Nonenzymatic RNA Oligomerization at the MineralâWater Interface

Subscriber access provided by YORK UNIV

C: Surfaces, Interfaces, Porous Materials, and Catalysis

Non-Enzymatic RNA Oligomerization at the Mineral-Water Interface: An Insight into the Adsorption-Polymerization Relationship Hussein Kaddour, Selim Gerislioglu, Punam Dalai, Toshikazu Miyoshi, Chrys Wesdemiotis, and Nita Sahai J. Phys. Chem. C, Just Accepted Manuscript • DOI: 10.1021/acs.jpcc.8b10288 • Publication Date (Web): 05 Dec 2018 Downloaded from http://pubs.acs.org on December 12, 2018

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 46 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

Non-Enzymatic RNA Oligomerization at the Mineral-Water Interface: An Insight into the Adsorption-Polymerization Relationship

Hussein Kaddour1,2,+ Selim Gerislioglu3, Punam Dalai1, Toshikazu Miyoshi1, Chrys Wesdemiotis1,3, and Nita Sahai1,4,5, *

1 2

Department of Polymer Science, 170 University Ave., University of Akron, Akron, OH 44325

Present address: Department of Pharmacology, Stony Brook University, 101 Nicolls Rd, Stony Brook, NY, 11794–8651, USA. 3 4 5

Department of Chemistry, University of Akron, Akron, OH 44325

Department of Geosciences, University of Akron, Akron, OH 44325

Integrated Bioscience Program, University of Akron, Akron, OH 44325 *

Corresponding author: [email protected]

+

Present address: Department of Pharmacological Sciences, Stony Brook University, Stony Brook, NY 11794, USA

ACS Paragon Plus Environment

1

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 46

ABSTRACT Non-enzymatic RNA polymerization, a major challenge in understanding the origin of life on early Earth, was previously achieved in the presence of montmorillonite clay and high magnesium concentrations (~75 mM) but the molecular interactions promoting ribonucleotide oligomerization remain unknown. High adsorption capacity of minerals is generally assumed to favor polymerization efficiency; however, this relationship was never shown. Here we examined the relationship between ribonucleotide adsorption affinity and the role of magnesium in the polymerization catalytic efficiency of minerals. Results showed that adsorption of the activated mononucleotide,

2-methylimadzolide

of

adenosine

monophosphate

(2-MeImPA),

is

approximately 10 times higher on zincite (ZnO) than on montmorillonite clay, but only montmorillonite acts as a catalyst for 2-MeImPA polymerization. In the presence of montmorillonite, oligomers formed even in pure water without any salts present. Attenuated Total Reflectance FTIR and Cross Polarization-Magic Angle Spinning 31P NMR spectroscopy showed that 2-MeImPA mononucleotides adsorption on ZnO directly involves the phosphate moiety, making it unavailable for polymerization. In contrast, mononucleotides adsorbed on montmorillonite by weak amine interactions leaving the phosphate moiety available for polymerization. Thus, providing a favorable orientation of the monomer, rather than a high adsorption capacity, as well as providing a nanoconfined environment and outer-sphere complex formation with the nucleotide are requisite properties for a mineral to be catalytic to nucleotide polymerization.


2

Page 3 of 46 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


1. INTRODUCTION One of the major challenges in the origins of life field is the non-enzymatic synthesis of RNA oligomers long enough to bear functionality.1-2 In the absence of enzymes, minerals have long been proposed to have played the role of catalysts.3 Pioneering work of James Ferris and coworkers at Rensselaer Polytechnic Institute for the past thirty years has demonstrated that montmorillonite clay promotes the non-enzymatic polymerization of activated RNA mononucleotides in the presence of magnesium or alkali cations.4-19 Early work from the Ferris group also showed only short oligomers up to about 10- or 12-mers.7, 20 Long oligomers up to approximately 40-mers were subsequently claimed, based on gel electrophoresis results without mass spectrometry for confirmation10, but those results have not been reproducible and recently only short oligomers up to 9-mers were shown unambiguously.18, 21 Furthermore, the detailed molecular-level interactions at the mineral-organic interface responsible for the catalytic effect of montmorillonite on activated ribonucleotide polymerization remain unknown to date. Based on the results obtained for montmorillonite, it has been repeatedly hypothesized in the literature that the catalytic activity of minerals relies on their capacity to adsorb prebiotic organics22, with much attention given to the amount of adsorption or adsorption capacity of a mineral.9, 23-24 Adsorption is hypothesized to lead to the emergence of “surface organisms” or “surface metabolists”.25 While high adsorption capacity may be important for other reasons, such as protection of the RNA from degradation26 the assumed correlation of adsorption capacity to polymerization efficiency remains untested. Thus, most studies simply examine RNA monomer (ribonucleotide or ribonucleoside) adsorption but not the polymerization reaction.27-38 Even these studies focus only on the simple mononucleotide whereas clay-catalyzed polymerization requires the activated mononucleotide.


3


Page 4 of 46

Conversely, other studies investigated the potential catalytic activity of various minerals (magnesite, calcite, dolomite, olivine, brucite, talc, montmorillonite, goethite, hematite, magnetite, siderite, galena, sphalerite and chalcocite) to promote non-enzymatic RNA polymerization and found only montmorillonite to be catalytic13, 20, but these studies did not include a detailed investigation of the adsorption characteristics of the various minerals. Thus, a relationship between the adsorption capacity and the catalytic activity of minerals in promoting ribonucleotide polymerization has yet to be determined. From a surface chemistry standpoint, the apparently unique catalytic activity of montmorillonite is intriguing because montmorillonite and the nucleotides are both negatively-charged at the optimal pH of polymerization (pH 7-8), hence, adsorption should be limited. This is probably the need for cations (Na+, Mg2+, Ca2+) in the polymerization reaction in order to facilitate the coordination of the nucleotides to the mineral surface.9 Furthermore, not even all montmorillonites are excellent catalysts–the efficiency depends on the provenance of the clay, which reflects the extent of isomorphous chemical substitution in the crystal structure during the formation of the clay, and on its treatment prior to the polymerization experiment by the so-called “Banin procedure”.16, 39 These factors apparently affect the cation-exchange capacity and acid-base properties of the clay. Finally, most of the reported non-enzymatic RNA polymerization reactions on montmorillonite are slow (days) and the yields are low (< 1%)1, urging therefore the search for better catalysts from the plausible prebiotic mineral inventory.18, 21 An understanding of the montmorillonite-catalyzed polymerization mechanism would help to identify other minerals with potential catalytic activity. In an attempt to identify the polymerization mechanism, the pH dependence of nucleotide adsorption and subsequent polymerization on three different montmorillonites was studied by


4

Page 5 of 46 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Joshi and co-workers.17 The authors investigated the role of each functional group of both the mineral and the nucleotide in the mechanism of adsorption. With these results, together with XRD data, it was proposed that polymerization is acid-base catalyzed and a mechanism was proposed wherein a hydroxyl edge site of montmorillonite triggers the nucleophilic attack. This mechanism, however, does not address how the nucleotides are arranged at the mineral surface or what would be the active conformation of the monomer to trigger the nucleophilic attack. Moreover, dissolved magnesium at concentrations of up to 50-75 mM is required for nonenzymatic RNA polymerization in the presence of montmorillonite as shown by Ferris and coworkers6,

16

or in the presence of a soluble RNA template.40 This is much higher than the

concentrations expected at the Earth’s surface either prebiotically or on modern Earth. The main goals of the present study were to shed light on any potential relationship between adsorption capacity and polymerization efficiency of various ribonucleotide monomers, and to determine the adsorbed conformation of the monomers required for polymerization. We also investigated the effects of aqueous magnesium concentration on polymerization. In detail, we determined adsorption of the 2-methylimidazolide of 5’-adenosine monophosphate (2-MeImPA), as well as the polymerization of 2-MeImPA and of 3’,5’-cyclic adenosine monophosphate (3’,5cAMP) (Figure S1). Cyclic nucleotide monomers generally accepted to be prebiotically plausible41-42 and 3’5’-cyclic guanosine monophosphate (cGMP) monomers are have been found recently to polymerize non-enzymatically in water43 and under dry conditions at moderate temperatures (50 – 80 °C).44 Adsorption and polymerization were examined on montmorillonite, which is known to promote polymerization and on zincite, which shows strong affinity for 5’-AMP based on preliminary results in our laboratory. The conformation of the adsorbed molecule is expected to be


5


Page 6 of 46

a key determinant of polymerization efficiency because the phosphate moiety should be available for polymerization. Conformation of adsorbed monomer was determined by Fourier transform infra-red (FTIR) and solid state nuclear magnetic resonance (ss-NMR) spectroscopy. These spectroscopic methods have been used previously for probing the structure of mononucleobases and mononucleosides35, 45 but, rarely, that of adsorbed mononucleotides.30, 46-48

2. MATERIALS AND METHODS 2.1. Materials. All chemical reagents used in the synthesis of activated nucleotides were purchased either from Sigma-Aldrich (Saint Louis, MO, USA) or from Fisher Scientific (Waltham, MA, USA), except 3’,5’-cyclic AMP free acid form (Biolog-Life Science Institute, Germany). Zincite (ZnO, 99.999 %, trace metal basis) was purchased from Sigma-Aldrich. Natural montmorillonite (Wyoming, Volclay SPV-200) was purchased from the American Colloid Company (Troy, IN, USA). All water used was ultrapure (18 M -cm) (Nanopore UV, Barnstead, USA). 2.2. Mineral Characterization. Minerals were characterized as described previously in detail.49 The identity of both minerals was confirmed by X-Ray Diffraction (Ultima IV Rigaku, DE, USA). Their specific surface areas were determined by multipoint N2 gas adsorption fit to BET isotherm (Micromeritics, GA, USA). Particle sizes and crystal morphology were estimated by Transmission Electron Microscopy (TEM, JSM-1230, ~ 120 kV, JEOL, MA, USA). Two µL of 0.1 mg.mL-1 of each mineral suspension were placed on 300-mesh copper grids coated with formvar/carbon film and dried in air. Isoelectric points (IEP) were determined by measuring the zeta potential of mineral suspensions (1 mg.mL-1) as a function of pH in ultrapure water and identifying the point of zero potential (Zetasizer NanoSeries ZS, Malvern Instruments, UK).49


6

Page 7 of 46 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Catalytic montmorillonite was prepared according to the Banin procedure39 as described by Joshi et al.16, which involves titrating the H+-form of the clay with 0.1 M NaOH to pH 8. 2.3. Synthesis of 2-MeImPA. The 2-methylimidazolide of 5’-adenosine monophosphate (5’-AMP) was prepared following the original published procedure50 with slight modifications adapted from Joshi et al.15 In detail, 5′-AMP monohydrate (611 mg) and 2-methylimidazole (1.8 g) were dissolved in 10 mL of anhydrous dimethylformamide (DMF) and the solvent was evaporated to dryness in a rotary evaporator (Heidolph, HeiVap, Germany). The evaporation was repeated twice with anhydrous DMF to remove residual water and the flask was kept overnight in a vacuum desiccator. The residue was again dissolved in 10 mL anhydrous DMF and stirred with triphenylphosphine (1.2 g), 2,2′-dithiodipyridine (1.4 g) and triethylamine (0.9 mL) at room temperature for 16 hours. The incubation duration was increased from 4h to 16 h to maximize the yields. The resulting product was precipitated in a solution of anhydrous diethyl ether (300 mL), acetone (99.5% pure, 200 mL) and sodium perchlorate (3.5 g), washed with 1:1 acetone/ether solution, then with ether alone until the yellowish color disappeared, and then ether was finally dried in a vacuum desiccator overnight. The formation of 2-MeImPA was confirmed by Electron Spray Ionization Mass Spectrometry analysis (ESI-MS) and its purity (~97 ± 3 %) was determined by reverse phase HPLC on a Kinetex biphenyl column (Phenomenex) (Figure S2). The activated nucleotide was used without purification. 2.4. Adsorption Isotherms. Adsorption of 2-MeImPA on minerals was measured through batch experiment with 10 mg.mL-1 mineral particle loading, and a nucleotide concentration ranging from 0-4 mM in ultrapure water. Reaction mixtures were prepared in 1.5 mL eppendorf tubes from stock solutions of 10 mM nucleotide and stock suspensions of 20 mg.mL-1 of mineral, prepared in water. Accurate pipetting was assured using electronic multichannel micropipettes (E1


7


Page 8 of 46

Cliptip, Thermo Scientific) to achieve maximum reproducibility of results. Samples were briefly vortexed and incubated at room temperature for 24-48 hours with occasional vortexing. Samples were then centrifuged at 13,300 rpm (Accuspin Micro 17, Fisher ScientificTM, Waltham, MA, USA) for 25 min at 25 °C and the supernatants were collected. The UV-Vis absorbance of the supernatant solutions was measured using a NanoDrop 2000c spectrophotometer (Thermo Fisher Scientific, Waltham, MA, USA). Error bars represent standard deviation of triplicate samples. The experiment was repeated three times in triplicate. 2.5. Polymerization Reactions. For polymerization of 2-MeImPA, the procedure was adapted from Joshi et al.15 In 1.5 mL eppendorf tubes, 15 µL of 100 mM 2-MeImPA stock solution (freshly prepared) were mixed with 10 µL of 10X polymerization buffer (2 M NaCl, 0.75 M MgCl2) and diluted with water to reach final concentrations of 200 mM NaCl, 75 mM MgCl2 (polymerization buffer) and 15 mM 2-MeImPA, in a reaction volume of 100 µL. To this reaction, 5 ± 0.3 mg mineral was added, yielding a mineral loading of 50 mg.mL-1. The mixture was vortexed briefly and kept at room temperature for 3 days. It is worthwhile to mention that experiments where pH was adjusted to 8 by the presence of 0.2 M buffer (HEPES or bicine) were conducted, but only a monomer-buffer covalently-bonded complex was obtained and apparently this complex inhibited the formation of oligomers, as evidenced by the results of ESI-MS analysis (Figure S3). Samples without minerals and/or without any salts were similarly prepared. RNA monomer and oligonucleotides were extracted prior to their identification as follows. At the end of the above-described polymerization reaction, the samples were centrifuged at 4000 rpm (Accuspin Micro 17, Fisher Scientific, Waltham, MA, USA) for 5 min at 25 °C and the supernatants were collected. To the mineral pellet, 200 µL extraction solution (0.1 M EDTA pH 11) were added and the samples were thoroughly vortexed for at least 1 hour. The extraction was


8

Page 9 of 46 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


repeated twice, the supernatants from each extraction cycle were combined and the volume was adjusted to 1 mL with water. The pH was adjusted to 4 with ~20 µL of 1 M perchloric acid and the sample was incubated at 37 °C for 1 hour to hydrolyze the un-reacted 2-MeImPA. Samples were finally filtered through cellulose acetate centrifuge tube filters (SpinX, Costar, Corning, NY, USA) with a pore size of 0.22 µm. 3’,5’-cyclic AMP H+-form was polymerized in dry conditions following the method of Morasch et al.44 Briefly, 15 mM of 3’,5’-cyclic AMP were incubated in the presence and absence of minerals (50 mg.mL-1) and in the presence and absence of polymerization buffer (200 mM NaCl, 75 mM MgCl2), in a 200 µL reaction volume and the mixture was incubated under vacuum (Eppendorf, Vacufuge Plus) at 60 °C for 16 hours. RNA was extracted from the minerals with 0.1 M EDTA pH 11 (200 µL, 1 hour, 3 times) and the extracts were filtered. 2.6. HPLC Analysis. HPLC analysis was performed using a Shimadzu Nexera 2 system equipped with a Photo Diode Array UV/Vis detector. The polymerization products were separated on a Dionex DNAPac-200 RS, 4 µm (4.6 × 250 mm2) analytical anion exchange column, using a gradient of 0–0.2 M NaClO4 with 2 mM Tris at pH 8, a method adapted from Joshi et al.15. All samples were filtered through 0.22 µm cellulose acetate filters (SpinX, Costar) prior to analysis. Analysis of the 2-MeImPA was performed on a Kinetex Biphenyl 2.6 µm, 100 Å column, 100 x 4.6 mm (Phenomenex) using a mobile phase of H2O/MeCN (98:2) acidified with 0.1 % TFA at a flow rate of 1 mL.min-1. Both separations were carried out at 40 °C. 2.7. Mass Spectrometry. Matrix assisted laser desorption/ionization mass spectrometry (MALDI-MS) experiments were performed on a Bruker Ultraflex III MALDI tandem time-offlight (TOF/TOF) mass spectrometer (Bruker Daltonics, Billerica, MA, USA) equipped with a Nd:YAG laser emitting at 355 nm. All samples were analyzed before and after desalting using a


9


Page 10 of 46

“ziptip.” The samples were desalted using C18-Millipore Ziptips (Sigma-Aldrich, St Louis, MO) according to the manufacturer’s guidelines. 6-aza-2-thiothimidine (ATT) (99 %; Sigma-Aldrich, St. Louis, MO, USA) served as matrix. The matrix solution was prepared in H2O/MeCN (50:50, v/v) at 2 mg·mL-1 concentration. The matrix and sample solutions were mixed in the ratio 5:1 (v/v), and 0.5–1.0 µL of the final mixture were applied to the MALDI target plate and allowed to dry at ambient room conditions before spectral acquisition at negative ion mode. This sample preparation protocol with or without desalting led to the formation of [M - H]- or [M + mNa – nH](n – m = 1) ions, respectively, where M corresponds to the mass of the compound of interest and –H indicates that the compound was ionized by deprotonation to form an anion. Spectral acquisition was carried out at reflectron mode and ion source 1 (IS 1), ion source 2 (IS 2), source lens, reflectron 1, and reflectron 2 potentials were set at 25.03 kV, 21.72 kV, 9.65 kV, 26.32, and 13.73 kV, respectively. The MALDI-MS data were analyzed using Bruker’s flexAnalysis v3.3 software. Electrospray ionization MS analyses were carried out by Bruker HCT Ultra II quadrupole ion trap mass spectrometer (Bruker Daltonics, Billerica, MA) equipped with ESI source. Samples were diluted to 1 µg·mL-1 in H2O/MeCN (50:50, v/v). The sample solutions were injected into the ESI source by direct infusion, using a syringe pump, at a flow rate of 3 µL·min-1. The tip of the ESI needle was grounded, and the entrance of the capillary, through which ions enter in the vacuum system of the mass spectrometer, was held at 3.5 kV. The pressure of the nebulizing gas (nitrogen) was set at 10 psi, and the flow rate and temperature of the drying gas (nitrogen) were 8 L·min-1 and 300 °C, respectively. Data collection was performed on negative ion mode. The ESI-MS data were analyzed using Bruker’s DataAnalysis v4.0 software.


10

Page 11 of 46 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


2.8. ATR-FTIR Spectroscopy. Four mM of 2-MeImPA or 5’-AMP-Na in 15 mL pure water was mixed with 150 mg of zincite. Because of weak adsorption, 10 mM of 2-MeImPA in 15 mL pure water was mixed with 150 mg of Banin-treated montmorillonite. All suspensions were incubated on a rotary shaker (SB3 Rotator, Stuart, Staffordshire, UK) overnight. Samples were then centrifuged at 10,000 rpm for 15 min and supernatants were discarded. The pellets obtained were frozen with liquid nitrogen and dried using lyophilizer (FreeZone 2.5, Labconco, Kansas City, MO, USA) for 24 hours. Measurements were performed with a Thermo Nicolet 380 FTIR instrument (Thermo Fisher Scientific, Waltham, MA, USA). Each spectrum was recorded in ATRmode with the use of an ATR Smart Orbit between 500 and 4000 cm-1 range with 64 scans with a resolution of 4 cm-1 at room temperature. 2.9. NMR Spectroscopy. Solid-State Magic Angle Spinning (MAS) NMR experiments were performed on a Brucker Avance 300 instrument with resonance frequencies of 121.5 and 300.1 MHz for

31

P and 1H, respectively. A 7 mm double-resonance probe was used for Cross-

Polarization Magic Angle Spinning (CPMAS) experiment at a MAS frequency of 4 kHz. The radio frequency field strength was set to 62.5 kHz for both 1H and 31P channels. The 1H–31P CP contact time was set to 2 ms. 1H two-phase pulse modulation (TPPM) with a field strength of 62.5 kHz was applied during the 31P signal acquisition time of 10 ms. Accumulation numbers of signals (NS) and recycle delay (RD) were varied depending on the sample. For 2-MeImPA, the CP-MAS conditions were: 2-MeImPA standard, NS = 512, RD = 14s; zincite, NS = 4,000, RD = 3s; montmorillonite, NS = 500,000, RD = 500 ms; The CP static conditions were: 2-MeImPA standard, NS = 2048; zincite = 16,000, RD = 3s. Analysis was performed on samples prepared similarly as for FTIR. Briefly, 4 mM of 2-MeImPA or 5’-AMP-Na in 15 mL pure water was mixed with 150 mg of zincite or Banin-treated montmorillonite and the suspensions allowed to equilibrate


11


Page 12 of 46

overnight with end-over-end rotation. Samples were then centrifuged at 10,000 rpm for 15 min and the pellets obtained were freeze dried using lyophilizer for 24 hours.


12

Page 13 of 46

3. RESULTS AND DISCUSSION 3.1. Adsorption Isotherms. The adsorption of 2-MeImPA mononucleotide on zincite and montmorillonite in water was examined as a function of increasing monomer concentration at room temperature (Figure 1). Adsorption on zincite was greater than on montmorillonite at all equilibrium concentrations (Ceq) of the monomer up to a factor of 10X greater at Ceq = 2 mM. No adsorption maximum was obtained within the concentration range examined.

Z in c ite

-2

)

80

q a d s o r b e d ( µ m o l.m q ads (µmol.m-2)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


60

M o n tm o rillo n ite

40

20

0 0 .0

0 .5

1 .0

1 .5

2 .0

2 .5

C e q (m M ) Figure 1. Adsorption of activated nucleotide 2-MeImPA (A) on zincite (circles) and montmorillonite (squares) in ultrapure water. Mineral loading = 10 mg.mL-1. Isotherms are reported as mass adsorbed, qe (µmol.m-2), as a function of the equilibrium solution concentration, Ceq. The error bars represent the standard deviation from the mean of three replicate samples.

3.2. Non-enzymatic RNA Polymerization. Polymerization of 2-MeImpA and 3’,5’cAMP mononucleotides was investigated on zincite and montmorillonite under two different experimental systems, representing two different schools in the origins of life literature: the Ferris school and the Di Mauro school.4-5, 7-16, 18-20, 43, 51 In the first system, the experimental conditions consisted of a mixture of 10 mM 2-MeImPA, 50 mg.mL-1 mineral loading, 200 mM NaCl, 75 mM MgCl2, that was incubated for 3 days at room temperature (see Methods).16 HPLC analysis


13


Page 14 of 46

revealed the presence of oligomers up to only trimers in the presence of ZnO, which was similar to the control system without minerals present (Figure 2a, b). Longer oligomers up to octamers were obtained in the presence of montmorillonite (Figure 2c). MALDI-MS analysis (Figure 2d-f) was carried out to confirm the HPLC results. The theoretical and the experimental mass-to-charge ratio (m/z) values of oligomer ions from 2-mer to 8-mer are summarized in Table S1. Again, only trimers were detected from the ZnO system similar to the control system (Figure 2d, e), whereas longer oligomers were observed with montmorillonite (Figure 2f; Table S1). Peaks corresponding to the theoretical m/z of oligomers up to 6-mer were identified from montmorillonite samples. Two oligomeric ion series were observed in MALDIMS spectrum (Figure 2f). The an ion series corresponds to linear oligomers (from 2-mer to 6-mer) of 2-MeImPA, whereas the an* series indicates ions with loss of one water molecule (an - 18 Da), showing the presence of oligomers that contain one cyclic monomer. Note that peaks for longer oligomers (7-mer and 8-mer) were found in the samples that had not been purified by “zip-tipping” (Figure S4). The absence of the 7-mer and 8-mer in the zip-tipped sample (Figure 2f) is explained by their low abundance and low recovery of these oligomers during the ziptip process. MALDI-MS analysis of the non-ziptipped sample showed aggregates of oligomers with differing lengths and multiple sodium adducts because of high salt concentration in the sample (Figure S4). Formation of such aggregates and salt adducts also resulted in a more convoluted spectrum. However, zooming in to the higher m/z values (2000-2900 m/z) allowed the identification of 7- and 8-mers (Figure S4). The amount of sodium adducts are indicated as a superscript on each oligomer ion observed in the spectrum of the non-ziptipped samples. For example, the ion observed at 2057.84 m/z, a63Na, corresponds to the 6-mer with three Na+ adducts ([M + 3Na - 4H]- (Figure S4). The presence of the ion series, which is 18 Da higher than an ions


14

Page 15 of 46

(labelled as bn), showed the formation of aggregates by non-covalent interaction of two oligomers. For example, the ion observed at 2075.88 m/z corresponds to an aggregate of two oligomers with a total number of 6 monomers and 3 Na+ adducts. 3.5

uV(x10,000)

(a) no mineral 75 mM Mg2+ 200 mM NaCl

3.0

2.5

Intens. [a.u.]

1

2.0

breakdown

1.5

1.0

0.5

3

0.0 0.0

5.0

10.0

15.0

20.0

25.0

30.0

35.0

(b) zincite 75 mM Mg2+ 200 mM NaCl

3.0

1 breakdown

2.5

2.0

(d) no mineral (a) 75 mM Mg2+ 200 mM NaCl

250 200

a2

150

675.13

50

1.5

0

(e) zincite (b) 75 mM Mg2+ 200 mM NaCl

250 200 150

2c/ 2l 3

1.0

300

100

m in

uV(x10,000)

Intens. [a.u.]

3.5

100

0.5

50 0.0 0.0 3.5

5.0

10.0

15.0

20.0

25.0

30.0

35.0

m in

uV(x10,000)

3.0

2.5

breakdown1

2.0

2c/ 2l

3

(c) mmte 75 mM Mg2+ 200 mM NaCl

4 5

1.5

0 4 x10 1.5 1.0 0.5

a2*

(f) mmte (c) 75 mM Mg2+ 200 mM NaCl

657.17

a2

675.17

a3*

a

3 986.23 1004.25

a4

1333.35

a5 a5* 1662.46

1644.45

0.5

0.0

7 8

0.0 5.0

10.0

15.0

20.0

25.0

30.0

a6 1991.56

6

1.0

0.0

Intens. [a.u.]

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


750 35.0

1000

1250

1500

1750

2000

2250

2500

2750

m/z

m in

Figure 2. Non-enzymatic polymerization of 2-MeImPA in the presence of 200 mM NaCl and 75 mM MgCl2 (classical polymerization buffer). (a)-(c) anion-exchange HPLC chromatograms and (d)-(f) MALDI-TOF MS spectra. (a, d) in the absence of mineral; (b, e) in the presence of zincite; and (c, f) in the presence of Banin-treated montmorillonite. Mineral loading = 50 mg.mL-1. In the HPLC chromatograms, “breakdown” refers to the monomer hydrolysis products. The numbers refer to the n-mer oligomer and 2c and 2l refer to the cyclic dimer and the linear dimer, respectively (see Figure S1 for structures). In the MALDI-TOF MS spectra, an is the ion series corresponding to oligomers with n-mers. The ions that are labelled with asteriks (an*) correspond to oligomers with a water loss. All of the ions observed in the spectra are deprotonated, [M - H]-.

In the second polymerization system following Di Mauro and co-workers43-44, 52, the H+form of the cyclic monomer (15 mM 3’,5’-cAMP) was incubated in dry conditions in the absence or presence of 50 mg.ml-1 montmorillonite and zincite at 60 °C for 16 hours (see Methods). Controls with and without salts (200 mM NaCl, 75 mM MgCl2) were also tested. No polymers above dimers could be detected by HPLC (Figure S5) or MALDI-MS (Figure S6). The structure


15


Page 16 of 46

of these dimers, however, cannot be ascertained definitively from the mass spectra alone, since covalently- and non-covalently bound dimers of 3’,5’-cAMP have the same empirical formula and mass. Therefore, the ion observed at 657 m/z could also correspond to a non-covalently associated cluster of 2 monomers (Figure S7).21 The difference between the previously reported polymerization of cyclic monomers and the present study may be because the original authors used 3’,5’-cGMP whereas 3’,5’-AMP is used here. The absence of polymerization even in the presence of montmorillonite indicates that different mechanistic pathways are involved in the polymerization of the activated linear (2-MeImPA) and the cyclic (3’,5’-cAMP) monomers.43-44, 51-52

Other

experimental

conditions

consisting

of

one-week

incubation

time

and

hydration/dehydration cycles with 3’,5’-cAMP were also tested (Figure S8). Only a peak of 313 m/z was dominant, corresponding to a ribosyl-pyrophosphate resulting from RNA dipurination at acidic conditions and high temperature, a finding that has been previously reported.53 However, no oligomers were detected, in contrast to previous reports.43-44 3.3. Effect of Mg2+ Concentration on Polymerization. The catalytic efficiency of zincite and montmorillonite to promote 2-MeImPA polymerization was assessed in pure water, at Mg2+ concentrations of 10 and 100 mM MgCl2, and in polymerization buffer (75 mM Mg2+ and 200 mM NaCl) on montmorillonite and zincite (Figures 3 and 4). The HPLC chromatograms (Figure 3) of the systems in pure water showed that montmorillonite catalyzes the polymerization up to pentamers (Figure 3c) while polymerization did not exceed trimers in the presence of zincite (Figure 3b). The HPLC chromatograms (Figure 4 a-d) and the MALDI-TOF spectra (Figure 4 eh) at various Mg2+ concentrations show up to tetramers in pure water (Figure 4a, e) and slightly longer oligomers of up to about 5-mers or 7-mers in the presence of 10 mM Mg2+ (Figure 4b, f).


16

Page 17 of 46 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


However, further increasing the Mg2+ concentration up to 100 mM (Figure 4c, g) or the presence of the classical polymerization buffer (Figure 4d, h) does not significantly improve catalytic efficiency, yielding up to 8-mers. Interestingly, it was also observed that up to tetramers are formed at 100 mM Mg2+ in the absence of any mineral but not at lower (10 mM Mg2+) concentration (Figure S9). 3.5

uV(x10,000)

(a) no mineral water (no salts)

3.0

1

2.5

breakdown

2.0

1.5

1.0

0.5

3 0.0

uV(x10,000)5.0 0.0 3.5

10.0

15.0

20.0

25.0

30.0

35.0

m in

(b) zincite water (no salts)

3.0

1 breakdown

2.5

2.0

1.5

2c/ 2l 3

1.0

0.5

0.0 0.0 uV(x10,000)5.0 3.5

10.0

15.0

20.0

25.0

breakdown1

2.5

35.0

m in

(c) mmte water (no salts)

2c/ 2l

3.0

30.0

3

2.0

4

1.5

1.0

0.5

5

0.0 0.0

5.0

10.0

15.0

20.0

25.0

30.0


35.0

m in

17


Figure 3. Anion-exchange HPLC chromatograms of non-enzymatic polymerization of 2-MeImPA in water (no salts present). (a) in the absence of mineral; (b) in the presence of zincite; and (c) in the presence of Banin-treated montmorillonite; the corresponding MALDI-TOF spectrum is shown in Figure 4a. Mineral loading = 50 mg.mL-1. Breakdown refers to the monomer hydrolysis. The numbers refer to the n-mer oligomer and 2c and 2l refer to the cyclic dimer and the linear dimer, respectively (see Figure S1 for structures).

In summary, the results presented above demonstrated that Mg2+ is not essential for the formation of at least short oligomers of activated, linear monomer (2-MeImPA) and that montmorillonite is a better catalyst than zincite, even though the latter was found to have a much

0.5

6

1991.927

2321.008

1333.657

1315.626

a7

a5

a7 2321.148

1992.038

a6 1814.895

1644.879

a4* a4

1662.895

(h) mmte 75 mM Mg2+ 200 mM NaCl

1662.924

a6 1992.106

1315.626

1333.725

1253.725

a5

a7

2058.071

0.5

5

a6

(g) mmte 100 mM Mg2+

1359.738

4

a4

a4*

1333.789

1.0

2 3

1004.527

986.502

a3* a3

a2*

1315.626

1.5

1253.780

(d) mmte 75 mM Mg2+ 200 mM NaCl

0 x10 4

1135.034

8

1004.615

7

1068.642

657.334

4

2

a4*

a3*

986.592

6

a2*

1048.595

breakdown

5

6

715.273 761.411

4

a5

0.0 x10 4

657.377

3

Inte ns . [a .u.]

(c) mmte 100 mM Mg2+

(f) mmte 10 mM Mg2+

1662.756

0.5

4 5

a4

a4* a4

a3* a3

1.5 1.0

3

1

2.0

986.555

1 breakdown 2

(e) mmte 0 mM Mg2+

0.0 x10 4

697.308

(b) mmte 10 mM Mg2+

Inte ns . [a .u.]

3 4

1 2 breakdown

739.372

1.0

breakdown

a4*

2321.272

1.5

1315.616

a2*

a3* a3

1333.629 1381.575

a2*

986.494 1004.502 1068.566

2.0

657.284

x10 4

657.299

1

(a) mmte 0 mM Mg2+

2

Inte ns . [a .u.]

higher adsorption capacity.

Inte ns . [a .u.]

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 18 of 46

0.0

800

1000

1200

1400

1600

1800

2000

2200

2400 m /z

Figure 4. Anion-exchange HPLC chromatograms (a-d) MALDI-TOF spectra (e-h) of 2-MeImPA reacted in the presence of Banin-treated montmorillonite at various concentrations of Mg2+ (a, e) 0 mM Mg2+; (b, f) 10 mM Mg2+; (c, g) 100 mM Mg2+; and (d, h) 75 mM Mg2+ and 200 mM NaCl. Mineral loading = 50 mg.mL-1.

Taken together, the results presented above indicate that the amount of adsorbed nucleotide is not the determining factor in the polymerization catalytic efficiency of a mineral. This suggests


18

Page 19 of 46 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


that the orientation of the nucleotides on the surface is a more important factor. In order to gain information concerning the binding conformation of nucleotides to the minerals and to correlate this binding information to the polymerization efficiency of the corresponding mineral, ATR-FTIR and

31

P ssNMR spectroscopy was performed for 2-MeImPA adsorbed on zincite and

montmorillonite. 3.4. Adsorbed Monomer Conformation by ATR-FTIR Analysis. The FTIR spectra were interpreted based on band assignments for nucleobase, nucleoside and nucleotides in solution as reported previously in the literature.36, 54-60 The bands between ~1000 and 1180 cm-1 correspond to PO4 bands of the mononucleotide (Figure 5). These bands overlap to some extent with the ribose bands between 960 and 1100 cm-1. Bands corresponding to the nucleobase ring occur at ~ 15801604 cm-1 and the amine group appears at ~ 1649 cm-1. The spectrum on ZnO shows that the bands corresponding to the phosphate group of the nucleotide are significantly altered upon adsorption to zincite, whereas the band assigned to the amine group is not affected. This suggests an innersphere complex between the phosphate moiety of the nucleotide monomer and the zinc atom of the zincite surface. If there were an outer-sphere complex in which the nucleotide is separated from the surface by an intervening water molecule, then the phosphate band would not have been shifted significantly. The spectrum of montmorillonite shows broad bands at 900-1100 cm-1 corresponding to the Si-O-Si framework (Figure 5). These bands overlap the phosphate and ribose bands of the ribonucleotide. Interestingly, the ~ 1649 cm-1 band of the ribonucleotide is shifted to 1634 cm-1 upon adsorption to montmorillonite clay indicating interaction of the amine group with the surface, indicating a H-bonded interaction.


19


Thus, the activated monomer interacts strongly by the phosphate group on zincite forming an inner-sphere surface complex and weakly through the amine group on montmorillonite forming an H-bonded complex. (a) 743

708

728 696

*

683 649

*

727

*

756

824 800

960

*

2’-MeImPA

905 875

*

*

667 649

798 783

830

*

*

*

1101

1042

*

*

*

995

1166 1203

1265

1375

1335 1302

1412

1535

1482

1649

1604 1580

*

1710

914

*

871

1018 989

1082

1039

*

*

890 875

ZnO + 2’-MeImPA

1169

1206

1268

1335

1380

1415

1647

%Transmittance

1076

1013 992

ZnO

*

1800

1600

1400

1200 1000 Wavenumbers (cm-1)

800

624

800

Mmte.

*

618

*

Mmte. + 2’-MeImPA

*

683 649

1101

756 727

824 800

905 875

2’-MeImPA

1042

*

960

995 1001

1166 1203

1265

1375

1335 1302

1412

1482

1535

1604 1580

1649

1710

916

882

850

*

798 782

1016

1122

732

*

697

1634

1118

917 882 850

1640

(b)

%Transmittance

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 20 of 46

1800

1600

1400

1200 1000 Wavenumbers (cm-1)

800

600

Figure 5. ATR-FTIR spectra of 2-MeImPA adsorbed on minerals in water. (a) zincite; and (b) Banin-treated montmorillonite


20

Page 21 of 46 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


3.5. Adsorbed Monomer Conformation by ssNMR Analysis. The NMR chemical shift interaction is characteristic in terms of the shielding effect of electron clouds surrounding the nucleus. Hence, anisotropic and isotropic chemical shifts are very sensitive to specific intermolecular interactions and environments surrounding target nuclei. 31P CP NMR spectra of 2MeImPA corresponding to magic angle spinning conditions (Figure 6a) show the nucleotide standard (bottom), adsorbed on zincite (middle) and adsorbed on montmorillonite (top). The 2MeImPA standard sample has very intense signals. Interestingly, 2-MeImPA adsorbed on zincite also provided relatively strong 31P signals, whereas the signal was very weak on montmorillonite and took 20-30 times larger accumulations than those on zincite. The CPMAS spectra of 2MeImPA on montmorillonite took about 2 days to collect. The isotropic chemical shift trend for 2MeImPA in bulk state (-9.5 ppm), adsorbed on montmorillonite (-0.5 ppm) and adsorbed on zincite (2.5 ppm) showed increasingly positive values. These chemical shift trends suggest stronger interactions between phosphate and the zincite surface than those between phosphate and the montmorillonite surface. 31

P NMR spectra collected under static conditions are shown for 2-MeImPA in Figure 6b

for the standard (bottom) and adsorbed on zincite (top). In the standard, the chemical shift anisotropy (CSA) principle values (s11, s22, s33 =103, 16, -147 ppm) are largely different from those of 2-MeImPA adsorbed on zincite (s11, s22, s33 = 63, -16, -35 ppm). This fact, again, suggests strong interaction between phosphate and surface of zincite. Note that very weak adsorption of 2MeImPA on the surface of montmorillonite does not allow us to measure CSA pattern within a reasonable time period.


21


(a) mmte

Page 22 of 46

(b)

zincite zincite

2-MeImPA

2-MeImPA

Figure 6. 31P CP NMR spectra of 2-MeImPA. Magic Angle spinning spectra (a) of 2-MeImPA in bulk (bottom), adsorbed on zincite (middle), and adsorbed on Banin-treated montmorillonite (top); red solid line indicates 31P chemical shift of 2-MeImPA; and (b) static NMR spectra for 2-MeImPA in bulk (bottom) and adsorbed on zincite (top).

3.6. Relating Adsorption Capacity, Adsorbed Conformation and Polymerization Efficiency. The primary focus of the present study was not to achieve long oligomers, but to elucidate relationships between adsorption capacity, conformation of the adsorbed monomer and polymerization catalytic efficiency of the minerals, if any. Zincite adsorbed up to 10X as much nucleotide as did montmorillonite under the present experimental conditions. However, polymerization efficiency was lower or absent on zincite. The present results are consistent with a body of results in the literature. For example, it was shown by Daniel and co-workers33, 37 that another clay mineral, nontronite, adsorbs twice as much nucleic acids as montmorillonite but, in another earlier study, nontronite was found to show little or no catalytic activity compared to montmorillonite in RNA polymerization.13


22

Page 23 of 46 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


In order to understand the adsorption capacity trends, we consider the surface charge characteristics of the two minerals. The isoelectric point (IEP) of zincite is ~ 8.449, so the surface is positively charged in water. Adsorption of the negatively charged nucleotide is, therefore, expected. Based on the crystal structure of ZnO, about 9 Zn sites per nm2 of the ZnO surface can be estimated. Assuming adsorption of one nucleotide per surface Zn atom and complete surface coverage, the apparent maximum adsorption capacity on zincite translates to ~50 molecules of nucleotide per nm2 of zincite or roughly 4 to 5 layers of monomers. In contrast, the negatively charged surface of montmorillonite (IEP ~ 2.549) is expected to repel the monomer. Yet, a small amount of adsorption is seen, presumably, due to weak van der Waals and H-bonding interactions as indicated by the spectroscopic results. Some adsorption may also occur by the formation of Mg2+ - 2MeImPA surface complexes. The present results show that adsorption of negatively charged nucleotides increases as the mineral surface charge becomes more positive suggesting a major contribution from electrostatic forces in controlling the adsorption capacity of a mineral. Since adsorption capacity is not the factor controlling polymerization catalytic efficiency of the minerals, the conformation of the adsorbed nucleotides is considered next. The FTIR and ssNMR data indicate that the phosphate moiety of the nucleotide is strongly attracted to the metal cation surface sites on positively charged surfaces, leading to covalent bond formation as seen for zincite here, and well-established in the literature for other positively charged minerals, such as goethite (a-FeOOH) and gamma-alumina (a-Al2O3).49 The spectroscopic analyses also reveal that the phosphate moiety does not interact strongly with the negativelycharged montmorillonite surface. This result indicates that Van der Waals interactions or Hbonding between the surface –OH sites and nucleobase or amine moieties contribute, albeit


23


Page 24 of 46

weakly, to adsorption on negatively-charged mineral surfaces. Our result is consistent with those from a recent molecular dynamics simulation, which showed that the amine group and the nucleobase in an activated ribonucleotide are oriented towards the silicate tetrahedral plane in the interlayer sites.61 Based on surface complexation model fits to a comprehensive set of adsorption data obtained over a range of pH and surface loadings conditions, it was previously concluded that nucleotides adsorb on swelling clays such as montmorillonite and nontronite by two mechanisms depending on the pH.33, 37 At low pH, it was proposed that nucleotides adsorb to the octahedral layer of clay at >FeOH or MgOH edge sites via phosphate forming an inner-sphere monodentate surface complexes and by ion exchange in the interlayer sites. At higher pHs, which is relevant to the pH 8 of the present study, Daniel and co-workers30, 37 proposed that nucleotides bind by forming bidentate outer-sphere complexes at >FeOH or MgOH edge sites through the phosphate group. If such outer-sphere complexes do form, spectroscopy would not capture it, hence, our study does not identify such complexes. Conversely, the previous studies of Daniel and coworkers30, 37 is based solely on model fits to bulk adsorption data without any spectroscopic analysis, so they may have missed the presence of interlayer adsorption of nucleotides at high pH as shown by our FTIR data because it is a weak adsorption. Thus, it is likely that both interlayer adsorption of nucleotides by interaction between the nucleobase and interlayer sites of montmorillonite as shown here as well as outer-sphere complexation via the phosphate group to FeOH or MgOH edge sites at the octahedral layer of montmorillonite are occurring. In summary, despite a much smaller adsorption capacity, polymerization on montmorillonite is much more efficient than on zincite because of the conformational differences of the adsorbed activated mononucleotide between the two minerals.


24

Page 25 of 46 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


3.7. Model for Catalysis of Nucletoide Polymerization. Based on the data obtained in our study, a model for the non-enzymatic polymerization on minerals is proposed herein (Figure 7). Nucleotides adsorb on montmorillonite through their nitrogenous bases or amine functional group via hydrogen bonding, consistent with the results of molecular dynamics (MD) simulations61, thus leaving the phosphate group unconstrained and available for formation of the phosphodiester bond. In contrast, nucleotides adsorb strongly through their phosphate group on positively charged minerals such as zincite, thus making the phosphate unavailable for polymerization. Taken together, our results suggest that adsorption capacity of a mineral is not positively correlated with the catalytic efficiency for polymerization as is widely held in the literature. Rather, detailed molecular-level interactions at the montmorillonite-nucleotides interface are at play wherein the orientation of the phosphate moiety makes it accessible for interaction with another monomer to allow for polymerization. It is important to note that not all negatively charged minerals catalyze RNA polymerization; indeed, not even all clays are catalysts.13, 20 This suggests that the extent and the nature of isomorphous substitution in the aluminosilicate framework affects the interlayer charge of the clays. Even among the montmorillonites, only certain clays pretreated in a specific manner (the “Banin procedure”) showed catalytic properties and it was conjectured that an acid-base catalysis mechanism involving –OH sites of clay is responsible.16-17 The –OH sites are found only at the edges of clay platelets but, in another study from the same research group, the nucleotide was proposed to occupy the inter-layer sites of clay8 (Figure 7). Thus, the precise location of surface sites where polymerization occurs remain unknown. Previous MD simulation results suggest a preference of the interlayer sites for polymerization.61 In summary, the ideal mineral catalyst should allow the phosphate moiety to be available for polymerization while possibly also


25


Page 26 of 46

providing a mechanism for acid-base catalysis. The catalytic role of montmorillonite in this model would be to facilitate the proper alignment of nucleotides, similar to their alignment on a complementary RNA template, as it was originally proposed by Ferris.14

Figure 7. Proposed model for adsorption-polymerization relationship on zincite (a) and montmorillonite (b). On zincite, The phosphate moeity is strongly bonded to mineral surface, thus unavailable for polymerization. On montmorillonite, adsorption on montmorillonite occurs via the nucleobases or amine functional group of the monomer such that the phosphate moeity is accessible to an incoming (or adjacent adsorbed) monomer for polymerization. The schematic is intended only to imply association of the activated nucleotide with the surface. Yellow circle represents 2-Methylimidazole phosphate.

We have shown above that, in order to be catalytic to nucleotide polymerization, a mineral must adsorb the nucleotide with the proper conformation where the phosphate remains available for polymerization. Furthermore, we propose that a mineral must also provide a nano-confined environment in which the activity of water is reduced in order to be catalytic. The decreased activity of water facilitates the elimination of a water molecule between the two nucleotides undergoing polymerization. The ~1 nm spacing of the interlayer sites of montmorillonite provide


26

Page 27 of 46 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


such a nano-confined environment. Thus, montmorillonite fulfils the various requirements for a mineral to be catalytic. Finally, we address the fact that Banin-treated clays are catalytic, whereas clays treated by simple ion-exchange are not. In the Banin procedure, the clay is washed thoroughly with a strong acid at low pH and then titrated back to pH 7 using NaOH. This ensures thorough removal of all cations from the interlayer sites and results in a pure Na-exchanged clay. The standard ionexchange process may leave traces of other cations. Taken together, these observations suggest that cations that are strongly hydrated, such as alkali cations or Mg2+, and form outer-sphere ion complexes with the phosphate moiety of the ribonucleotide are permissible in the polymerization reaction. The formation of the ion complex reduces the repulsion between the nucleotide and the negatively charged montmorillonite surface or second nucleotide, while still allowing the cation to be displaced easily by the second nucleotide during the formation of the phosphodiester bond. Other cations, such as Ca2+, which are less strongly hydrated are more easily desolvated and may form inner sphere complexes with the phosphate moiety of the nucleotide, such that the phosphate is less available for polymerization. 3.8. Role of Magnesium. It was found here that magnesium was not required to achieve short oligomers up to pentamers in the presence of montmorillonite (Figure 3c, Figure 4a). The addition of 10 mM Mg2+ promoted the formation of octamers and oligomer length did not increase up to 100 mM Mg2+ (Figure 4). Thus, magnesium plays the role of a co-catalyst to montmorillonite and a small concentration is as efficient as a much higher concentration. This inference is in agreement with the fact the polymerization up to 10-mers was also achieved at high monovalent alkali cation concentrations (0.8-1.2 M) in the presence of montmorillonite but without any divalent cations present.19


27


Page 28 of 46

Indeed, the formation of up to pentamers in pure water (in the absence of any salts) shown for the first time in the present study indicates that montmorillonite is inherently catalytic (Figure 3c, 4a). The presence of salts aids the polymerization by reducing charge repulsion as the oligomers get longer. The ability to achieve oligomers even at low magnesium concentrations in the presence of montmorillonite extends the range of geochemical environments in which montmorillonitecatalyzed polymerization may have occurred on early Earth. Interestingly, 100 mM Mg2+ was able to catalyze the formation of up to tetramers (Figure S9) in the absence of montmorillonite though not at lower concentrations. However, such a high concentration of magnesium is implausible in almost any prebiotic geochemical environment. 3.9. Oligomer Length. It is worth noting that only oligomers up to approximately 10-mers have been unambiguously identified in the literature.5, 18, 21 One study claimed longer polymers up to about 40-mers10, but only gel electrophoresis and HPLC were used in that study without mass spectrometry to confirm the nature of the gel electrophoresis products. In a different approach, it has been stated that unactivated 5’-AMP polymerizes in the presence of lipid molecules during repeated wetting-drying cycles53, 62-63 and a potential mechanism has been proposed.64 In these studies, again, only very short RNA oligomers (< 8-mers) have been definitively identified by mass spectroscopy.53 The result that only short oligomers are produced non-enzymatically by montmorillonite catalysis has implications for the RNA world theory, where large RNA polymers are assumed to have played the role of both informational and catalytic molecules. Clearly, obtaining such large polymers remains a challenge in the origins of life field. On the other hand, conversion of nonfunctional short RNA strands to functional RNA by primer extension has been recently


28

Page 29 of 46 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


demonstrated2, so it is possible that very long polymers of RNA are not required. However, obtaining even oligomers as long as the original2 template has currently not been shown nonenzymatically. 4. CONCLUSIONS In summary, the results presented in the present work indicate that greater adsorption of nucleotides on minerals does not necessarily lead to increased polymerization. Rather, the orientation of the monomers at the mineral/water interface is critical, where the phosphate moiety should be available for polymerization. Magnesium appears not to be essential for oligomerization but acts, rather, as a co-catalyst in the presence of Banin-treated montmorillonite. Short oligomers were obtained even in pure water in the absence of any salts. The present results allow us to propose that at least three properties of a mineral must be fulfilled in order for it to be catalytic in nucleotide polymerization. The mineral must adsorb the mononucleotide in the proper orientation where phosphate is still available for formation of the phosphodiester bond, a nano-confined environment (e.g., the interlayer sites of montmorillonite) which promotes the dehydration reaction between two nucleotides, and outer sphere complexation of cations (e.g., Mg2+ or high alkali cations) with the phosphate moiety of the nucleotide, so that the cation can be easily displaced to form the phosphodiester bond. In the absence of any mineral, up to tetramers were obtained in the presence of 100 mM Mg2+. Finally, obtaining long RNA polymers remains an open challenge in the origins of life field. The results of the present study help resolve an open question for three decades as to what makes a mineral catalytic toward activated ribonucleotide polymerization and should help the narrow the search for other catalytic minerals among the thousands of known minerals.


29


Page 30 of 46

ASSOCIATED CONTENT Supporting Information Table S1 Supplementary Figures S1 – S9 The Supporting Information is available free of charge on the ACS Publications website. ACKNOWLEDGMENTS N.S. is grateful for financial support from the Simons Foundation Origins of Life Initiative award # 290359, the NSF EAR grant # EAR-1251479 and “start-up” funds from the University of Akron; T. M. gratefully acknowledges NSF DMR 1708999; C. W. gratefully acknowledges NSFCHE-1808115 .


30

Page 31 of 46 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


REFERENCES (1) Kaddour, H.; Sahai, N. Synergism and Mutualism in Non-Enzymatic RNA Polymerization. Life 2014, 4, 598-620. (2) Adamala, K.; Engelhart, A. E.; Szostak, J. W. Generation of Functional RNAs from Inactive Oligonucleotide Complexes by Non-Enzymatic Primer Extension. J. Am. Chem. Soc. 2014, 137, 483-489. (3) Cleaves II, H. J.; Scott, A. M.; Hill, F. C.; Leszczynski, J.; Sahai, N.; Hazen, R. Mineral– Organic Interfacial Processes: Potential Roles in the Origins of Life. Chem. Soc. Rev. 2012, 41, 5502-5525. (4) Ferris, J. P.; Ertem, G.; Agarwal, V. Mineral Catalysis of the Formation of Dimers of 5′AMP in Aqueous Solution: The Possible Role of Montmorillonite Clays in the Prebiotic Synthesis of RNA. Orig. Life Evol. Biosph. 1989, 19, 165-178. (5) Ferris, J. P.; Ertem, G., Oligomerization of Ribonucleotides on Montmorillonite: Reaction of the 5′-Phosphorimidazolide of Adenosine. Science 1992, 257, 1387-1389. (6) Ferris, J. P.; Hill Jr, A. R.; Liu, R.; Orgel, L. E. Synthesis of Long Prebiotic Oligomers on Mineral Surfaces. Nature 1996, 381, 59. (7) Ertem, G.; Ferris, J. P. Synthesis of RNA Oligomers on Heterogeneous Templates. Nature 1996, 379, 238. (8) Ertem, G.; Ferris, J. P. Formation of RNA Oligomers on Montmorillonite: Site of Catalysis. Orig. Life Evol. Biosph. 1998, 28, 485-499. (9) Franchi, M.; Ferris, J. P.; Gallori, E. Cations as Mediators of the Adsorption of Nucleic Acids on Clay Surfaces in Prebiotic Environments. Orig. Life Evol. Biosph. 2003, 33, 1-16. (10) Huang, W.; Ferris, J. P. Synthesis of 35–40 Mers of RNA Oligomers from Unblocked Monomers. A Simple Approach to the RNA World. Chem. Comm. 2003, 1458-1459. (11) Huang, W.; Ferris, J. P. One-Step, Regioselective Synthesis of up to 50-Mers of RNA Oligomers by Montmorillonite Catalysis. J. Am. Chem. Soc. 2006, 128, 8914-8919. (12) Ferris, J. P. Mineral Catalysis and Prebiotic Synthesis: Montmorillonite-Catalyzed Formation of RNA. Elements 2005, 1, 145-149. (13) Miyakawa, S.; Joshi, P. C.; Gaffey, M. J.; Gonzalez-Toril, E.; Hyland, C.; Ross, T.; Rybij, K.; Ferris, J. P. Studies in the Mineral and Salt-Catalyzed Formation of RNA Oligomers. Orig. Life Evol. Biosph. 2006, 36, 343-361. (14) Ferris, J. P. Montmorillonite-Catalysed Formation of RNA Oligomers: The Possible Role of Catalysis in the Origins of Life. Philos. Trans. R. Soc. B: Biol. Sci. 2006, 361, 1777-1786. (15) Joshi, P. C.; Pitsch, S.; Ferris, J. P. Selectivity of Montmorillonite Catalyzed Prebiotic Reactions of D, L-Nucleotides. Orig. Life Evol. Biosph. 2007, 37, 3-26. (16) Joshi, P. C.; Aldersley, M. F.; Delano, J. W.; Ferris, J. P. Mechanism of Montmorillonite Catalysis in the Formation of RNA Oligomers. J. Am. Chem. Soc. 2009, 131, 13369-13374. (17) Aldersley, M. F.; Joshi, P. C.; Price, J. D.; Ferris, J. P. The Role of Montmorillonite in Its Catalysis of RNA Synthesis. Appl. Clay Sci. 2011, 54, 1-14. (18) Aldersley, M. F.; Joshi, P. C.; Huang, Y. The Comparison of Hydrochloric Acid and Phosphoric acid Treatments in the Preparation of Montmorillonite Catalysts for RNA Synthesis. Orig. Life Evol. Biosph. 2017, 1-8. (19) Joshi, P. C.; Aldersley, M. F. Significance of Mineral Salts in Prebiotic RNA Synthesis Catalyzed by Montmorillonite. J. Mol. Evol. 2013, 76, 371-379.


31


Page 32 of 46

(20) Ferris, J. P.; Ertem, G. Oligomerization Reactions of Ribonucleotides: The Reaction of the 5′-Phosphorimidazolide of Nucleosides on Montmorillonite and Other Minerals. Orig. Life Evol. Biosph. 1992, 22, 369-381. (21) Burcar, B. T.; Cassidy, L. M.; Moriarty, E. M.; Joshi, P. C.; Coari, K. M.; McGown, L. B. Potential Pitfalls in Maldi-Tof MS Analysis of Abiotically Synthesized RNA Oligonucleotides. Orig. Life Evol. Biosph. 2013, 43, 247-261. (22) Bernal, J. D. The Physical Basis of Life. Proc. Phys. Soc. A 1949, 62, 537. (23) Sowerby, S. J.; Cohn, C. A.; Heckl, W. M.; Holm, N. G. Differential Adsorption of Nucleic Acid Bases: Relevance to the Origin of Life. PNAS 2001, 98, 820-822. (24) Maurel, M.-C.; Leclerc, F. From Foundation Stones to Life: Concepts and Results. Elements 2016, 12, 407-412. (25) Wächtershäuser, G. Before Enzymes and Templates: Theory of Surface Metabolism. Microbiol. Rev. 1988, 52, 452-484. (26) Biondi, E.; Branciamore, S.; Maurel, M.-C.; Gallori, E. Montmorillonite Protection of an UV-Irradiated Hairpin Ribozyme: Evolution of the RNA World in a Mineral Environment. BMC Evol. Biol. 2007, 7, S2. (27) Graf, G.; Lagaly, G. Interaction of Clay Minerals with Adenosine-5-Phosphates. Clays Clay Miner. 1980, 28, 12. (28) Holm, N. G.; Ertem, G.; Ferris, J. P. The Binding and Reactions of Nucleotides and Polynucleotides on Iron Oxide Hydroxide Polymorphs. Orig. Life Evol. Biosph. 1993, 23, 195215. (29) Winter, D.; Zubay, G. Binding of Adenine and Adenine-Related Compounds to the Clay Montmorillonite and the Mineral Hydroxylapatite. Orig. Life Evol. Biosph. 1995, 25, 61-81. (30) Perezgasga, L.; Serrato-Díaz, A.; Negron-Mendoza, A.; Gal’N, L. D. P.; Mosqueira, F. Sites of Adsorption of Adenine, Uracil, and Their Corresponding Derivatives on Sodium Montmorillonite. Orig. Life Evol. Biosph. 2005, 35, 91-110. (31) Plekan, O.; Feyer, V.; Šutara, F.; Skála, T.; Švec, M.; Cháb, V.; Matolín, V.; Prince, K. The Adsorption of Adenine on Mineral Surfaces: Iron Pyrite and Silicon Dioxide. Surf. Sci. 2007, 601, 1973-1980. (32) Cleaves, H. J.; Jonsson, C. M.; Jonsson, C. L.; Sverjensky, D. A.; Hazen, R. M. Adsorption of Nucleic Acid Components on Rutile (TiO2) Surfaces. Astrobiology 2010, 10, 311-323. (33) Feuillie, C.; Daniel, I.; Michot, L. J.; Pedreira-Segade, U. Adsorption of Nucleotides onto Fe–Mg–Al Rich Swelling Clays. Geochim. Cosmochim. Acta 2013, 120, 97-108. (34) Feuillie, C. c.; Sverjensky, D. A.; Hazen, R. M. Attachment of Ribonucleotides on ΑAlumina as a Function of pH, Ionic Strength, and Surface Loading. Langmuir 2014, 31, 240-248. (35) Hashizume, H. Adsorption of Nucleic Acid Bases, Ribose, and Phosphate by Some Clay Minerals. Life 2015, 5, 637-650. (36) Iqubal, M. A.; Sharma, R. Studies on Interaction of Ribonucleotides with Zinc Ferrite Nanoparticles Using Spectroscopic and Microscopic Techniques. Karbala International Journal of Modern Science 2015, 1, 49-59. (37) Pedreira-Segade, U.; Feuillie, C.; Pelletier, M.; Michot, L. J.; Daniel, I. Adsorption of Nucleotides onto Ferromagnesian Phyllosilicates: Significance for the Origin of Life. Geochim et Cosmochim. Acta 2016, 176, 81-95. (38) Biondi, E.; Furukawa, Y.; Kawai, J.; Benner, S. A. Adsorption of RNA on Mineral Surfaces and Mineral Precipitates. Beilstein J. Org. Chem. 2017, 13, 393.


32

Page 33 of 46 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


(39) Banin, A.; Lawless, J. G.; Mazzurco, J.; Church, F. M.; Margulies, L.; Orenberg, J. B. pH Profile of the Adsorption of Nucleotides onto Montmorillonite. Orig. Life Evol. Biosph. 1985, 15, 89-101. (40) Adamala, K.; Szostak, J. W. Nonenzymatic Template-Directed RNA Synthesis inside Model Protocells. Science 2013, 342, 1098-1100. (41) Powner, M. W.; Gerland, B.; Sutherland, J. D. Synthesis of Activated Pyrimidine Ribonucleotides in Prebiotically Plausible Conditions. Nature 2009, 459, 239-242. (42) Scott, W. G.; Szöke, A.; Blaustein, J.; O'Rourke, S. M.; Robertson, M. P. RNA Catalysis, Thermodynamics and the Origin of Life. Life 2014, 4, 131-141. (43) Costanzo, G.; Saladino, R.; Botta, G.; Giorgi, A.; Scipioni, A.; Pino, S.; Di Mauro, E. Generation of Rna Molecules by a Base-Catalysed Click-Like Reaction. ChemBioChem 2012, 13, 999-1008. (44) Morasch, M.; Mast, C. B.; Langer, J. K.; Schilcher, P.; Braun, D. Dry Polymerization of 3′, 5′-Cyclic Gmp to Long Strands of RNA. ChemBioChem 2014, 15, 879-883. (45) Hashizume, H.; Theng, B. K. Adenine, Adenosine, Ribose and 5′-Amp Adsorption to Allophane. Clays Clay Miner. 2007, 55, 599-605. (46) Arora, A. K. Interaction of Ribose Nucleotides with Zinc Oxide and Relevance in Chemical Evolution. Colloids Surf. A: Physicochem. Eng. Asp. 2007, 298, 186-191. (47) Clayden, N. J.; Waugh, J. S. Solid State 31P NMR Spectroscopy of 5′-AMP Adsorbed on Zinc-Exchanged Clay. J. Chem. Soc., Chem. Comm. 1983, 292-293. (48) Fry, R. A.; Kwon, K. D.; Komarneni, S.; Kubicki, J. D.; Mueller, K. T. Solid-State NMR and Computational Chemistry Study of Mononucleotides Adsorbed to Alumina. Langmuir 2006, 22, 9281-9286. (49) Sahai, N.; Kaddour, H.; Dalai, P.; Wang, Z.; Bass, G.; Gao, M. Mineral Surface Chemistry and Nanoparticle-Aggregation Control Membrane Self-Assembly. Sci. Rep. 2017, 7, 43418. (50) Joyce, G.; Inoue, T.; Orgel, L. Non-Enzymatic Template-Directed Synthesis on Rna Random Copolymers: Poly (C, U) Templates. J. Mol. Biol. 1984, 176, 279-306. (51) Costanzo, G.; Pino, S.; Timperio, A. M.; Šponer, J. E.; Šponer, J.; Nováková, O.; Šedo, O.; Zdráhal, Z.; Di Mauro, E. Non-Enzymatic Oligomerization of 3’, 5’cyclic AMP. PloS one 2016, 11, e0165723. (52) Šponer, J. E.; Šponer, J. í.; Giorgi, A.; Di Mauro, E.; Pino, S.; Costanzo, G. Untemplated Nonenzymatic Polymerization of 3′,5′cGMP: A Plausible Route to 3′, 5′-Linked Oligonucleotides in Primordia. J. Phys. Chem. B 2015, 119, 2979-2989. (53) Mungi, C. V.; Rajamani, S. Characterization of RNA-Like Oligomers from Lipid-Assisted Nonenzymatic Synthesis: Implications for Origin of Informational Molecules on Early Earth. Life 2015, 5, 65-84. (54) Angell, C. 100. An Infrared Spectroscopic Investigation of Nucleic Acid Constituents. J. Chem. Soc. (Resumed) 1961, 504-515. (55) Tsuboi, M. Vibrational Spectra of Phosphite and Hypophosphite Anions, and the Characteristic Frequencies of PO3--and PO2- Groups. J. Am. Chem. Soc. 1957, 79, 1351-1354. (56) Morgan, R.; Blout, E. Infrared Spectra and the Structure of Polyriboadenylic Acid. J. Am. Chem. Soc. 1959, 81, 4625-4629. (57) Khalil, F. L.; Brown, T. L. Infrared Spectra of Adenosine Triphosphate Complexes in Deuterium Oxide Solution. J. Am. Chem. Soc. 1964, 86, 5113-5117. (58) Kundu, J.; Neumann, O.; Janesko, B.; Zhang, D.; Lal, S.; Barhoumi, A.; Scuseria, G.; Halas, N. Adenine− and Adenosine Monophosphate (Amp)− Gold Binding Interactions Studied


33


Page 34 of 46

by Surface-Enhanced Raman and Infrared Spectroscopies. J. Phys. Chem. C 2009, 113, 1439014397. (59) Swoboda, A.; Kunze, G. Infrared Study of Pyridine Adsorbed on Montmorillonite Surfaces. Clays Clay Miner. 1964, 13, 277. (60) Theophanides, T.; Tajmir-Riahi, H. Spectroscopic Properties of Metal-Nucleotide and Metal-Nucleic Acid Interactions. In Spectroscopy of Biological Molecules, Springer: 1984; pp 137-152. (61) Mathew, D. C.; Luthey-Schulten, Z. Influence of Montmorillonite on Nucleotide Oligomerization Reactions: A Molecular Dynamics Study. Orig. Life Evol. Biosph. 2010, 40, 303317. (62) Rajamani, S.; Vlassov, A.; Benner, S.; Coombs, A.; Olasagasti, F.; Deamer, D. LipidAssisted Synthesis of Rna-Like Polymers from Mononucleotides. Orig. Life Evol. Biosph. 2008, 38, 57-74. (63) Deamer, D. The Role of Lipid Membranes in Life’s Origin. Life 2017, 7, 5. (64) Toppozini, L.; Dies, H.; Deamer, D. W.; Rheinstädter, M. C. Adenosine Monophosphate Forms Ordered Arrays in Multilamellar Lipid Matrices: Insights into Assembly of Nucleic Acid for Primitive Life. PloS one 2013, 8, e62810.


34

Page 35 of 46 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


TOC Graphic


35


Page 36 of 46

Supporting Information Non-Enzymatic RNA Polymerization at the Mineral-Water Interface: An Insight into the Adsorption-Polymerization Relationships Hussein Kaddour1,2,+ Selim Gerislioglu3, Punam Dalai1, Toshikazu Miyoshi1, Chrys Wesdemiotis1,3, and Nita Sahai1,4,5, * 1 2

Department of Polymer Science, 170 University Ave., University of Akron, Akron, OH 44325

Present address: Department of Pharmacology, Stony Brook University, 101 Nicolls Rd, Stony Brook, NY, 11794–8651, USA. 3 4 5

Department of Chemistry, University of Akron, Akron, OH 44325

Department of Geosciences, University of Akron, Akron, OH 44325

Integrated Bioscience Program, University of Akron, Akron, OH 44325 *

Corresponding author: [email protected]

+

Present address: Department of Pharmacological Sciences, Stony Brook University, Stony Brook, NY 11794, USA This file includes: Table S1 Supplementary Figures S1 – S9


S1

Page 37 of 46 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Table S1. Measured and Theoretical m/z Values of the Oligomers of 2-MeImPA that are Observed in MALDI-MS Spectra. Oligomer a2*

Measured m/z 657.17*

Theoretical m/z 657.10*

Observed Ion Type

a2

675.17

675.11

[M - H]

a3* a3 a4* a4

986.23* 1004.25 1315.34* 1333.35

986.15* 1004.16 1315.20* 1333.21

[M - H] [M - H] [M - H] [M - H] [M - H]

a5*

1644.45*

1644.26*

[M - H]

a5

1662.45

1662.27

[M - H]

a6 a7

3Na

a74Na a8

1991.56 2387.01 2409.00

1991.32 2386.32 2409.00

[M - H]

-

n 2

-

2

-

3

-

3

-

4

-

4

-

5

-

5

-

[M + 3Na - 4H] [M + 4Na - 5H]

6 -

7

-

7

-

5Na

2759.34 2760.11 8 [M + 5Na - 6H] The m/z values with the asterisk “*” correspond to oligomer ions with one water loss. n: number of monomers present in each oligomer chain.


S2


(a)

(b)

(d)

Page 38 of 46

(c)

(e)

Figure S1. Structure of nucleotides used in this study: (a) 5’-AMP; (b) 2-MeImPA; (c) 3’,5’-cAMP; (d) cyclic and (e) linear dimers, which are products of 2-MeImPA polymerization.


S3

Page 39 of 46

Datafile Name:water.lcd Sample Name:water no pH adj Sample ID:1 mM

m AU

2-MeImPA

40

30

97.096

psi

(a)

3250

3000

dimer

0.442

2750

1.106

cAMP

10

1.356

20

AMP

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


2500

0 0.0

2.5

5.0

Intens. 4 x10

7.5

10.0

m in

(b)

410.0

2.5

2.0 1.5 1.0

0.5 311.5 346.1 0.0

200

300

400

500

600

700

800

m/z

Figure S2. (a) Reverse phase HPLC analysis and (b) ESI mass spectrometry of 2MeImPA. 346.1 and 410.0 correspond to unreacted 5’-AMP and 2-MeImPA, respectively.


S4


[MB - H]-

[MB + Na - 2H]-

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 40 of 46

[ML - H]-

[2ML - H]-

[MA - H]-

Figure S3. ESI analysis of the polymerization of the 2MeImPA for 3 days at 25 oC in the presence of montmorillonite, 200 mM NaCl, 75 mM MgCl2 and 200 mM bicine pH 8. ML: linear monomer; MA: activated monomer; MB: bicine-monomer; 2ML: linear dimer.


S5

x10 4

(a)

(a)

1377.62

1048.47

1.25 1.00 0.75

697.37

1728.72

0.50 2079.81

0.25

2470.94

2800.25

0.00 1000 Intens. [a.u.]

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Intens. [a.u.]

Page 41 of 46

b6

2500

3Na

a6

1500

2075.88 2079.81

a6

2000

2000

2500

3000

3500 m/z

4Na

(b) (b)

b64Na

2097.83

3Na

a65Na

2057.84

b75Na

2119.82

2448.97

b74Na

1500

b76Na

2470.94

b85Na

2778.23

2427.02

a74Na

1000

b86Na

2800.25

a85Na

2409.00

2760.11

500

2000

2100

2200

2300

2400

2500

2600

2700

2800

m/z

Figure S4. (a) MALDI-MS of non-ziptipped sample of Figure 3f and (b) mass spectrum zoomed in between 2000-2900 m/z. anmNa: ion series corresponding to oligomers with n-mers. bnmNa: ions series formed by non-covalent interaction of two oligomers, where n indicates the total of monomers present in the aggregate. The subscript “mNa” on each ion indicates the amount of sodium adducts. No covalently bound oligomer ions (anmNa) observed above 8-mers (above ~3000 m/z in top spectrum).


S6


pure water (no salts) (a) no mineral

uV(x1 0 ,0 0 0 ) 3 .5 3 .0

Absorbance 260 nm (x10,000 µAU)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

2 .0

2 .0

1 .5

1 .5

1 .0

1 .0

0 .5

0 .5 0 .0 5 .0

1 0 .0

1 5 .0

2 0 .0

2 5 .0

m in

(b) zincite

3 .0

0uV(x1 .0 0 ,0 0 0 ) 3 .5

2 .5

2 .0

2 .0

1 .5

1 .5

1 .0

1 .0

0 .5

0 .5

0 .0

0 .0 5 .0

1 0 .0

1 5 .0

2 0 .0

2 5 .0

m in

(c) mmte

3 .0

0uV(x1 .0 0 ,0 0 0 ) 3 .5

1 0 .0

1 5 .0

2 .5

2 .0

2 .0

1 .5

1 .5

1 .0

1 .0

0 .5

0 .5

0 .0

2 0 .0

2 5 .0

m in

(e) zincite

5 .0

1 0 .0

1 5 .0

2 0 .0

2 5 .0

m in

(f) mmte

3 .0

2 .5

0 .0

5 .0

3 .0

2 .5

0uV(x1 .0 0 ,0 0 0 ) 3 .5

(d) no mineral

3 .0 2 .5

0uV(x1 .0 0 ,0 0 0 ) 3 .5

with salts

uV(x1 0 ,0 0 0 ) 3 .5

2 .5

0 .0

Page 42 of 46

0 .0 5 .0

1 0 .0

1 5 .0

2 0 .0

2 5 .0

m in

0 .0

5 .0

Retention time (min)

1 0 .0

1 5 .0

2 0 .0

2 5 .0

m in

Figure S5. HPLC chromatograms of dry polymerization of 3’,5’-cAMP in the absence of mineral and the presence of 50 mg.ml-1 zincite or montmorillonite at 60°C for 16 hours in the pure water without any salts (left panel) and in the presence of polymerization buffer (200 mM NaCl, 75 mM MgCl2) (right panel). No oligomers above dimers could be detected.


S7

4

328.01

x10 4 1.25

Inte ns . [a .u.]

no salts (a) no mineral

x10 4

327.98

390.88

0.75

2

390.91

1

0.50

463.99

0.25

589.07 657.18 0 400

500

600

700 m /z

(b) zincite

328.04

3

300

Inte ns . [a .u.]

Inte ns . [a .u.]

424.87

657.22

524.99

0.00

300

x10 4

with salts (d) no mineral

1.00

3

400

500

600 m /z

x10 4

327.98

(e) zincite

2.0 1.5

2

390.88 1.0

1

390.94

493.12

0.5

657.24

300

2.5

400

500

300

600

328.02

(c) mmte

390.92

2.0

400

500

600

700 m /z

m /z

Inte ns . [a .u.]

x10 4

657.14

493.04

0.0

0

Inte ns . [a .u.]

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Inte ns . [a .u.]

Page 43 of 46

x10 4

328.06

(f) mmte

3 390.97 2

1.5

446.01

1.0

1 479.98

0.5

567.09

464.06

657.19

589.12

0.0

657.26

0 300

400

500

600

700 m /z

300

400

500

600 m /z

Figure S6. MALDI-MS analysis of dry polymerization of 3’,5’-cAMP in the absence of mineral and the presence of 50 mg.ml-1 zincite or montmorillonite at 60°C for 16 hours in pure water without any salts present (left panel) and in the presence of polymerization buffer salts (200 mM NaCl, 75 mM MgCl2) (right panel). No oligomers above dimers could be detected.


S8


(a)

(b)

Page 44 of 46

(c)

Figure S7. Possible structure for 657.1 m/z: (a) 5’,2’l-dimer with one 3’,5’-cAMP monomer; (b) 3’,5’c-dimer; and (c) non-covalent aggregate of two 3’,5’-cAMP monomers.

Figure S8. ESI-MS analysis of the polymerization of the 3’,5’-cAMP H+-form after seven hydration/dehydration conditions at 85 °C. m/z values of 164 and 313 correspond to adenine and ribosyl-pyrophosphate, respectively.


S9

Page 45 of 46

(a)

0 mM Mg2+

10 mM Mg2+ Intens. [a.u.]

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


x10

4

2.5

a2

(b)

675.20

100 mM Mg2+ 2.0

1

100 mM Mg2+

2 breakdown

1.5

1.0

3 4

a3

0.5

b41Na

1004.32 797.16

1373.37

1076.25

b51Na 1702.42

0.0 750

1000

1250

1500

1750

2000

2250

2500

2750

m/z

Figure S9. Polymerization of of 2-MeImPA in the absence of any mineral and the presence of 100 mM MgCl2 for 3 days at 25 °C. (a) HPLC chromatogram and (b) MALDI-TOF spectrum at 100 mM MgCl2. In the MALDI-TOF spectra, bnmNa is the ions series formed by non-covalent interaction of two oligomers where n indicates the total of monomers present in the aggregate. The subscript “mNa” on each ion indicates the amount of sodium adducts. No covalently bound oligomer ions (an) observed above trimer.


S10


TOC Graphic

Adsorption of activated RNA nucleotide monomer in the proper conformation, as on montmorillonite, is more critical for non-enzymatic mineral-promoted polymerization rather than a greater amount of monomer adsorption with the incorrect conformation, as in the case of zincite.


Page 46 of 46

Nonenzymatic RNA Oligomerization at the MineralâWater Interface

Recommend Documents

Nonenzymatic RNA Oligomerization at the MineralâWater Interface