ASTROBIOLOGYVolume 9, Number 2, 2009© Mary Ann Liebert, Inc.DOI: 10.1089/ast.2007.0143

Research Article

Role of Metal Oxides in Chemical Evolution: Interaction of Ribose Nucleotides with Alumina

Avnish Kumar Arora and Kamaluddin

Abstract

Abstract

Interaction of ribonucleotides—namely, 5�-AMP, 5�-GMP, 5�-CMP, and 5�-UMP—with acidic, neutral, and ba-sic alumina has been studied. Purine nucleotides showed higher adsorption on alumina in comparison withpyrimidine nucleotides under acidic conditions. Adsorption data obtained followed Langmuir adsorptionisotherm, and Xm and KL values were calculated. On the basis of infrared spectral studies of ribonucleotides,alumina, and ribonucleotide-alumina adducts, we propose that the nitrogen base and phosphate moiety of theribonucleotides interact with the positive charge surface of alumina. Results of the present study may indicatethe importance of alumina in concentrating organic molecules from dilute aqueous solutions in primeval seasin the course of chemical evolution on Earth. Key Words: Chemical evolution—Origin of life—Metal oxides—Alumina—Ribonucleotides. Astrobiology 9, 165–171.

165

1. Introduction

IT HAS BEEN PROPOSED that a catalyst played a central role inthe formation of biopolymers during the course of chem-

ical evolution and origin of life. Catalysts tend to direct thereaction along a few reaction pathways, so that only a lim-ited array of products is obtained. Catalysts bind specifictypes of compounds to their surfaces and then convert themto a limited number of products. The involvement of claysand other inorganic minerals in adsorption and catalytic re-actions of biomolecules in molecular evolution was first pro-posed by Bernal (Bernal, 1951). Ferris studied the role of claysin catalyzing reactions of nucleotides (Ferris and Hagan,1986; Ferris et al., 1989, 1990; Ferris and Ertem, 1992; Ertemand Ferris, 1997). The availability and relevance of solublemineral salts in a primordial environment was proposed byLahav and Chang (Lahav and Chang, 1976, 1982) and Lahavand White (Lahav and White, 1980).

Aluminum is an important component of clay on Earth,where it is coordinated with oxygen and exists in octahedralform. The existence of alumina on Mars has also been re-ported (Keener, 1997). Alumina is found as a mineral and

has a rhombohedral structure similar to Cr2O3, �-Fe2O3, andTi2O3. It is likely that alumina was present on primitive Earthand catalyzed various reactions. Prebiotic synthesis ofpurine, adenine, cytosine, and 4(3H)-pyrimidinone fromformamide has been found to be catalyzed by alumina, aswell as other oxides and minerals, and its application in theorigin of life has been discussed (Saladino et al., 2001). Un-der mild conditions, alumina is found to catalyze peptideformation (Basiuk and Sainz-Rojas, 2001). Meng et al. (2004)examined adsorption and thermal condensation of glycineon silica, and Tzvetkov et al. (2004) studied the interactionbetween glycine and alumina.

We propose that alumina and other metal oxides on prim-itive Earth may have provided surfaces onto which bio-monomers could have concentrated through selective ad-sorption. In this work, we have studied the interaction of theribonucleotides 5�-AMP, 5�-GMP, 5�-CMP and 5�-UMP withaluminum oxide, which is an important subject in view ofthe expected low concentrations of these compounds inaqueous solutions and the associated problem of their poly-merization during the course of chemical evolution. Ribosenucleotides have negatively charged phosphate groups, lone

Department of Chemistry, Indian Institute of Technology Roorkee, Roorkee, India.

pair that contain nitrogen atoms, and aromatic rings with �electron clouds. So interaction might have taken place withthe negatively charged moieties of nucleotide and the posi-tively charged surface of alumina. The present work showsaluminum oxide to be a good adsorbent for ribonucleotidesand, therefore, supports the possible role of metal oxides inchemical evolution.

2. Experimental

Materials

Aluminum oxide acidic [70–290 mesh size, pH of aqueoussuspension 4.5, Sisco Research Laboratories (SRL), Mumbai,India], neutral (70–290 mesh size, pH of aqueous suspension6.40, SRL), basic (70–290 mesh size, pH of aqueous suspen-sion 8.20, SRL), 5�-AMP (SRL), 5�-CMP (SRL), 5�-GMP (SRL),and 5�-UMP (SRL) were used as received. All other chemi-cals were of analytical reagent grade.

2.1. Spectral studies

The electronic spectra of 5�-AMP, 5�-GMP, 5�-CMP, and5�-UMP were recorded with a Shimadzu UV-1601 spec-trophotometer, and their characteristic �max values were 259,254, 278, 262 nm, respectively. The infrared spectra of ad-sorbents, adsorbates, and adsorption adducts were recordedin KBr discs on a Perkin Elmer Fourier transform infraredspectrophotometer (Model Perkin Elmer-1600 Series).

2.2. CHN analysis

Carbon, hydrogen, nitrogen analysis of the adduct wascarried out on an Elementar Vario ELHI CHNS analyzer.This showed presence of C, H, and N on the surface of alu-minum oxide.

2.3. BET surface area

The BET surface area of the samples was measured by ni-trogen adsorption isotherms on micromeritics ASAP 2010(UK). Samples were activated at 473 K for 4 h prior to themeasurement.

2.4. Adsorption studies

Adsorption of nucleotides on aluminum oxide was studiedwith use of varying concentrations of nucleotides (400–640�M). Twenty-five milligrams of oxide were added to 5 ml ofdifferent concentrations of aqueous solution of nucleotide.Mixtures were shaken for about 2 minutes. The mixture after24 h was centrifuged on a REMI R4C centrifuge for 30 min-utes at 4500 rpm. Absorbance of the supernatant was noted atcharacteristic �max of the nucleotide. The concentration of nu-cleotides after adsorption for 24 h was determined with thehelp of a calibration curve made by plotting concentration vs.optical density (O.D.) of nucleotides. The amount of nucleo-tide adsorbed was calculated with the absorbance of nucleo-tide before and after adsorption. Nucleotide-alumina adductwas washed with water, dried, and subjected to IR studies.

3. Results

Adsorption of nucleotides on aluminum oxide was stud-ied over a range of concentrations of adsorbate (400–640 �M)

on acidic, neutral, and basic alumina. All three forms of alu-mina are equivalent with respect to nucleotide adsorptionand were verified by putting any one of them in the condi-tions of desired pH. Adsorption isotherms were obtained byplotting the amount of nucleotides adsorbed, Xe (mg/g), ver-sus their equilibrium concentration, Ce (mol L�1). Plots weremade with Origin 6.0 software and were extrapolated towardthe origin of the graph with zero absorbance against zeroconcentration of the nucleotides. At extremely low concen-trations of the nucleotide, estimation of absorbance was verydifficult. From the adsorption isotherm, percent adsorptionof nucleotides on the metal oxides was determined, whichis reported in Table 1. Percent binding was calculated fromabsorbance values of nucleotides before and after adsorptionat the point of the curve where saturation starts. The initialportions of the isotherms show a linear relationship betweenthe amount adsorbed in mg/g and the equilibrium concen-tration. At a higher concentration range, saturation phe-nomenon was observed (Figs. 1–4). The adsorption data werefitted in Langmuir adsorption isotherm (Atkins and dePaula, 2002) as given below.

� � kL

where Ce is equilibrium concentration of nucleotide(mole/liter), Xe is amount of nucleotide adsorbed per gramweight of adsorbent (mg), Xm is amount of nucleotide ad-sorbed at saturation (mg/g), and kL is the Langmuir ad-sorption constant (L/mg).

A typical graph of Ce/Xe vs. Ce was a straight line [Figs.5–8]. From the curve Xm and kL, values were calculated andare presented in Table 1. Higher Xm values indicate moreamount of nucleotide adsorbed, while KL value is related tothe enthalpy of adsorption.

Adsorption of nucleotides on all three forms of aluminawas found to be maximum in acidic pH (4.0) as comparedto the neutral and basic forms. All three forms of aluminaare equivalent and could be obtained by putting any one ofthem in the conditions of desired pH. Adsorption of all fournucleotides in acidic (pH �6.0), neutral (pH �7.0), and ba-sic (pH �8.0) alumina was studied, and the results are shownin Table 2. Higher adsorption of the nucleotide is in accor-dance with the surface area of alumina in their three forms.

1�Xm

Ce�Xm

Ce�Xe

ARORA AND KAMALUDDIN166

TABLE 1. LANGMUIR CONSTANTS FOR ADSORPTION OF

NUCLEOTIDES ON ALUMINUM OXIDES

Alumina Nucleotide kL (L mg�1) Xm (mg/g)

Acidic 5�-AMP 9.86 60.605�-GMP 9.53 62.895�-CMP 6.61 38.315�-UMP 7.11 32.26

Neutral 5�-AMP 9.87 35.335�-GMP 9.31 41.845�-CMP 5.63 34.485�-UMP 6.76 29.23

Basic 5�-AMP 8.21 28.575�-GMP 9.24 30.765�-CMP 6.74 28.085�-UMP 7.10 25.83

RIBONUCLEOTIDE-ALUMINA INTERACTION 167

FIG. 1. Adsorption isotherms for 5�-AMP on alumina.FIG. 2. Adsorption isotherms for 5�-GMP on alumina.

FIG. 3. Adsorption isotherms for 5�-CMP on alumina. FIG. 4. Adsorption isotherms for 5�-UMP on alumina.

FIG. 5. Langmuir isotherm plots for 5�-AMP on alumina. FIG. 6. Langmuir isotherm plots for 5�-GMP on alumina.

Surface area of acidic, neutral, and basic alumina was foundto be 146, 127, and 119 m2/g.

To show that the behavior of the three forms of aluminais the same and pH of the suspension is responsible for thediffering behavior of acidic, neutral, and basic forms, a typ-ical experiment was carried out where adsorption of 5�-AMPon all three alumina under acidic, neutral, and basic condi-tions was studied. It was found that adsorption of 5�-AMPwas maximum in acidic, less in neutral, and the least in ba-sic medium (Table 3). Percent adsorption values of 5�-AMPon the three alumina under acidic, neutral, and basic condi-tions were nearly equal. In general, alumina was found tohave an extremely high affinity toward ribonucleotides asthe extent of surface covered (� � kLCe/1 � kLCe) was de-termined to be about 100% irrespective of the pH of themedium.

4. Discussion

In an aluminum oxide/water system, the OH groups onthe solid surface are the most important sites for surface in-

ARORA AND KAMALUDDIN168

M

FIG. 8. Langmuir isotherm plots for 5�-UMP on alumina.

TABLE 2. PERCENT BINDING OF NUCLEOTIDES ON ALUMINUM OXIDE

Alumina Nucleotide Percent binding

Acidic 5�-AMP 715�-GMP 725�-CMP 685�-UMP 59

Neutral 5�-AMP 605�-GMP 605�-CMP 545�-UMP 55

Basic 5�-AMP 475�-GMP 475�-CMP 415�-UMP 37

% Binding � � � � 100O.D. of ribonucleotide before adsorption � O.D. of ribonucleotide after adsorption����������

O.D. of ribonucleotide before adsorption

TABLE 3. PERCENT ADSORPTION OF 5�-AMP ON ALUMINA

Percent adsorption

Alumina pH 4.0 pH � 7 pH 8.0 pH 11.0

Acidic 72.0 63.0 47.0 4.0Neutral 71.0 62.0 47.0 3.0Basic 71.0 62.0 47.0 3.0

FIG. 7. Langmuir isotherm plots for 5�-CMP on alumina.

teractions. These groups can act as acid or base, dependingon the pH of the solution. Zero point charge (ZPC) for alu-mina is 8.9 (Huang et al., 1996). In acidic conditions, positivesurface charge on alumina is mainly due to the protonatedaluminol group (AlOH�

2), whereas in alkaline medium be-yond pH 8.9, the surface of alumina becomes negativelycharged due to the formation of AlO� (Forland et al., 1996).Below ZPC (8.9), interaction with nucleotides may take placethrough negatively charged oxygen atoms in the phosphategroup, � electrons of the aromatic ring, and lone pair of elec-

RIBONUCLEOTIDE-ALUMINA INTERACTION 169

FIG. 9. IR spectra of 5�-AMP (a), Alumina (b), and 5�-AMP-alumina adduct (c). Adsorption performed at pH 4.0.

trons on N and O atoms with the positively charged surfaceof alumina. Purines (5�-AMP, 5�-GMP), which have one morering and lone pair of electrons (N-7), could adsorb morestrongly on alumina compared to pyrimidines (5�-CMP and5�-UMP).

Under different experimental conditions of pH, adsorp-tion of nucleotides on the alumina surface may occurthrough different types of interactions. Negatively chargedoxygen of the nucleotide may interact with the protonatedalumininol groups (AlOH�

2) through electrostatic forces, orit may establish hydrogen bonding with the AlOH groupsof alumina surfaces. At pH above 8.9 beyond ZPC, the sur-face of aluminum oxide becomes negatively charged due tothe formation of AlO� groups on its surface. There will berepulsion between AlO� and the negatively charged oxygenatoms of nucleotides. So the adsorption is extremely low(1–5%), as shown in Table 3.

The nature of the interaction between ribonucleotides andalumina has been studied by means of infrared absorption.The infrared spectra of adenosine monophosphate, acidicalumina, and alumina-nucleotide adduct are depicted in Fig.9a, 9b, and 9c, respectively. The shift in wavelengths of thecharacteristic frequencies of ribonucleotides in nucleotide-alumina adduct indicates interactions between the ribonu-cleotide and alumina and are summarized in Table 4. Itseems that the binding sites of nucleotides may be N-1 andN-7 for 5�-AMP, N-1 and N-7 for 5�-GMP, and N-3 for 5�-CMP and 5�-UMP. The adsorption is presumably related tothe involvement of N-1, N-3, and N-7 of the base residues,lone pair of nitrogen atom, and to the dissociation of twoavailable protons of the phosphate group. A shift towardslower wavelength of N-H bending from 1647 cm�1 to 1617cm�1 shows the involvement of the NH2 group in an inter-action of ribonucleotide with alumina. A strong band at 1084cm�1 in 5�-AMP corresponding to C-O-P vibration has beenshifted to 1108 cm�1. So interaction takes place throughamino and phosphate groups of ribonucleotides with alu-mina.

The absence of a significant change in the typical infraredfrequencies of the alumina suggests that the ribonucleotidemolecules do not enter into the lattice. Ribonucleotides in-teract through their purine or pyrimidine residue and phos-phate group with the positively charged surface of material.Greater adsorption of 5�-AMP and 5�-GMP may also be dueto a higher number of � electrons in the aromatic ring sys-tem.

Carbon, hydrogen, nitrogen analysis of the AMP-aluminaadduct shows 0.604% N, 0.153% H, and 1.045% C on the sur-face of acidic alumina. This confirms the presence of AMPon alumina because C:N molar ratio corresponds to that ex-pected for 5�-AMP.

5. Conclusion

Alumina may have played a role in selectively adsorbingand concentrating molecules in the primeval soup duringchemical evolution. The results of this study on the adsorp-tion of ribonucleotides on alumina lend support to this hy-pothesis by demonstrating a strong degree of adsorption.This process may have protected these biomolecules fromdegradation and helped to preserve them for further poly-merization reactions.

Abbreviations

O.D., optical density; ZPC, zero point charge.

References

Atkins, P. and de Paula, J. (2002) Atkins’ Physical Chemistry, 7th

ed., Oxford University Press, Oxford.Basiuk, V.A. and Sainz-Rojas, J. (2001) Catalysis of peptide for-

mation by inorganic oxides: high efficiency of alumina undermild conditions on the Earth-like planets. Adv. Space Res.27:225–230.

Bernal, J.D. (1951) The Physical Basis of Life, Routledge and KeganPaul, London.

Ertem, G. and Ferris, J.P. (1997) Template-directed synthesis us-ing the heterogeneous templates produced by montmoril-lonite catalysis. A possible bridge between the prebiotic andRNA worlds. J. Am. Chem. Soc. 119:7197–7201.

Ferris, J.P. amd Ertem, G. (1992) Oligomerization of ribonu-cleotides on montmorrillonite: reactions of the 5�-phospho-rimidazolide of adenosine. Science 257:1387–1389.

Ferris, J.P. and Hagan, W.J. (1986) Adsorption and reaction ofadenine nucleotide on montmorillonite. Orig. Life Evol. Biosph.17:69–84.

Ferris, J.P., Ertem, G., and Agarwal, V.K. (1989) The adsorptionof nucleotides and polynucleotides on montmorillonite clay.Orig. Life Evol. Biosph. 19:153–164.

Ferris, J.P., Kamaluddin, and Ertem, G. (1990) Oligomerizationreactions of deoxyribonucleotides on montmorillonite clay:the effect of mononucleotide structure, phosphate activationand montmorillonite composition on phosphodiester bondformation. Orig. Life Evol. Biosph. 20:279–291.

ARORA AND KAMALUDDIN170

TABLE 4. TYPICAL INFRARED SPECTRAL FREQUENCIES (CM�1) OF RIBONUCLEOTIDES

BEFORE AND AFTER ABSORPTION ON ALUMINUM OXIDE

Ribonucleotides vNH2 vNH vC-O-P

5�-AMP 3221* 3343 1647 1084(3414)* (3482) (1617) (1108)

5�-GMP 3143* 3433 1605 1083(3390)* (3478) (1584) (1099)

5�-CMP 3122* 3334 1539 1091(3275)* (3462) (1530) (1102)

5�-UMP 1473 1082(1468) (1085)

*Values in parentheses indicate typical infrared spectral frequencies after adsorption.

Førland, G.M., Rahman, T., Høiland, H., and Børve, K.J. (1996)Adsorption of sodium dodecyl sulphate and butanol ontoacidic and basic alumina. J. Colloid Interface Sci. 182:348–355.

Huang, L., Maltesh, C., and Somasundaram, P. (1996) Adsorptionbehavior of cationic and nonionic surfactant mixtures at the alumina-water interface. J. Colloid Interface Sci.177:222–228.

Keener, J. (1997) Martian soil compositions (from Pathfinder1997) by weight percentage. Available online at http://www.redcolony.com/features.php?name=compositions.

Lahav, N. and Chang, S. (1976) The possible role of solid sur-face area in condensation reactions during chemical evolution:reevaluation. J. Mol. Evol. 8:357–380.

Lahav, N. and Chang, S. (1982) The possible role of soluble saltsin chemical evolution. J. Mol. Evol. 19:36–46.

Lahav, N. and White, D.H. (1980) A possible role of fluctuatingclay-water systems in the production of ordered prebioticoligomers. J. Mol. Evol. 16:11–21.

Meng, M., Stievano, L., and Lambert, J.-F. (2004) Adsorption andthermal condensation mechanisms of amino acids on oxidesupports. 1. glycine on silica. Langmuir 20:914–923.

Saladino, R., Crestini, C., Costanzo, G., Negri, R., and Mauro,E.D. (2001) A possible prebiotic synthesis of purine, ade-nine, cytosine, and 4(3H)-pyrimidinone from formamide:implications for the origin of life. Bioorganic Med. Chem.9:1249–1253.

Tzvetkov, G., Koller, G., Zubavichus, Y., Fuchs, O., Casu, M.B.,Heske, C., Umbach, E., Grunze, M., Ramsey, M.G., and Net-zer, F.P. (2004) Bonding and structure of glycine on orderedAl2O3 film surfaces. Langmuir 20:10551–10559.

Address reprint requests to:Dr. Kamaluddin

Department of ChemistryIndian Institute of Technology

Roorkee 247667 (U.A)India

E-mail: kamalfcy@iitr.ernet.in

RIBONUCLEOTIDE-ALUMINA INTERACTION 171