Introduction

Since the discovery of its catalytic properties, RNA has attracted significant interest in studies of the origins of terrestrial life as it has the ability to serve as both a genetic blueprint and an enzyme to catalyze the reactions necessary to replicate this blueprint (Guerrier-Takada et al. 1983; Kruger et al. 1982), possibly leading to a world biochemically dominated by RNA. Such an “RNA World” is appealing because it provides a simpler system based solely on RNA when compared to modern biology’s dependence on RNA, DNA, and proteins (Orgel 1986; Gesteland et al. 2006). RNA-based organisms would evolve to produce DNA and proteins, superior molecules for their specific biochemical roles, ultimately leading to the biochemical diversity seen within modern organisms (Olsen and Woese 1997; Forterre 2001, 2002). This evolutionary pathway allows all life as we know it to be derived from a much simpler world whose ecosystem would owe its existence to the unique properties of RNA.

A RNA World requires abiotic routes to long, catalytic RNA oligomers in order to give rise to the biochemical diversity needed to support early life. Intensive research efforts have led to reports of formation of monomeric ribonucleotides under prebiotic conditions (Powner et al. 2009), albeit with low (<1 %) yields and many competing side products. Oligomerization of ribonucleotides has been achieved on a small scale using circular ribonucleotides (Costanzo et al. 2009; Morasch et al. 2014), imidazole-activated ribonucleotides with a montmorillonite catalyst (Ferris 2002, 2005; Joshi et al. 2011), and ribonucleotides in the presence of lipids (Rajamani et al. 2008). However, these reactions generally require activation of the monomers in order to achieve significant yields of longer oligomers.

This work focuses on RNA oligomerization from a collection of pre-formed ribonucleotides, directly building on work demonstrating oligomerization of imidazole-activated ribonucleotides using a montmorillonite clay catalyst (Ferris 2002, 2005; Joshi et al. 2011). In that work, organic synthesis was used to replace an oxygen atom in the phosphate group of ribonucleotides with an imidazole group (Pfeffer et al. 2005). Imidazole has been shown to be a good leaving group leading to oligomer formation when used in conjunction with catalytically activated montmorillonite clay (MMC) (Fig. 1). Since the imidazole-activated ribonucleotides were generated using organic solvents and controlled reactions, the prebiotic formation of these modified nucleotides on early Earth is questionable. Previously, researchers have attempted to address this issue by adding water-soluble carbodiimides to create high-energy reaction intermediates, but were able to produce only dimers in solution (Ferris and Kamaluddin 1989; Kawamura and Okamoto 2000).

Fig. 1
figure 1

Oligomerization using pre-activated ribonucleotides (ImpX) with montmorillonite clay catalyst

Full size image

Imidazolation can be accomplished in aqueous solution using 1-Ethyl-3-(3-dimethylaminopropyl)carbodiimide (EDC) with imidazole. This approach is commonly used for cross-linking in biochemical applications (Hermanson 2008). Although EDC is not generally recognized as a prebiotic molecule, pathways have been demonstrated for synthesizing carbodiimides through isomerization of cyanamide under prebiotic conditions (Aylward 2012; Duvernay et al. 2004). Here, we applied this chemistry to create imidazole-activated ribonucleotides in situ for RNA oligomerization and tested it in the absence and presence of catalytic montmorillonite clay (Fig. 2). In the context of prebiotic chemistry, this approach offers important advantages over pre-activation of the nucleotides prior to the oligomerization reaction. This activation reaction is performed in aqueous solution rather than organic solvents, and it occurs in situ, as does the oligomerization reaction, which may be a more realistic model for prebiotic chemistry.

Fig. 2
figure 2

Oligomerization using in situ EDC activation of ribonucleotides with montmorillonite clay catalyst

Full size image

Materials and Methods

Materials

Adenosine monophosphate (AMP, free acid) and guanosine monophosphate (GMP, disodium salt) and all other reagents, except where noted, were purchased from Sigma (St. Louis, MO, USA). Montmorillonite in the form of volclay SPV-200 was a gift from The American Colloid Company (Arlington Heights, IL, USA). Low molecular weight oligonucleotide calibration standard comprising 4-mer, 5-mer, 7-mer, 9-mer, and 11-mer DNA oligonucleotides was purchased from Bruker Daltonics (Billerica, MA, USA). Ultrapure, 18 MΩ deionized was produced in-house and used for all sample preparations and experiments.

Montmorillonite Clay Activation

Catalytically active montmorillonite clay, containing a mixture of Na+ and H+ within its interlayer, was prepared according to the procedure of Banin et al. (1985).

Ribonucleotide Activation and XMP Oligomerization

The ribonucleotides were either activated in situ or pre-activated. In situ activation of ribonucleotides, in which the imidazole activation of the ribonucleotides occurs directly in the same aqueous solution in which the oligomerization reaction occurs, is based upon modification of published protocols (Hermanson 2008). Solutions of 20 mM or 200 mM disodium XMP (X = A, G, C or U) ribonucleotides were made up in 10 mM 2-(N-morpholino)ethanesulfonic acid (MES), 150 mM NaCl buffer solution and the pH was adjusted to 5.5 or 6.5 using 0.1 M NaOH. 75 μL of this solution was added to RNAse-free plastic microcentrifuge tubes containing 1–3 mg 1-Ethyl-3-(3-dimethylaminopropyl) carbodiimide (EDC). In some experiments, the solution also contained 2.5–5.0 mg of activated montmorillonite clay. 25 μL of 0.1 M imidazole buffer at pH 6.0 (adjusted using 0.1 M HCl) was added to each vial. The vials were then vortexed for 10s, briefly centrifuged, and agitated for 3d. For reactions using pre-activated ribonucleotides, GMP was first converted from the disodium salt to the free acid by using hydrogen-saturated Dowex 50WX8 cation exchange resin. The free acid forms of monomeric AMP and GMP were then activated by adding imidazole to the 5′-phosphate (creating ImpX where X = A or G), as described in the literature (Pfeffer et al. 2005). The oligomerization reactions were performed in the same manner as with the in situ activated ribonucleotides, except that the addition of EDC was omitted and 25 μL of pH 5.5 or 6.5 buffer was used instead of the imidazole buffer.

MALDI-TOF MS Analysis

For Matrix Assisted Laser Desorption Ionization—Time-of-Flight Mass Spectrometry (MALDI-TOF MS) analysis, samples were immediately staged for analysis upon completion of the reactions. MALDI-TOF MS was performed using a Bruker Autoflex II instrument (Bruker Daltonics, Billerica, MA, USA). To reduce adduct formation, the samples were all centrifuged to sediment the clay and subjected to C18 zip tip (EMD Millipore, Darmstadt, Germany) desalting procedures (following the manufacture’s protocols). The samples were then plated on an AnchorChip var/387 steel target (Bruker Daltonics). The MALDI matrix was 2,4,6-trihydroxyacetophenone (246-THAP) with an ammonium citrate co-matrix (Castleberry et al. 2008; Burcar et al. 2013). All analyses were performed in the negative ion mode. The reaction solutions were analyzed in reflectron mode with the following settings: ion source 1, 19.00 kV; ion source 2, 16.85 kV; lens, 8.5 kV; reflector, 20.0 kV; system energy, 115.7 ± 5.0 μJ; pulsed ion ext., 80–400 ns. In all cases, relative laser power for desorption was 31–55 % for all targets. 500 laser shots were applied per location. The shots were summed and smoothed using the Savitsky Golay polynomial regression algorithm with a width of 0.2 m/z. Baseline subtraction was performed on all spectra. Prior to analysis, the mass spectrometer was calibrated using the low molecular weight oligonucleotide standards as external calibrants. For each reaction solution, the observed peaks were compared to the known m/z values of the 2-mer, 3-mer, 4-mer, and 5-mer of the linear oligomers of the ribonucleotides used in that reaction, and these peaks were used for an additional internal calibration as long as their m/z values were within 0.5 m/z. Once the peaks were calibrated, successive oligomers could be identified in the spectra by observing peaks that are consistent with known masses for the linear oligomer products.

Results and Discussion

Oligomerization experiments were first conducted using 15 mM XMP (X = A, C, U or G) or ImpX (X = A or G) at pH 6.5 or 5.5 in the presence or absence of montmorillonite clay (MMC). Results of MALDI-TOF MS analysis of the reaction products are summarized in Fig. 3. Oligomerization was observed for all of the in situ activated and pre-activated ribonucleotides at pH 6.5, while at pH 5.5 no oligomers were observed for either in situ activated GMP or pre-activated ImpG. Not surprisingly, at both pH values, longer oligomers were obtained in the presence of MMC with the exception of X = C, for which no appreciable difference was observed. The failure of either in situ activated GMP or pre-activated ImpG to form oligomers at pH 5.5 is likely due to the unique Hoogsteen hydrogen bonded, G-quadruplex structures that guanosine compounds can form in solution and that are promoted with decreasing pH (Lane et al. 2008).

Fig. 3
figure 3

Length of linear oligomers for in situ EDC activated and pre-activated 15 mM ribonucleotides run at pH 6.5 and 5.5, with and without montmorillonite clay (MMC)

Full size image

Taking a closer look at the reaction products, MALDI-TOF MS spectra are shown in Fig. 4 for in situ activated AMP and for ImpA at pH 6.5. In the absence of MMC (Fig. 4a and b, respectively, also shown in Fig. 5a and b with an expanded m/z axis zooming in on the low m/z range), oligomers as long as the 4-mer are detected for in situ activated AMP but only as long as the 3-mer for ImpA. The most striking feature of the in situ activated oligomerization products is the presence of a large number of additional peaks that are not observed for ImpA. These unidentified peaks are attributed to side products of the in situ activation reaction. In the presence of MMC (Fig. 4c and d, also shown in Fig. 6a and b with an expanded m/z axis zooming in on the high m/z range), oligomers as long as the 8-mer are detected for in situ activated AMP and as long as the 10-mer for ImpA. Again, many additional, unidentified peaks are detected for in situ activated AMP especially in the lower m/z region (Figs. 4c and 6a) that are not observed for ImpA (Figs. 4d and 6b). The relative intensities of the m/z peaks in the MALDI-TOF MS spectra corresponding to each linear oligonucleotide are less intense for the in situ activated ribonucleotides, implying lower concentrations of oligomers compared to the corresponding pre-activated ImpX reactions. This decrease in oligomerization efficiency is most likely due to the marked increase in side products observed in the MALDI-TOF MS spectra. These side reactions are likely caused by non-specific activation of other reactive sites on ribose or the nucleobases targeted by EDC in solution, or by reactions with isourea side products in solution.

Fig. 4
figure 4

MALDI-TOF MS spectra of oligomerization products from 15 mM ribonucleotide reactions at pH 6.5. (a) in situ EDC activated AMP (m/z values are 3-mer: 1004.68, 4-mer: 1333.97). (b) Pre-activated ImpA (m/z values are 3-mer: 1004.66). (c) in situ EDC activated AMP with montmorillonite catalyst (m/z values are 3-mer: 1004.68, 4-mer: 1333.83, 5-mer: 1663.14, 6-mer: 1992.31, 7-mer: 2321.10, 8-mer: 2650.62). (d) Pre-activated ImpA with montmorillonite catalyst (m/z values are 3-mer: 1004.68, 4-mer: 1333.83, 5-mer: 1663.13, 6-mer: 1992.36, 7-mer: 2321.48, 8-mer: 2651.21, 9-mer: 2980.36, 10-mer: 3310.22)

Full size image
Fig. 5
figure 5

Expanded view of low m/z region for pH 6.5, 15 mM (a) in situ EDC activated AMP and (b) ImpA reactions without montmorillonite catalyst. Note the increase in peaks attributed to side products for the EDC activated reaction

Full size image
Fig. 6
figure 6

Expanded view of high m/z region for pH 6.5, 15 mM (a) in situ EDC activated AMP and (b) ImpA reactions in the presence of montmorillonite clay. The prominent peaks immediately to the right of the linear oligomers in the ImpA reaction are attributed to aggregates of circular and linear oligomers in solution (Burcar et al. 2013)

Full size image

Experiments were then conducted on in situ EDC activation of ribonucleotides with an increased concentration of ribonucleotide or EDC in pH 6.5 solution. When XMP concentrations were increased from 15 to 150 mM, there was a dramatic increase in the peaks due to side products in the MALDI-TOF MS spectra. This is evident in Fig. 7, which shows the MALDI-TOF MS spectra for in situ activated AMP at both concentrations. Figure 8 summarizes the effects of increasing concentration on the oligomerization products for all of the investigated ribonucleotides. In the absence of MMC, oligomer length increases for in situ activated GMP and UMP while no change is observed for AMP and CMP. However, with the exception of CMP, there is an overall decrease in oligomer length upon increasing ribonucleotide concentration when the reactions are run in the presence of MMC. These results can be attributed to an increased concentration of non-activated ribonucleotides in the catalytic interlayer of the clay that could be limiting oligomer growth not only through chain termination, but also by preventing access to the catalytic surfaces through saturation of the binding sites. In addition, there would also be an increase in in situ activated ribonucleotides reacting with non-specifically activated ribonucleotides in solution, thereby generating larger concentrations of side products.

Fig. 7
figure 7

Comparison of peak intensity for non-linear oligomer side reactions for in situ EDC activation of reactions at pH 6.5 utilizing (a) 15 mM AMP and (b) 150 mM AMP

Full size image
Fig. 8
figure 8

Length of linear oligomers for in situ EDC activated ribonucleotides run at 15 mM and 150 mM concentrations at pH 6.5, with and without montmorillonite clay (MMC)

Full size image

Tripling the EDC concentration had no effect upon the oligomer length or the intensity of side reaction peaks in the MALDI-TOF MS spectra (data not shown). These results indicate that the concentration of ribonucleotides for in situ activation reactions is the limiting factor in determining the extent of side reactions and the extent and yield of oligomers at the concentrations and conditions of these experiments.

In conclusion, this work demonstrates that RNA oligomerization can be achieved in situ in aqueous solution using EDC activation of ribonucleotides, yielding oligonucleotides of comparable length to those obtained using ImpX ribonucleotides that were pre-activated in organic solvent. This important result supports the possibility of an RNA world by providing a more feasible pathway for prebiotic generation of highly-reactive ribonucleotides on early Earth than has previously been demonstrated. In addition, even at lower concentrations, the in situ activated ribonucleotides generated oligomers in comparable yields to its detectable side products. This is important from the perspective of plausible, local nucleotide concentrations on early Earth.