work has shown that Boron (i.e., Boric acid, B) is an essential and multifunctional microelement for vascular plant development. In addition to its other functions, which include xylem development and lignin biosynthesis, we now know that B is involved in phytohormone-signaling and influences the mechanical properties of intercellular pectins. From these data, we conclude that B played an important role during the evolutionary development of lignified tissues, and that it may have been involved in the evolution of vascular plant roots, as hypothesized by D. H. Lewis in 1980. Herein, we review the data pertaining to Lewis’ hypothesis, present experimental results on the role of B in root (vs. rhizoid) formation in sunflower vs. a liverwort, and describe the appearance of roots in the fossil record. Open questions are addressed, notably the lack of our knowledge concerning soil microbes and their interactive roles with the micronutrient B during root formation.

KEYWORDS:

• Boron
• evolution
• root development
• epiphytic microbes

Introduction

Thirty-seven years ago, Lewis¹ hypothesized that the micronutrient Boron (i.e., Boric acid, B) played an essential role during the evolutionary transition from non-vascular (bryophyte-like) plants to vascularized embryophytes during the early Paleozoic. Specifically, Lewis¹ summarized evidence indicating that B was required for the biosynthesis of lignin. Moreover, in conjunction with the phytohormone auxin, B may induce the differentiation of xylem-elements. Since B was found to be necessary for the development of adventitious roots, Lewis¹ postulated that this micronutrient may have been a key factor for the origin of the root system in the earliest vascular plants.

In a Review Article on “Early Root Evolution," Kenrick and Strullu-Derrien² provide paleontological data documenting the occurrence of rhizoids in bryophyte-like Rhynie Chert fossils, such as the ca. 407 million-year (Myr)-old Horneophyton (Fig. 1), and roots on sporophytes of some early vascular plants, such as Zosterophyllum of about the same age. In 2016, Gerrienne et al.³ summarized this topic from a paleontological perspective. With respect to the physiologic basis of rhizoid/root formation in the most ancient embryophytes, Kenrick and Strullu-Derrien² discuss the possible role of polar auxin transport. However, in both articles, the authors^2,3 do not mention the hypothesis of Lewis,¹ nor some of the other important functions Boron performs in developing land plants. These topics, notably the effects of this micronutrient (as well as epiphytic microbes) on root development in seed plants,^4,5 such as Arabidopsis thaliana or Helianthus annuus, are the focus of the present article.

Figure 1. Vascular strand and abaxial cluster of rhizoids on a transverse section through the rhizome-like portion of the Rhynie Chert plant Horneophyton. (Specimen from the Cornell University fossil plant collection).

Evolutionary gains and losses of roots

Roots and root-like organs have evolved independently many times among the vascular plants (i.e., twice in the lycopods and perhaps as many as 3 times in the euphyllophytes).^6,7 The capacity to form roots has also been lost or repressed among some extant tracheophytes, as for example among some eusporangiate “ferns” (e.g., Psilotum nudum) and some angiosperms (e.g., Wolffia arrhiza).⁸ The evolutionary origins of roots are unfortunately ambiguous because it is often difficult to determine if roots are absent, as in some extant pteridophytes, not preserved for a particular fossil, or perhaps may not be recognized in a fossilized state. Another factor contributing to the problematic nature of root evolution is the paucity of definitive developmental features that can be used to define accurately and precisely what is meant by “roots." This ambiguity is attributable in part to the fact that roots have evolved many times, and therefore in different ways among tracheophytes. For example, root development is typically described as endogenous, i.e., the apical meristems of lateral roots develop from cells within the pericycle. However, the roots of some ferns (e.g., Ceratopteris) emerge from cells in the outermost layer of the cortex just below the epidermis (whose cells divide anticlinally to keep pace with divisions in the hypodermis) in a manner that is essentially indistinguishable from the ontogeny of eudicot leaves,^9,10 wherein the initial bulge of the leaf primordium forms as a result of sub-epidermal and epidermal cell divisions. In passing, it is worth noting that the sub-epidermal origin of the roots of Ceratopteris and some other leptosporangiate ferns is virtually indistinguishable from that of stigmarian roots, which are considered by many workers to be leaf homologs.¹⁰ Likewise, the application of other criteria typically used to distinguish roots from other organ types is equally problematic. For example, an endodermis with a Casparian strip occurs in the stems of some lycopods (e.g., Hupertzia) and the leaves of some conifers (e.g., Pinus), and the aerial roots of some grasses lack or possess poorly developed rootcaps (which would have a low probability of being preserved as fossils in some geological contexts).

Finally, it should not escape notice that even among the monophyletic angiosperms, the root apical meristem has at least 15 demonstrably different anatomic configurations.¹⁰ Consequently, the application of the concept of endogenous development or any other developmental or anatomic criteria to define “roots” can be misleading. It should also be noted that, based on a single model organism such as Arabidopsis thaliana, the multiple evolutionary origins of organs called “roots” may preclude drawing undisputed conclusions about their evolution.

Root evolution and the fossil record

If we visualize the biology of the first terrestrial embryophytes (Fig. 2) based on the biology of extant non-vascular embryophytes (i.e., liverworts, mosses, and hornworts), the first plants to actually live on the land (literally) were gametophytes and not their associated sporophytes.^11,12 This postulate emerges from the fact that the sporophytes of liverworts, mosses, and hornworts invariably begin their development within archegonia and gain water and other nutrients by a basipetal “foot” (via matrotrophy) during their early ontogeny. This “model” for the dependency of the first embryophyte sporophytes on their gametophytes is consistent with (albeit not substantiated by) the megafossil remains of some of the earliest known vascular plant sporophytes, e.g., Cooksonia (Fig. 3). With the exception of one report purporting to show Cooksonia sporophytes attached at their base to a gametophyte,³ the proximal portions of all other Cooksonia sporophytes are broken and provide no evidence for what the morphology of the basal portions of the sporophyte or the gametophyte looked like. In passing, it is worth noting that the genus “Cooksonia” should be considered a form genus and not a “true genus." The morphological diversity of the sporangia on structures described as Cooksonia is so great that it is unlikely that all morphs belong to the same organism. It is nevertheless certain that many of the first vascular plants lacked roots owing to the exquisite preservation of the plants preserved in the Rhynie Chert (e.g., Rhynia and Horneophyton). These early land plants absorbed water and nutrients by means of rhizoids and their association with mycorrhizae.

Figure 2. Morphology of the sporophyte of the Lower Devonian land plant Zosterophyllum shengfengense, which may have inhabited moist (semi-aquatic) areas. Note that the basal part of the plant is depicted as consisting of numerous negatively gravitropic root-like organs. Adapted from Hao et al..¹³

Figure 3. Compression fossil assigned to the early Paleozoic genus Cooksonia and schematic of a stereotypical sporophyte (see Inset). (Specimen from the Cornell University fossil plant collection).

Finally, the oldest known fossil plant with roots, Zosterophyllum shengfengense from Southern China,¹⁴ predates the Rhynie chert and lacked leaves (see Fig. 2). Thus, until future discoveries in the fossil record indicate otherwise, the evolution of roots preceded the phylogenetic development of the organographic distinction between what is conventionally meant by “leaf” and “stem."

Boron as a multifunctional microelement

Seven years ago, we summarized paleobotanical as well as developmental and molecular data indicating that embryophytes originated from a common ancestor similar to extant freshwater charophycean algae.^11,12 In that context, we discussed the morphology of mature sporophytes of Zosterophyllum rhenatum, an Early Devonian (ca. 410 Myr-old) tracheophyte – the root-less specimen depicted in our article was collected in the Ahrtal/Rhineland, Germany. Concomitantly, Hao et al.¹³ described a newly discovered, uprooted specimen of Zosterophyllum from Southern China, Z. shengfengense. This ca. 413 Myr-old land plant has a large rhizome-like rooting system, and is considered to be the oldest known tracheophyte.² The re-constructed phenotype of Z. shengfengense (Fig. 2) depicts positively gravitropic axes attached to the shoot, which are reminiscent of adventitious roots in the majority of extant tracheophytes.

What is the significance of Lewis’ “Boron-hypothesis” with respect to the evolutionary appearance of roots? Before answering this question, it must be stated that the precise metabolic function of B (translocated via aquaporins) in plant development is unclear.⁵ Empirical observations, however, indicate that it is involved in cell expansion, nucleic acid synthesis, membrane function, hormone responses, pollen tube growth, and lignin biosynthesis (see, refs 14–18). For this reason, B-deficiency exhibits a variety of symptoms including the necrosis of leaf lamina and terminal buds, the suppression of apical dominance and cell expansion, and the breakdown of parenchyma in fleshy roots, tubers, and fruits. In Arabidopsis, B-limitation results in an inhibition of root elongation, accompanied by a boost in root hair elongation (which can also be induced by methylotrophic bacteria).^4,5

These phenomena are explicable in part because Boron plays an important role in embryophyte cell-to-cell adhesion. Specifically, B forms a covalent complex with Ca²⁺–rhamnogalacturonic II (RG-II), and the structure of the B-RG-II complex (which consists of 2 molcules of Boron, Ca²⁺, and 2 chains of monomeric RG-II) is ubiquitous in the sporophyte and gametophyte generations of vascular land plants.¹⁶

Root formation vs. the occurrence of rhizoids

In addition to the numerous roles B plays, Lewis¹ drew attention to the classic study of Middleton et al.,¹⁷ wherein it is shown that the growth of adventitious roots in bean (Phaseolus aureus)-cuttings is promoted by B. More recent studies verified and extended this observation. Josten and Kutschera¹⁸ documented that when light-grown seedlings of sunflower (Helianthus annuus) are de-rooted and incubated on nutrient solution containing or lacking boric acid ( ± B), adventitious roots developed in the + B-samples (up to 10 per cutting), but not in the – B-samples. This B-induced development of adventitious roots occurred in the absence of any externally applied phytohormones (e.g., auxins, gibberellins, cytokinins). Cytological studies revealed that B stimulated the meristematic activity of the internal cells of the hypocotyl, and hence initiated the development of juvenile roots at the cut end.¹⁸

Fig. 4A shows that 4-d-old light-grown seedlings, de-rooted and placed in an upright position 5 mm deep into moist vermiculite (nutrient solution ± B), develop a large system of adventitious roots in the presence of the micronutrient. The cuttings with adventitious roots were planted in soil and incubated in a light/day-growth chamber. All of these seedlings survived and flowered. Under similar experimental conditions, pieces of the liverwort Marchiantia polymorpha, placed on agar with a nutrient solution ( ± B), as described by Kutschera and Koopmann,¹⁹ developed numerous rhizoids, both in the presence and absence of the micronutrient (Fig. 4 B). These results show that adventitious roots are induced by B, whereas rhizoid development is independent of this micronutrient. This finding is consistent with the facts that borate-cross-linked pectic polysaccharides are present in the cell walls of sporophytes of vascular plants (sunflower etc.), but largely absent in the walls of the gametophytes of the nonvascular bryophytes (liverworts, mosses, hornworts).²⁰

Figure 4. Effect of boric acid (B) on the development of adventitious roots in cuttings from light-grown sunflower (Helianthus annuus) plants. Four-day-old seedlings were de-rooted, placed with the cut end in vermiculite moistened with nutrient solution ( ± B, 0.1 mM), and photographed one week later (A). Using the same nutrient solution ( ± B, 0.1 mM), sterile pieces of the thallus of the liverwort Marchantia polymorpha were incubated on agar-plates. Five weeks later, in both treatments numerous rhizoids had developed on the lower side of the bryophyte (original experiments).

Since the earliest vascular plants, such as Z. shengfengense (Fig. 2), may have developed roots that we today would call “adventitious," it is possible that the micronutrient B played a major role during the evolutionary development of root systems, as suggested by Lewis.¹ As noted, the molecular mode of B-action is largely unknown.²¹ However, additional functionalities for B that can affect root development have been discovered. For example, B deficiency is known to repress cytokinin receptor genes resulting in defective xylem differentiation,²² results in leaf growth defects,²³ and may cause an induction of aquaporin-genes.⁵ It also prevents rhamnogalacturon-II cross-linking by means of borate diesters, which form normally either intra-protoplasmically or during secretion.²⁴ Thus, B is critical to proper cell-to-cell signaling and adhesion, thereby influencing the mechanical properties of cells walls via the modulation of the metabolism of pectins.^25,26

Conclusions and outlook

Based on numerous experimental studies, we now know that B is involved in phytohormone signaling and influences the mechanical properties of intercellular pectins. From these data, we conclude that the role of Boron during the evolutionary development of lignified plant organs, i.e., roots and shoots,¹ should be explored in more detail to resolve root form vs. function relationships that are currently poorly understood.²⁷ Finally, we want to stress that root development is, under natural conditions, also regulated by epiphytic microbes that inhabit the rhizosphere.^4,28,29 The interaction between the micronutrient B and bacteria that live in (or are attached to) the root system is unknown. Hence, more integrative “plant-nutrients-microbe”-studies are required to resolve the puzzle of root development during the evolution of land plants.^2,3,7

Disclosure of potential conflicts of interest

No potential conflicts of interest were disclosed.

Funding

The cooperation of the authors is supported by the Alexander von Humboldt-Foundation, Bonn, Germany (AvH-Fellowship Stanford 2014 to U. K.).