Elements Covers

Borate and the Origin of RNA: A Model for the Precursors to Life

According to the RNA World hypothesis, ribonucleic acid (RNA) played a critical role in the origin of life. However, ribose, an essential component of RNA, is easily degraded: finding a way to stabilize it is critical to the viability of the hypothesis. Borate has been experimentally shown to have a strong affinity for ribose, and, thus, could have protected ribose from degradation in the formose reaction, a potential process for prebiotic ribose formation. Accumulation of borate on Hadean Earth (prior to ~4,000 Ma) might have been a key step in the chemical evolution of the biotic sugar. Proto-arcs are suggested as a geological setting sufficiently rich in borate to stabilize ribose during the Hadean.

DOI: 10.2138/gselements.13.4.261

Keywords: Hadean Earth, proto-arc, ribose, RNA, prebiotic


Many questions remain concerning the prebiotic spontaneous formation of the building blocks of life and their polymerization to form functional polymers on Earth prior to ~4,000 Ma, i.e. during the Hadean. More specifically, according to the RNA World hypothesis, ribose (a type of sugar) [(1) in Fig. 1] played a critical role. However, it is one of the most unstable molecules among all the major potential building blocks. Thus, both the spontaneous formation and the accumulation of ribose is thought to be essential for the origin of life. In this article, we review experiments showing how borate plays an important chemical role in the formation and accumulation of ribose and of borate’s role in nucleoside phosphorylation, a critical step in assembling a phosphate and a nucleoside to form a monomeric unit of RNA. Furthermore, this article discusses the surface environments on Hadean Earth and introduces a plausible geological scenario for how and where borate might have accumulated during the Hadean.

  • Facebook
  • Twitter
  • Google+

Figure 1. Schematic of the formose reaction. Formaldehyde (5) and glycolaldehyde (6) react to form sugars including ribose (1). Ribose is a minor product. Successive reactions degrade ribose into brown caramel.

Prebiotic Chemistry on Borate and Ribose

The RNA World Hypothesis and the Difficulty of Stabilizing Ribose

Sugars are an indispensable part of the functional biomolecules that are used for the storage and expression of genes, for cell membranes, and for carrying energy. In ribonucleic acid (RNA) and deoxyribonucleic acid (DNA), ribose or 2-deoxyribose forms a structural framework with phosphate and is connected with nucleobases to form the sequence of the gene code. RNA is usually a single-stranded molecule that has a sequence of nucleobases comparable to DNA. The potential importance of RNA in the origin of life was proposed in the 1960s and later named the RNA World hypothesis (Crick 1968; Joyce 1989) In all living forms, gene information is stored in DNA as a sequence of nucleobases, whereas proteins are constructed from the gene information via RNA. Proteins catalyze many biochemical reactions, including the formation of DNA and RNA. Thus, the three complex biopolymers of RNA, DNA, and proteins work together in all living forms.

It does not seem plausible that these three complex biopolymers arose simultaneously and then self-assembled into a complementary system within a cell compartment on prebiotic Earth. But, a scenario has been proposed whereby one type of polymer (RNA) works both as catalyst and gene-code template such that the system composed of RNA might have been ancestral to the present DNA–RNA–protein system of life (Crick 1968). Subsequently, researchers discovered that several RNAs catalyze biochemical reactions, as do enzymes.

Ribose is an aldopentose—a five-carbon monosaccharide with an aldehyde functional group—and the sole sugar in RNA (Fig. 1). It has many structural isomers, i.e. sugars having the same chemical composition but different structures, which can be seen in such compounds as xylose [(2) in Fig. 1], lyxose [(3) in Fig. 1], arabinose [(4) in Fig. 1], ribulose, and xylulose. They all form in the formose reaction in which formaldehyde [(5) in Fig. 1] and glycolaldehyde [(6) in Fig. 1] react in alkaline solutions with a cation catalyst such as Ca2+ and Mg2+ (e.g. Shapiro 1988). Experimental formation of detectable amounts of ribose in the formose reaction has been reported by Shapiro (1988). However, the reaction is sequential and the successive polymerization consumes previously formed sugars, including ribose. Ultimately, all the products polymerize into a brown caramel substance, a complex mixture of sugar polymers and their degradation products. Furthermore, the yield of ribose is limited in the formose reaction products compared with other aldopentoses because ribose is the least stable aldopentose (Larralde et al. 1995). Due to this low stability of ribose, some researchers have questioned the spontaneous accumulation of ribose and the subsequent formation of RNA on the prebiotic Earth (e.g. Shapiro 1986; Larralde et al. 1995).

Borate Stabilization of Ribose may be Critical to Forming Prebiotic RNA

Experiments support the suggestion that the preservation of ribose might have been mediated by borate. Borate is an oxyanion of B3+ and, in a wider sense, ‘borate’ refers to any compound containing an oxyanion and hydroxyl anion of boron, e.g. B(OH)3 and B(OH)4. Thus, in this wider definition, the term ‘borate’ is also used to refer to minerals containing boron oxide/hydroxide, such as colemanite [CaB3O4(OH)3·H2O]. In the present paper, oxyanions of B3+ in solution, such as B(OH)3 and B(OH)4, are referred to as ‘borate’, whereas solids containing borate, such as colemanite, are referred to as borate minerals.

Speciation of borate in solution depends on the solution pH: at pH < 9, trigonal B(OH)3 is the dominant species, whereas at pH > 9, tetragonal B(OH)4 dominates. The B(OH)4 species can combine with a hydroxyl to form an ester (Van Duin et al. 1984), with cis-diols, in particular, forming a stable cyclic ester (Fig. 2B). Thus, sugars typically form complexes with tetragonal B(OH)4. Research into how pentose–borate complexes form has shown that borate combines with the 1',2'-diol and with the 2',3'-diol of ribose to form various ribose–borate complexes (Chapelle and Verchere 1988) (Fig. 2B). In aqueous solutions, each aldopentose appears in five different configurations (a-furanose, b-furanose, a-pyranose, b-pyranose, and a linear aldehyde form). The proportions of ribose at equilibrium in water are 9% in the aldehyde form, 70% in the a- and b-pyranose form, and 21% in a- and b-furanose form (Fig. 2). When ribose forms a complex with borate, all ribose molecules that are fixed in the furanose form are far less reactive than when in the aldehyde form, thereby decreasing the degradation rate of ribose (Chapelle and Verchere 1988). Among the four aldopentoses, borate forms complexes preferentially with ribose (Chapelle and Verchere 1988; Furukawa et al. 2013).

  • Facebook
  • Twitter
  • Google+

Figure 2. Epimeric forms of ribose. In solution, ribose molecules have five different configurations: two five-member rings (a- and b-ribofuranose), two six-member rings (a- and b-ribopyranose), and a linear form (aldehyde). Ribose molecules continuously change their epimeric form via the linear aldehyde even after the fractions of the emipers reach equilibrium. (A) Equilibrium of epimers of ribose in borate-free solutions. The a- and b-pyranose are dominant but other forms, including the aldehyde form, are present. The aldehyde form is the most reactive and works as the channel for degradation of pentoses to brown caramel. (B) Epimers of ribose in borate-rich solutions. More than the other sugars, ribose is fixed in the furanose form by complexing with borate through ester bonds. The fraction of aldehyde form is reduced, minimizing ribose degradation to brown caramel. (C) Incubation of formaldehyde/glycolaldehyde solution containing Ca(OH)2 (at pH 12) at 60 °C for 1 hour. The left bottle contains sodium borate (80 mM borate), which stabilize the sugars that form as a result of the formose reaction. Conversely, the right bottle does not contain any borate. In the borate-free solution, aldehydes and sugars are polymerized, producing brown caramel.

The formose reaction is promoted in high-pH solutions (Fig. 2C). However, the stability of ribose and other sugars decreases with increasing pH (Larralde et al. 1995) because the reactivity of ribose increases with increasing pH. The dissolved borate species B(OH)4 does, in fact, lead to the condensation of glyceraldehyde and glycolaldehyde to form ribose in better yields than the reactions in borate-free solutions (Ricardo et al. 2004). The stability of ribose itself was shown to increase in borate solutions (Scorei and Cimpoias u 2006). Furukawa et al. (2013) compared the effects of borate, B(OH)4, on all four aldopentoses and found that ribose is selectively stabilized by borate in a concentrated borate solution (80 mM) but the other aldopentoses continued to degrade. The strong affinity of borate does selectively stabilize ribose, the least stable of the aldopentoses, yet ribose is the only aldopentose found in RNA. Dissolved silicate, Si(OH)5, has also been proposed as a stabilizer of sugars in the formose reaction under different conditions from those experienced by borate (Lambert et al. 2010). Quantitative comparison experiments showed that silicate certainly improves the stability of pentoses, but the effect is smaller than that of borate and does not lead to the selective stabilization of ribose (Nitta et al. 2016). Phosphate is also an essential component of RNA because it combines with ribose to form the sugar–phosphate backbone of RNA. But, the effects of dissolved phosphate on the stabilization of pentoses are negligible (Nitta et al. 2016). These results suggest that borate is the most effective simple compound for stabilizing ribose in abiotic environments.

Experiments also reveal that borate can affect the phosphorylation of ribonucleoside, a reaction for which a phosphate group combines with nucleoside, itself composed of a ribose and a nucleobase. This reaction is a critical step in the assembly of ribonucleotide, a monomeric unit of RNA. During phosphorylation of ribonucleoside, phosphate randomly forms esters at three different hydroxyl positions in ribonucleoside: 2'-, 3'-, or 5'-hydroxyl, as shown for the adenosine structure in Figure 3. The products of phosphorylation at 3'- and 5'-hydroxyl are the ribose–phosphate linkages found in terrestrial biota. Curiously, the 2'-hydroxyl form of ribose–phosphate has not been reported.

  • Facebook
  • Twitter
  • Google+

Figure 3. Formation of nucleoside–borate complex and selective phospholylation. In solutions free from borate, phosphorylation can occur at all hydroxyl groups: 2'-, 3'-, and 5'-hydroxyl. However, in the presence of borate, borate forms a complex with the 2'-hydroxyl and 3'-hydroxyl of a ribose moiety, forming a nucleoside–borate complex (7). Subsequently, 5'-hydroxyl is selectively phosphorylated (8).

Ribonucleoside can form a complex with borate at its 2',3'-diol position when in water and also with formamide [(7) in Fig. 3] (Furukawa et al. 2015; Kim et al. 2016). As illustrated in Figure 3, the bound borate inhibits phosphorylation at 2'- and 3'-hydroxyl, which results in the selective phosphorylation at the 5'-hydroxyl to form 5'-monophosphate [(8) in Fig. 3] (Furukawa et al. 2015; Kim et al. 2016). Activated 5'-monophosphate can form long ­oligonucleotides when dried in the presence of montmorillonite, a clay mineral catalyst (Ferris et al. 1996). Thus, the formation of 5'-monophosphate would be an important step to forming primordial RNA. And although long oligomers can be formed from 2',3'-cyclic monophosphate, further investigations are needed to understand exactly how. Unlike borate, dissolved silicate has almost no effect on the selection of the phosphorylation site of a ribonucleoside (Furukawa et al. 2015).

Borate also mitigates another serious problem in phosphorylation: the low reactivity of phosphate minerals. There is not sufficient phosphate dissolved in natural aqueous environments for phosphorylation to easily proceed because most phosphate is bound in the relatively insoluble minerals of the apatite group. However, Kim et al. (2016) showed that the magnesium phosphate borate mineral lüneburgite {Mg3[B2(OH)6(PO4)2]·6H2O} is effective in phosphorylating a nucleoside: it can selectively yield good amounts of nucleoside 5'-monophosphate in aqueous solutions. In this latter case, borate is extracted from lüneburgite and subsequently bound to the 2'-hydroxyl and 3'-hydroxyl of ribose. The phosphate released from lüneburgite is then utilized for phosphorylation of a nucleoside before it can precipitate as insoluble phosphates. The formation of Mg-borate mineralsduring the Hadean has been suggested by Holm (2012). Thus, borate could have mediated selective phosphorylation on the Hadean Earth.

Borate In The Early Archean And Hadean

Borate-rich Environments in the Early Archean

The strong chemical affinity between ribose and borate suggests that ribose could have accumulated in places where borate is highly concentrated. Enrichment of borate is commonly found in modern continental evaporite basins, followed by the precipitation of borate minerals, e.g. colemanite. Such evaporite basins have been considered as ideal localities to promote prebiotic ribose formation (Ricardo et al. 2004). But Grew et al. (2011) questioned whether continental evaporites were present during the Hadean. This skepticism has now resulted in contradictory models as to where prebiotic ribose might have formed.

The surface environments of early Mars are often considered as “suitable” for prebiotic ribose formation. The NASA Curiosity rover detected 10–100 ppm B in calcium sulfate veins in Mars’ Gale Crater (Gasda et al. 2016), which is comparable to amounts present in terrestrial clay-rich sediments (~100 ppm). This discovery supports the suggestion that the first prebiotic ribose may have formed on ancient Mars (Kirshvink et al. 2006). This model presumes not only that the water-rich ancient Earth could not concentrate borate and, therefore, was unsuitable for ribose genesis, but also that any RNA generated on Mars then survived the trip from Mars to Earth. However, there remains as much ambiguity regarding the availability of concentrated borate and the feasibility of the formose reaction on early Mars as on early Earth.

The Hadean Earth has attracted the interest of researchers in ribose genesis. Modern marine sediments can contain up to 258 ppm B (Williams et al. 2001). Seawater alteration of oceanic volcanic rocks is often accompanied by boron enrichment, particularly when associated with ultramafic rocks (Boschi et al. 2008). Boron in altered ultramafic rocks is released during the dehydration of a subducting plate, borate accumulating in the released alkaline fluids. The borate-rich alkaline fluids discharge on the seafloor to form mud volcanoes in the forearc regions. Holm (2012) suggested that such borate-rich alkaline fluids were present even in a Hadean subduction zone, and such fluids were responsible for ribose formation. However, relatively few scientists believe that a modern-style subduction system, with its dehydration at specific depths and concomitant borate release from altered rocks, was present during the Hadean.

Did Prebiotic Ribose Formation Occur in a Proto-arc?

The presence of an accreted ophiolite complex in the Isua Supracrustal Belt (ISB) of west Greenland provides evidence for plate tectonics in the early Archean (Furnes et al. 2007). A proto-arc model has been proposed to explain tectonic evolution of the 3.8 Ga to 3.7 Ga ISB (Nutman et al. 2015). In this model, segments of oceanic crust were accreted to form a proto-arc, while tonalite–trondhjemite–granite (TTG) suites were generated by melting accreted materials at the base of the proto-arc (Fig. 4).

  • Facebook
  • Twitter
  • Google+

Figure 4. Proto-arc model for the early Archean and Hadean, and the proposed locations of Hadean boron-rich environments (Grew et al. 2015; Nutman et al. 2015). White dots represent boron-rich clays. TTG = tonalite–trondhjemite–granite suite

Tourmaline, a chemically diverse group of borosilicate minerals, has been widely reported in the ISB (e.g. Appel 1995; Mishima et al. 2016). Based on boron isotope compositions, Grew et al. (2015) suggested that the concentration of boron necessary for the generation of abundant tourmaline in the ISB was attained in a partially isolated basin by hydrothermal processes in a proto-arc setting (Fig. 4). Boron could have been initially extracted by deep fluids from the TTG suites, from accreted sediments (shown brown with white dots in Fig. 4), and/or from mafic/ultramafic rocks (shown in green and blue with white dots in Fig. 4). These deep fluids were discharged into the ancient oceans as hydrothermal fluids. Nutman et al. (2016) reported primary evaporitic carbonate in the ISB that could have been deposited from alkaline brines in a shallow and partially isolated basin on the proto-arc. Therefore, an early Archean proto-arc could have provided marine environments that were both boron-rich and alkaline.

Hadean zircon crystals have been found from Jack Hills in Australia. These pre–4 Ga zircons have higher oxygen isotope ratios than zircons in oceanic crusts, they have similar U/Yb ratios to Phanerozoic and Archean continental and island arc crusts, and they have higher Li concentrations than normal magmatic ocean-crust zircon. Such geochemical characteristics of Hadean zircon indicate that they formed from a granitic parent magma in TTG-like proto-continental crust (Valley et al. 2010). The Hadean zircons do not necessarily show that a proto-arc was present during the Hadean, but a proto-arc is nonetheless a plausible candidate to produce Hadean juvenile crust.

We suggest that environments created by a Hadean proto-arc were ideal not only for TTG genesis but also for ­prebiotic ribose formation. In isolated and shallow basins on the proto-arc, evaporation may have helped to concentrate borate, supplemented by any continued influx of borate from hydrothermal fluids (Fig. 4). Formaldehyde, the source material of the formose reaction, is a water-soluble volatile and is formed by photochemical reactions between CO2 and H2O (Fig. 1). Formaldehyde in seawater in a shallow basin can be concentrated spontaneously through azeotropic evaporation (up to ~4 mol/L, theoretically). The water in this postulated isolated and shallow basin would be alkaline, as has been inferred for the ISB shallow basin, although the first open oceans were most likely acidic (Russell 2007). Therefore, such isolated alkaline conditions favoured the formation of sugars by the formose reaction. Concentrated borate in such an alkaline basin might have helped to form and preserve ribose selectively, as well as to form oligonucleotides through phosphorylation.

Boron-rich and alkaline environments are also expected locally around mud volcanoes on the slopes of a proto-arc (shown in yellow in Fig. 4), similar to the model proposed by Holm (2012). The presence of alkaline fluids at mud volcanos during the early Archean is indicated by variable zinc isotope compositions from ISB serpentines (Pons et al. 2011). Furthermore, deep marine sediments in the proto-arc would have provided a boron-rich alkaline environment (Mishima et al. 2016). Thus, the formose reaction could also have proceeded in relatively deep marine environments around a Hadean proto-arc. As the result, the spontaneous formation and accumulation of ribose could have been promoted in Hadean proto-arc environments. Preferential stabilization of ribose by borate in prebiotic Earth might explain why ribose became the sugar component in nucleic acids.


The authors thank Steven A. Benner and Hyo-Joong Kim for fruitful discussions on prebiotic chemistry. We also thank Edward Grew, Nils Holm, and Romulus Scorei for helpful comments on this manuscript. The authors acknowledge funding by JSPS KAKENHI (24244084 and 15H02144).


Appel PWU (1995) Tourmalinites in the 3800 Ma old Isua supracrustal belt, West Greenland. Precambrian Research 72: 227-234

Boschi C, Dini A, Früh-Green GL, Kelley DS (2008) Isotopic and element exchange during serpentinization and metasomatism at the Atlantis Massif (MAR 30°N): Insights from B and Sr isotope data. Geochimica et Cosmochimica Acta 72: 1801-1823

Chapelle S, Verchere J-F (1988) A 11B and 13C NMR determination of the structures of borate complexes of pentoses and related sugars. Tetrahedron 44: 4469-4482

Crick FHC (1968) Origin of genetic code. Journal of Molecular Biology 38: 367-379

Ferris JP, Hill AR Jr, Liu R, Orgel LE (1996) Synthesis of long prebiotic oligomers on mineral surfaces. Nature 381: 59-61

Furnes H, de Wit M, Staudigel H, Rosing M, Muehlenbachs K (2007) A vestige of Earth’s oldest ophiolite. Science 315: 1704-1707

Furukawa Y, Horiuchi M, Kakegawa T (2013) Selective stabilization of ribose by borate. Origins of Life and Evolution of Biospheres 43: 353-361

Furukawa Y, Kim HJ, Hutter D, Benner SA (2015) Abiotic regioselective phosphorylation of adenosine with borate in formamide. Astrobiology 15: 259-267

Gasda PJ and 9 coauthors (2016) First observations of boron on Mars and implications for Gale Crater geochemistry. American Geophysical Union 2016 Fall Meeting, San Francisco, 11–15 December, Abstract P21D-04

Grew ES, Bada JL, Hazen RM (2011) Borate minerals and origin of the RNA world. Origins of Life and Evolution of Biospheres 41: 307-316

Grew ES and 6 coauthors (2015) Boron isotopes in tourmaline from the ca. 3.7–3.8 Ga Isua supracrustal belt, Greenland: sources for boron in Eoarchean continental crust and seawater. Geochimica et Cosmochimica Acta 163: 156-177

Holm NG (2012) The significance of Mg in prebiotic geochemistry. Geobiology 10: 269-279
Joyce GF (1989) RNA evolution and the origins of life. Nature 338: 217-224

Kim H-J and 5 coauthors (2016) Evaporite borate-containing mineral ensembles make phosphate available and regiospecifically phosphorylate ribonucleosides: borate as a problem solver in prebiotic chemistry. Angewandte Chemie International Edition 55: 15816-15820

Kirshvink JL, Weiss BP, Beukes NJ (2006) Boron, ribose, and a Martian origin for terrestrial life. Geochimica et Cosmochimica Acta 70(18) Supplement: A320

Lambert JB, Gurusamy-Thangavelu SA, Ma K (2010) The silicate-mediated formose reaction: bottom-up synthesis of sugar silicates. Science 327: 984-986

Larralde R, Robertson MP, Miller SL (1995) Rates of decomposition of ribose and other sugars: implications for chemical evolution. Proceedings of the National Academy of Sciences of the United States of America 92: 8158-8160

Mishima S, Ohtomo Y, Kakegawa T (2016) Occurrence of tourmaline in metasedimentary rocks of the Isua Supracrustal Belt, Greenland: implications for ribose stabilization in Hadean marine sediments. Origins of Life and Evolution of Biospheres 46: 247-271

Nitta S, Furukawa Y, Kakegawa T (2016) Effects of silicate, phosphate, and calcium on the stability of aldopentoses. Origins of Life and Evolution of Biospheres 46: 189-202

Nutman AP, Bennett VC, Friend CR (2015) The emergence of the Eoarchean proto-arc: evolution of ca. 3700 Ma convergent plate boundary at Isua, southern West Greenland. In: Roberts NMW, Van Kranendonk M, Parman S, Shirey S, Clift PD (eds) Continent Formation Through Time. Geological Society, London, Special Publications 389, pp 113-133

Nutman AP, Bennett VC, Friend CRL, Van Kranendonk MJ, Chivas AR (2016) Rapid emergence of life shown by discovery of 3,700-million-year-old microbial structures. Nature 537: 535-538

Pons M-L and 7 coauthors (2011) Early Archean serpentine mud volcanoes at Isua, Greenland, as a niche for early life. Proceedings of the National Academy of Sciences of the United States of America 108: 17639-17643

Ricardo A, Carrigan MA, Olcott AN, Benner SA (2004) Borate minerals stabilize ribose. Science 303: 196

Russell MJ (2007) The alkaline solution to the emergence of life: energy, entropy and early evolution. Acta Biotheoretica 55: 133-179

Scorei R, Cimpoiaşu VM (2006) Boron enhances the thermostability of carbohydrates. Origins of Life and Evolution of the Biosphere 36: 1-11

Shapiro R (1986) Origins: A Skeptics Guide to the Creation of Life on Earth. Summit Books, New York, and Heinemann, London, 336 pp

Shapiro R (1988) Prebiotic ribose synthesis: a critical analysis. Origins of Life and Evolution of Biospheres 18: 71-85

Valley JW and 6 coauthors (2010) What can we agree on before 4 Ga? In: Tyler IM, Knox-Robinson CM (eds) Fifth International Archean Symposium Abstracts. Geological Survey of Western Australia, pp 5-7

Van Duin M, Peters JA, Kieboom APG, Van Bekkum H (1984) Studies on borate esters 1: the pH dependence of the stability of esteres of boric acid and borate in aqueous medium as studied by 11B NMR. Tetrahedron 40: 2901-2911

Williams LB, Hervig RL, Holloway JR, Hutcheon I (2001) Boron isotope geochemistry during diagenesis. Part I. Experimental determination of fractionation during illitization of smectite. Geochimica et Cosmochimica Acta 65: 1769-1782

Share This