Review

Why boron?

Luis Bolaños a,*, Krystyna Lukaszewski b, Ildefonso Bonilla a, Dale Blevins b

a Departamento de Biologia, Facultad de Ciencias, Universidad Autónoma de Madrid, 28049 Madrid, Spainb Plant Science Unit, University of Missouri, Columbia, MO 65211, USA

Received 9 June 2004; accepted 17 November 2004

Available online 16 December 2004

Abstract

It is now more than 80 years since boron was convincingly demonstrated to be essential for normal growth of higher plants. However, itsbiochemical role is not well understood at the moment. Several recent reviews propose that B is implicated in three main processes: keepingcell wall structure, maintaining membrane function, and supporting metabolic activities. However, in the absence of conclusive evidence, theprimary role of boron in plants remains elusive. Besides plants, growth of specific bacteria, such as heterocystous cyanobacteria and therecently reported actinomycetes of the genus Frankia, requires B, particularly for the stability of the envelopes that control the access of thenitrogenase-poisoning oxygen when they grow under N2-fixing conditions. Likewise, a role for B for animal embryogenesis and other devel-opmental processes is being established. Finally, a new feature of the role of boron comes from signaling mechanisms for communicationamong bacteria and among legumes and rhizobia leading to N2-fixing symbiosis, and it is possible that new roles for B, based on its specialchemistry and its interaction with Ca would appear in the world of signal transduction pathways. In conclusion, the diversity of roles playedby B might indicate that either the micronutrient is involved in numerous processes or that its deficiency has a pleiotropic effect. The arisingquestion is why such an element? Since all of the roles clearly established for B are related to its capacity to form diester bridges betweencis-hydroxyl-containing molecules, we propose that the main reason for B essentiality is the stabilization of molecules with cis-diol groupsturning them effective, irrespectively of their function.© 2004 Elsevier SAS. All rights reserved.

Keywords: Animal development; Bacteria signaling; B–Ca relationship; cis-Hydroxyl groups; Molecular linker; Signal transduction; Wall structure

1. Introduction

Boron is a member of the semiconductor group of ele-ments and has properties intermediate between metals andnon-metals. The boron atom is small with only three valenceelectrons. The chemistry of boron is unique and, after that ofcarbon, it might be the most intriguing and complex of anyelement [22]. Boron, along with other light elements likelithium and berilium, originates from the Big Bang nucleo-synthesis or galactic cosmic-ray events [34,40], and its abun-dance is extremely low: only about 10–9 times that of hydro-gen and about 10–6 that of carbon, nitrogen, or oxygen.However, in spite of its low cosmic abundance, boron iswidely distributed both in the Earth’s crust (from 5 mg kg–1

in basalts to 100 mg kg–1 in shales) [42] and in the ocean(~4.5 mg l–1) [26].

The boron requirement for plant growth was first demon-strated in the early 1920s [45], and since then boron has beenestablished as an essential micronutrient for all vascular plants.Boron deficiency causes a plethora of rapid biochemical,physiological, and anatomical aberrations, which, along withthe lack of relevant information on boron chemistry, havemade determining the primary function of boron in plantsone of the most difficult tasks in plant nutrition. While greatprogress has been made in the last few years, the primary roleof boron remains undefined. Ongoing research focuses onboron involvement in three main areas: organization of cellwalls, membrane function, and metabolic activities [4,14].

2. Boron and structures containing moleculeswith cis-hydroxyl groups

At the near-neutral pH found in most biological fluids,boron exists primarily (~96%) as boric acid, B(OH)3, plus asmall amount of borate anion B(OH)4

–. Both boric acid andborate readily form complexes with a wide variety of sugars

Abbreviations: FITC, fluorescein-isothiocyanate conjugate.* Corresponding author. Fax: +34 91 497 8344.

E-mail address: luis.bolarios@uam.es (L. Bolaños).

Plant Physiology and Biochemistry 42 (2004) 907–912

www.elsevier.com/locate/plaphy

0981-9428/$ - see front matter © 2004 Elsevier SAS. All rights reserved.doi:10.1016/j.plaphy.2004.11.002

and other compounds containing cis-hydroxyl groups (Fig. 1)[27]. Recent identification of the first boron-containing mol-ecules from plants has led to a breakthrough in boron research.Isolation and characterization of rhamnogalacturonan-II-boron (RG-II-B) complexes demonstrated boron cross-linkbetween apiose residues in pectin (Fig. 2), and confirmed invivo the proposed role of boron in cell wall architecture[24,31,32]. So far, no direct biochemical evidence has been

presented resolving boron action in membranes, but boronlinking or binding to hydroxyl-containing constituents, suchas phosphoinositides, glycoproteins, and glycolipids, has beenproposed to explain the altered membrane composition andtransport processes in boron deficient plants [14].

Although for most of the last eight decades boron require-ment has been recognized exclusively in plants, recent stud-ies demonstrate a nutritional importance of boron across abroad spectrum of organisms: yeasts [1], animals and human[29]. While the mechanisms of boron action in animals andhumans are unknown, a specific function in membranes hasbeen proposed [14,30].

At the opposite end of the biological spectrum, boronessentiality has been established for the growth of specifictypes of bacteria, such as heterocystous cyanobacteria [12],and actinomycetes of the genus Frankia [9]. Both types ofmicroorganisms require boron for the stability of the enve-lopes that prevent access of nitrogenase-poisoning oxygenwhen grown under N2-fixing conditions.As expected, the cellsthat specifically require boron are the nitrogenase-harboringheterocysts in Anabaena and vesicles in Frankia. Stabiliza-tion of these envelopes by boron is intriguing, considering

Fig. 1. Chemical structures of boric acid (A), borate anion (B), and their diolesters (C, D).

Fig. 2. Sites of boron attachment in biological structures: plant cell wall boron-rhamnogalacturonan II complex (A), bacterial quorum sensing signaling mole-cule autoinducer AI-2 (B), phloem boron transport complex with sorbitol (C), and hypothetical models of boron binding with second messenger GMP (D),bacteriohopanetetrol (E) and phosphoinositol IP3 (F).

908 L. Bolaños et al. / Plant Physiology and Biochemistry 42 (2004) 907–912

their different chemical composition. In the heterocysts, boronis present in an inner laminated layer formed of specific gly-colipids. In contrast, in Frankia, boron is found in a multil-aminate vesicle wall composed of glycolipids and neutral lip-ids with a very high proportion of long-chain polyhydroxyfatty acids or alcohols, including the hopanoid bacterio-hopanetetrol [2]. All of these components are rich in free diolgroups, ideal for bonding with borate (Fig. 2).

3. Boron and bacteria-plants talking

A role for boron has been demonstrated in the establish-ment of an effective legume–Rhizobium symbiosis [7]. As inplants, boron requirements have been reported for the main-tenance of nodule cell wall structure [13]. Furthermore, themicronutrient plays important roles for the correct establish-ment of the symbiosis.

The N2-fixing legume root nodule is the result of geneti-cally determined interactions between rhizobia and the hostplant [43]. As a result of molecular signaling, mediated byplant and bacteria derived glycoconjugates most of them richin cis-diols, the physical and metabolic integration betweenrhizobia and the host cells becomes progressively more inti-mate [23]. Boron is required for early symbiont/plant signal-ing, namely nod-gene activation by root plant exudates andnodule initiation [38]. Moreover, boron is required for infec-tion thread development and nodule invasion [5], due to arole of B as modulator of the interactions between plantsderived infection thread matrix glycoproteins and the bacte-ria cell surface. In the absence of B, the glycoproteins canattach to the cell surface of rhizobia. Therefore, the bacte-rium can be trapped and unable to interact with the plant cellmembrane and hence elicitation of the endocytosis processleading to cell invasion is inhibited. The presence of B (butnot Ca, pH changes, salt or high ionic strength) specificallyinhibits the in vitro bacteria-matrix glycoprotein attachmentand promotes the rhizobial interaction with the plant mem-brane prior cell invasion. Once Rhizobium is inside the cell Bpromotes symbiosome (containing N2-fixing bacteroids)development. Specifically, B is needed for the targeting ofnodule-specific plant derived glycoproteins [6] that are cru-cial as signals for bacteroid differentiation into a N2-fixingform [10].

Finally, the recent discovery of a boron-containing bacte-rial signal molecule, autoinducer AI-2, revealed an unex-pected role for boron in bacterial quorum sensing (Fig. 2)[15]. Quorum sensing allows bacterial populations to moni-tor cell density by means of diffusible pheromones, whichaccumulate in the extracellular space as the populationincreases. Higher levels of pheromones activate signalingpathways, leading to coordinated alteration of gene expres-sion throughout the population [16]. AI-2, identified as a fura-nosyl borate diester, is a novel signaling molecule for both,structure and function. In contrast to the previously knownbacterial pheromones, AI-2 is not exclusive to a single spe-

cies, but is produced by a multitude of diverse bacteria, andthe gene encoding the AI-2 synthase (luxS) is widely con-served. This raises a possibility that AI-2 might serve as auniversal bacterial signal for communication among species[15,47] could also serve as a boron transporter, a way of mov-ing boron in or out of the cell, depending on growth or envi-ronmental conditions [16]. It remains unclear at what stageboron binds to the carbohydrate moiety of the AI-2 mol-ecule; in the autoinducer-producing cell, in the surroundingenvironment, or in the recipient cell [16].

4. Boron (and the B–Ca relationship) in signaltransduction and gene expression

Another insight into boron function from the perspectiveof gene expression comes from the area of nitrogen fixation.Boron was found to be essential for bacteria-plant signalingand nod-gene induction in pea [38]. Addition of calcium ame-liorated some effects of boron deficiency on nodulation, andincreased nodule number [39]. Interaction between calciumand boron has also been reported in pea nodule establish-ment and development under salt stress [19,20]. Genetic stud-ies of nodulation of Medicago truncatula showed that expres-sion of 60% of the analyzed genes (including genes involvedin cell cycle, cell wall assembly, and ribosome biogenesis)was affected by boron deficiency, and, in some cases, over-expression could be reversed by supplemental calcium[36,37]. However, calcium did not reverse either the abnor-mal cell wall structure of boron deficient nodules or the dis-tribution of pectic polysaccharides immunolocalized bymonoclonal antibodies in cell [39].

Calcium association with boron has been observed inplants, bacteria, animals and humans, but the nature of thisinteraction is still debated. Various types of evidence suggesta role of calcium in stabilizing boron complexes. Pectin-associated calcium in tomato was reduced under boron defi-ciency [48]. Calcium inhibited decomposition of cell wallRG-II-B dimmers in vitro and in vivo [25,46]. Calcium medi-ated the recovery of boron deficient, nitrogen-fixing cyano-bacteria Anabaena [8], and non-fixing Synechococcus [11],by stabilizing their envelopes. Additionally, the amount ofmembrane-bound calcium in roots of Vicia faba was downwithin hours of boron deprivation [28]. Recent studies on thestability of different boron fractions in intact roots suggestthat boron cross-linking of RG-II results in a conformationchange that creates binding sites for calcium ions, increasingstrength of the B-RG-II complex [25]. It has been proposedthat the same mechanism could apply to boron complexes inmembranes [46].

In plants adequately supplied with boron, as much as 60%of the boron’s total content can be present in soluble form[17,31,33]. However, except for complexes with polyols, littleis known about boron binding inside the cell. Consideringthe calcium effect on boron-responsive genes in Medicago,we envision boron–calcium interplay in signaling events. Well

909L. Bolaños et al. / Plant Physiology and Biochemistry 42 (2004) 907–912

known regulatory and/or signaling molecules are potentialtarget for interaction with boron. Nucleic acids and nucle-otides, adenylates, guanylates, oxidized nicotinamides, arecapable of forming stable boron complexes, as their ribosecomponent has the optimal configuration for borate esterifi-cation (Fig. 2). Phosphoinositides are also potential targetsfor boron binding (Fig. 2), which would place boron in thePI, PIP, PIP2, IP3-modulated signal transduction, thus in theIP3-mediated calcium release from the endoplasmic reticu-lum or tonoplast into the cytosol, activation of calcium-dependent protein kinases, and changes in gene expression.This binding would also affect anchoring of certain proteinsto cell membranes. But these are just possibilities. Unfortu-nately, the very limited information currently available onboron biological complexation does not allow us to speculatewhich of the boron-attracting molecules are the bona fide tar-get in biological systems, and this is the conundrum of boronresearch.

5. The problem of discovering boron-dependentbiomolecules

The development of methods to identify boron ligandsamong molecules isolated from biological samples is impera-tive. Capillary electrophoresis has been used to detect, quan-tify and compare in vitro boron binding by adenylates andnicotinamides [35]. The development of markers that bind tocis-diols in the same manner than borate, such fluorescein-isothiocyanate-boronic-acid conjugate (B-FITC), may behelpful for mapping borate-binding sites in cells [21]. Edel-man et al. [18] used immunogold-labeled anti-FITC antibod-ies and electron microscopy to detect B-FITC binding notonly in the cell wall but also in membranes and cytoplasm.Plasma membrane glycoprotein with capacity to bind B couldtherefore been identified following electrophoresis, westernblotting, incubation with B-FITC and detection with anti-FITC antibodies.

Those techniques that show B-binding capacity, alwaysrequire artificial enrichment with either borate or binding-sites markers, and cannot be used for in vivo detection andidentification of borate diester complexes. The absence of aradioactive isotope makes difficult labeling of such com-plexes. However, natural boron is a mixture of two stable iso-topes, 10B and 11B, and the development of techniques thatdiscriminate between both isotopes appear as the most usefulmethodology. There are a number of techniques routinelyemployed for analysis of boron isotopes in materials sciencematrices, but only a few are sensitive enough to study boronat the trace level [44]. Among nuclear techniques, 11B NMRseems to be a powerful tool not only for detection of boratecross-linked biomolecules, but also for the analysis of the typeof borate complex, as recently reported for guar gum [3]. Thestriking finding by 11B NMR that AI-2 ligand is a furanosylborate diester [15] opens the possibility that new B com-plexes can be found in living organisms following applica-tion of 11B-NMR to biological samples.

6. Conclusion: boron gets cis-hydroxyl-containingmolecules working

The cumulative evidence from plant, bacteria, animal, andhuman experiments, shows boron as a dynamic trace elementaffecting an exceptionally large number of seemingly unre-lated biological functions. The diverse responses to boron dep-rivation in a broad range of organisms indicate either boroninvolvement in a broad spectrum of processes, or pleiotropiceffects. Our current hypothesis is that the primary role ofboron in biological systems is stabilization of molecules withcis-diol groups, independently on their function. It is pos-sible that new roles for B, based on its special chemistry wouldappear. In fact, last exciting report on this topic proposed thatborate minerals could play a crucial role in an early “RNAworld” of life on Earth by stabilizing cyclic ribose [41]. Therequirement of boron for cross-linking the pectin componentRG-II in plant cell walls [31], for vesicle targeting and trans-membrane transport in symbiosomes [6], or as a ligand in thecyclic furanosyl bacterial quorum sensing signal AI-2 [15],provides strong evidence for boron function as a “molecularlinker”, and supports our hypothesis.

Coulthurst et al. [16], referring to the newly found role forboron in bacterial signaling asked, “Why boron?” This ques-tion remains relevant in other biological contexts. We con-clude that boron chemistry makes it a perfect candidate foratomic diester bridging. Although other atoms, such as phos-phorus or sulfur, could form links through diester bridgesinstead of boron, the resulting configuration would be unstabledue to markedly greater electron density of those heavieratoms. At this moment, unraveling the occurrence of boroncomplexes in biological systems has only begun. While muchadditional research will be necessary, our model for boronfunction provides a single unifying explanation for what haspreviously seemed an incomprehensibly diverse range of bio-logical roles. Although analytical instruments and proce-dures have steadily improved during the last decade, furtherprogress greatly depends on the development of new meth-odology with higher capability for analytical imaging of boronisotopes at physiological concentrations in plant tissues.

Acknowledgements

This work was supported by MCYT no. BOS2002-04164-CO3-02. Ildefonso Bonilla was granted by MECD Programa“Salvador de Madariaga” 2002 for a stay in University ofMissouri, Columbia.

References

[1] A. Bennett, R.I. Rowe, N. Soch, C.D. Eckhert, Boron stimulates yeast(Saccharomyces cerevisiae) growth, J. Nutr. 129 (1999) 2236–2238.

[2] A.M. Berry, R.A. Moreau, A.D. Jones, Bacteriohopanetetrol: abun-dant lipid in Frankia cells and in nitrogen-fixing nodule tissue, PlantPhysiol. 95 (1991) 111–115.

910 L. Bolaños et al. / Plant Physiology and Biochemistry 42 (2004) 907–912

[3] M. Bishop, N. Shahid, F. Yang, A.R. Barron, Determination of themode and efficacy of the cross-linking of guar by borate using MAS11B NMR of borate cross-linked guar in combination with solution11B NMR of model systems, Dalton Trans 17 (2004) 2621–2634.

[4] D.G. Blevins, K.M. Lukaszewski, Boron in plant structure and func-tion, Annu. Rev. Plant Physiol. Plant Mol. Biol. 49 (1998) 481–500.

[5] L. Bolaños, N.J. Brevin, I. Bonilla, Effects of boron on Rhizobium–legume cell-surface interactions and nodule development, PlantPhysiol. 110 (1996) 1249–1256.

[6] L. Bolaños, A. Cebrián, M. Redondo-Nieto, R. Rivilla, I. Bonilla,Lectin-like glycoprotein PsNLEC-1 is not correctly glycosylated andtargeted in boron-deficient pea nodules, Mol. Plant-Microbe Interact.14 (2001) 663–670.

[7] L. Bolaños, E. Esteban, C. de Lorenzo, M. Fernández-Pascual,M.R. de Felipe, A. Gárate, et al., Essentiality of boron for symbioticdinitrogen fixation in pea (Pisum sativum)–Rhizobium nodules, PlantPhysiol. 104 (1994) 85–90.

[8] L. Bolaños, P. Mateo, I. Bonilla, Calcium-mediated recovery of borondeficient Anabaena sp. PCC 7119 grown under nitrogen fixing condi-tions, J. Plant Physiol. 142 (1993) 513–517.

[9] L. Bolaños, M. Redondo-Nieto, I. Bonilla, L.G. Wall, Boron require-ment in the Discaria trinervis (Rhamnaceae) and Frankia symbioticrelationship. Its essentiality for Frankia BCU110501 growth andnitrogen fixation, Physiol. Plant. 115 (2002) 563–570.

[10] L. Bolaños, M. Redondo-Nieto, R. Rivilla, N.J. Brewin, I. Bonilla,Cell surface interactions of Rhizobium bacteroids and other bacterialstrains with symbiosomal and peribacteroid membrane componentsfrom pea nodules, Mol. Plant-Microbe Interact. 17 (2004) 216–223.

[11] I. Bonilla, L. Bolaños, P. Mateo, Interaction of boron and calcium inthe cyanobacteria Anabaena and Synechococcus, Physiol. Plant. 94(1995) 31–36.

[12] I. Bonilla, M. García-González, P. Mateo, Boron requirement inCyanobacteria. Its possible role in the early evolution of photosyn-thetic organisms, Plant Physiol. 94 (1990) 1554–1560.

[13] I. Bonilla, C. Mergold-Villasenor, M.E. Campos, N. Sánchez,H. Pérez, L. López, et al., The aberrant cell walls of boron-deficientbean root nodules have no covalently bound hydroxyproline/proline-rich proteins, Plant Physiol. 115 (1997) 1329–1340.

[14] P.H. Brown, N. Bellaloui, M.A. Wimmer, E.S. Bassil, J. Ruiz, H. Hu,H. Pfeffer, F. Dannel, V. Romheld, Boron in plant biology, Plant Biol.4 (2002) 203–223.

[15] X. Chen, S. Schauder, N. Potier, A. Van Dorsselaer, I. Pelczer,B. Bassler, F.M. Hughson, Structural identification of a bacterialquorum-sensing signal containing boron, Nature 415 (2002) 545–549.

[16] S.J. Coulthurst, N.A. Whitehead, M. Welch, G.P.C. Salmond, Canboron get bacteria talking? Trends Biochem. Sci. 27 (2002) 217–219.

[17] F. Dannel, H. Pfeffer, V. Romheld, Update on boron in higherplants—uptake, primary translocation and compartmentation, PlantBiol. 4 (2002) 193–204.

[18] H.G. Edelman, M.A. Wimmer, H.E. Goldbach, Immunogold labellingof boric acid binding sites by immunoreaction to the FITC-boronicacid: first results, J. Trace Microprobe Tech. 18 (2000) 451–459.

[19] A. El-Hamdaoui, M. Redondo-Nieto, R. Rivilla, I. Bonilla,L. Bolaños, Effects of boron and calcium nutrition on the establish-ment of the Rhizobium leguminosarum-pea (Pisum sativum) symbio-sis and nodule development under salt stress, Plant Cell Environ. 26(2003) 1003–1011.

[20] A. El-Hamdaoui, M. Redondo-Nieto, B. Torralba, R. Rivilla, I. Bon-illa, L. Bolaños, Influence of boron calcium on the tolerance to salinityof nitrogen-fixing pea plants, Plant Soil 251 (2003) 93–103.

[21] K.H. Glüsenkamp, H. Kosegarten, K. Mengel, F. Grolig, A. Esch,H.E. Goldbach, A fluorescein boronic acid conjugate as a marker forborate binding sites in the apoplast of growing roots of Zea mays L.and Helianthus annuus L, in: R.W. Bell, B. Rerkasem (Eds.), Boron inSoils and Plants, Kluwer Academic Publishers, Dordrecht, 1997, pp.229–235.

[22] N.N. Greenwood,A. Earnshaw, Chemistry of the Elements, PergamonPress, New York, 1984.

[23] E.L. Kannenberg, N.J. Brewin, Host-plant invasion by Rhizobium: therole of cell-surface components, Trends Microbiol. 2 (1994) 277–283.

[24] M. Kobayashi, T. Matoh, J. Azuma, Two chains of rhamnogalactur-onan II are cross-linked by borate-diol ester bonds in higher plant cellwalls, Plant Physiol. 110 (1996) 1017–1020.

[25] M. Kobayashi, H. Nakagawa, T. Asaka, T. Matoh, Borate-rhamnogalacturonan II bonding reinforced by Ca2+ retains pecticpolysaccharides in higher plant cell walls, Plant Physiol. 119 (1999)199–204.

[26] D. Lemarchand, J. Gaillardet, E. Lewin, C.J. Allegre, The influence ofrivers on marine boron isotopes and implications for reconstructingpast ocean pH, Nature 408 (2000) 951–954.

[27] W.D. Loomis, R.W. Durst, Chemistry and biology of boron, Biofac-tors 3 (1992) 229–239.

[28] K.H. Muhling, M. Wimmer, H.E. Goldbach, Apoplastic andmembrane-associated Ca2+ in leaves and roots as affected by borondeficiency, Physiol. Plant. 102 (1998) 179–184.

[29] F.H. Nielsen, The nutritional importance and pharmacological poten-tial of boron for higher animals and human, in: H.E. Goldbach,B. Rerkasem, M.A. Wimmer, P.H. Brown, M. Thellier, R.W. Bell(Eds.), Boron in Plant and Animal Nutrition, KluwerAcademic/Plenum Publishers, New York, 2002, pp. 37–50.

[30] F.H. Nielsen, J.G. Penland, Boron supplementation of per-menopausal women affects boron metabolism and indices associatedwith macromineral metabolism, hormonal status and immune func-tion, J. Trace Elem. Exp. Med. 12 (1999) 251–261.

[31] M.A. O’Neill, S. Eberhard, P. Albersheim, A.G. Darvill, Requirementof borate cross-linking of cell wall rhamnogalacturonan II for Arabi-dopsis growth, Science 294 (2001) 846–849.

[32] M.A. O’Neill, D. Warrenfeltz, K. Kates, P. Pellerin, T. Doco, A. Dar-vill, et al., Rhamnogalacturonan-II, a pectic polysaccharide in thewalls of growing plant cell, forms a dimer that is covalently cross-linked by a borate ester, J. Biol. Chem. 271 (1996) 22923–22930.

[33] H. Pfeffer, F. Dannel, V. Romheld, Boron compartmentation in rootsof sunflower plants of different boron status: a study using the stableisotopes 10B and 11B adopting two independent approaches, Physiol.Plant. 113 (2001) 346–351.

[34] J.X. Prochaska, J.C. Howk, A.M. Wolfe, The elemental abundancepattern in a galaxy at z = 2.626, Nature 423 (2003) 57–59.

[35] N.V.C. Ralston, C.D. Hunt, Diadenosine phosphates andS-adenosylmethionine: novel boron binding biomolecules detected bycapillary electrophoresis, Biochim. Biophys. Acta 1527 (2001)20–30.

[36] M. Redondo-Nieto, Boron and Calcium Relationship in Rhizobium–Legumes Symbioses, Ph.D. Thesis, Universidad Autónoma deMadrid, Madrid, 2002.

[37] M. Redondo-Nieto, P. Mergaert, A. Kondorosi, E. Kondorosi, I. Bon-illa, L. Bolaños, Nutritional Influence of Boron and Ca2+ on NoduleOrganogenesis in Legumes, Fifth European Nitrogen Fixation Con-ference, Abstract 8.22, Norwich, 2002.

[38] M. Redondo-Nieto, R. Rivilla, A. El-Hamdaoui, I. Bonilla,L. Bolaños, Boron deficiency affects early infection events in thepea-Rhizobium symbiotic interaction, Aust. J. Plant Physiol. 28(2001) 819–823.

[39] M. Redondo-Nieto, A. Wilmot, A. El-Hamdaoui, I. Bonilla,L. Bolaños, Relationship between boron and calcium in the N2-fixinglegume–rhizobia symbiosis, Plant Cell Environ. 26 (2003) 1905–1915.

[40] H. Reeves, The origin of the light elements in the early Universe, in:The Century of Space Science, Kluwer Academic Publishers, Dor-drecht, 2001, pp. 423–440.

[41] A. Ricardo, M.A. Carrigan, A.N. Olcott, S.A. Benner, Borate mineralsstabilize ribose, Science 303 (2004) 196.

[42] V.M. Shorrocks, The occurrence and correction of boron deficiency,Plant Soil 193 (1997) 121–148.

[43] J. Stougaard, Regulators and regulation of legume root nodule devel-opment, Plant Physiol. 24 (2000) 531–540.

911L. Bolaños et al. / Plant Physiology and Biochemistry 42 (2004) 907–912

[44] M. Thellier, A. Chevalier, I. His, M. Jarvis, M.A. Lovell, C. Ripoll,D. Robertson, W. Sauerwein, M.C. Verdus, Methodological develop-ments for application to the study of physiological boron and to boronneutron capture therapy, J. Trace Microprobe Tech. 19 (2001) 623–657.

[45] K. Warington, The effect of boric acid and borax on the broad beenand certain other plants, Ann. Bot. (Lond.) 37 (1923) 629–672.

[46] M.A. Wimmer, H.E. Goldbach, Influence of Ca2+ and pH on thestability of different boron fractions in intact roots of Vicia faba L,Plant Biol. 1 (1999) 632–637.

[47] S. Winans, Bacterial esperanto, Nat. Struct. Biol. 9 (2002) 83–84.[48] T.Yamauchi, T. Hara,Y. Sonoda, Distribution of calcium and boron in

the pectin fraction of tomato leaf cell wall, Plant Cell Physiol. 27(1986) 729–732.

912 L. Bolaños et al. / Plant Physiology and Biochemistry 42 (2004) 907–912