Questions and Answers Related to the Prebiotic Production of Oligonucleotide Sequences from 3’,5’ Cyclic Nucleotide Precur- Life Sors

Home , Nucleobase, Oligomer, Polynucleotide, Ribonucleoside

Review Questions and Answers Related to the Prebiotic Production of Oligonucleotide Sequences from 3’,5’ Cyclic Nucleotide Precur- life sors

Review Judit E. Šponer 1,*, Jiří Šponer 1, Aleš Kovařík 1, Ondrej Šedo 2, Zbyněk Zdráhal 2, Giovanna Costanzo 3 and Ernesto Di MauroQuestions 3 and Answers Related to the Prebiotic Production of Oligonucleotide Sequences from 30,50 Cyclic Nucleotide Precursors1 Institute of Biophysics of the Czech Academy of Sciences, Královopolská 135, 61265 Brno, Czech Republic; [email protected] (J.E.S.); [email protected] (J.S.); [email protected] (A.K.) Judit E. Šponer 1,* , Jiˇrí Šponer2 Central1, Aleš European Kovaˇrík 1 Institute, Ondrej of Technology, Šedo 2 , ZbynˇekZdr Masaryk University,áhal 2, Giovanna Kamenice Costanzo 5, 62500 3Brno, Czech Republic; and Ernesto Di Mauro 3 [email protected] (O.S.); [email protected] (Z.Z.) 3 Institute of Molecular Biology and Pathology, CNR, Piazzale A. Moro 5, 00185 Rome, Italy; giovan- [email protected] (G.C.); [email protected] (E.D.M.) 1 Institute of Biophysics of the Czech Academy of Sciences, Královopolská 135, 61265 Brno, Czech Republic; * [email protected]: [email protected] (J.Š.); [email protected] (A.K.) 2 Central European Institute of Technology, Masaryk University, Kamenice 5, 62500 Brno, Czech Republic; Abstract:[email protected] Template-free (O.Š.); nonenzymatic [email protected] polymerization (Z.Z.) of 3’,5’ cyclic nucleotides is an emerging 3 Institute of Molecular Biology and Pathology, CNR, Piazzale A. Moro 5, 00185 Rome, Italy; [email protected] of the origin of life (G.C.);research. [email protected] In the last ten years, (E.D.M.) a number of papers have been published *addressingCorrespondence: various [email protected] aspects of this process. These works evoked a vivid discussion among scientists working in the field of prebiotic chemistry. The aim of the current review is to answer the most Abstract: Template-free nonenzymatic polymerization of 30,50 cyclic nucleotides is an emerging topic offrequently the origin of raised life research. questions In the lastrelated ten years, to the a number detectio of papersn and have characterization been published addressingof oligomeric products as variouswell as aspects to the of geological this process. co Thesentext worksof this evoked chemistry. a vivid discussion among scientists working in the field of prebiotic chemistry. The aim of the current review is to answer the most frequently raisedKeywords: questions polymerization; related to the detection cyclic and nucleotides; characterization prebiotic of oligomeric chemistry products as well as to the geological context of this chemistry. Citation: Šponer, J.E.; Šponer, J.; Kovaˇrík,A.; Šedo, O.; Zdráhal, Z.; Citation:Costanzo, Šponer, G.; DiJ. E.; Mauro, Šponer, E. Questions J.; Ko- Keywords: polymerization; cyclic nucleotides; prebiotic chemistry vařík, andA. Life Answers 2021 Related, 11, x. to the Prebiotic https://doi.org/10.3390/xxxxxProduction of Oligonucleotide 1. Introduction Sequences from 30,50 Cyclic Nucleotide Precursors. Life 2021, 11, 1. IntroductionProduction of oligonucleotide sequences from nucleotide precursors in the absence Academic Editor: Michele Fiore And 800. https://doi.org/10.3390/ of enzymesProduction and of oligonucleotide template molecules sequences is from one nucleotide of the most precursors challenging in the absence problems of of modern Emiliano Altamura life11080800 enzymesorigin of and life template research. molecules 2’,3’ iscyclic one of nucleotides the most challenging (see Figure problems 1) have of modern long origin been considered as ofpotent life research. precursors 20,30 cyclic of the nucleotides most ancient (see Figure oligonucleotide1) have long been sequences considered formed as potent on our planet [1– Received:Academic 19 July Editors: 2021 Michele Fiore and precursors4]. This motivated of the most ancient numerous oligonucleotide attempts sequences dealing formedwith the on ourreconstruction planet [1–4]. This of their synthesis Accepted:Emiliano 03 Aug Altamuraust 2021 motivated numerous attempts dealing with the reconstruction of their synthesis under under plausible prebiotic conditions [5–8]. Nonetheless, nucleotides containing a phos- Published: date plausible prebiotic conditions [5–8]. Nonetheless, nucleotides containing a phosphodiester Received: 19 July 2021 linkagephodiester confined linkage in a strained confined 5-membered in a strained ring are 5-me unstablembered in an acidicring environmentare unstable and in an acidic envi- Accepted: 3 August 2021 undergo a quick ring-opening reaction on the timescale of minutes [9]. Publisher’sPublished: Note: 8 August MDPI 2021 stays neu- ronment and undergo a quick ring-opening reaction on the timescale of minutes [9]. tral with regard to jurisdictional claimsPublisher’s in published Note: mapsMDPI and stays institu- neutral with regard to jurisdictional claims in tional affiliations. published maps and institutional affiliations.

SubmittedCopyright: for poss© 2021ible byopen the access authors. publicationLicensee under MDPI, the Basel, terms Switzerland. and con- This article is an open access article ditions of the Creative Commons At- distributed under the terms and tribution (CC BY) license (https://cre- conditions of the Creative Commons ativecommons.org/licenses/by/4.0/).Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ FigureFigure 1. 1.Structural Structural formulas formulas of 30,5 0ofand 3’,5’ 20,3 0andcyclic 2’,3’ nucleotides. cyclic nucleotides. 4.0/).

Life 2021, 11, 800. https://doi.org/10.3390/life11080800 https://www.mdpi.com/journal/life

Life 2021, 11, x. https://doi.org/10.3390/xxxxx www.mdpi.com/journal/life Life 2021, 11, 800 2 of 13

In contrast, the 30,50 cyclic form possesses sufficient stability to withstand extreme conditions as conferred by low-pH environments, most likely ubiquitous on the early Earth [10,11]. Partly because of its significance played in modern biology (cGMP and cAMP are cofactors of key biochemical reactions [12]) and partly because of its unique aggregation properties, 30,50 cyclic guanosine monophosphate (hereafter 30,50 cGMP, the rest of the abbreviations used are explained at the end of the paper) was the target of a series of studies aimed at unraveling the process that could lead to the formation of the first oligonucleotide sequences on the early Earth [13–18]. The concept of polymerization studies using 30,50 cyclic nucleotides is strikingly different from that of preceding works published on prebiotic polymerization processes, which set a goal to find high-yielding chemical routes to oligonucleotides. These latter studies (for an overview see e.g., [19]) use highly activated monomers, because they enable achieving higher polymerization yields. Nevertheless, the lower stability of highly activated precursors (e.g., phosphorimidazolides, nucleotide triphosphates or 20,30 cyclic nucleotides) raises questions regarding their accumulation and long-term survival in the prebiotic pool. Polymerization studies targeting 30,50 cyclic nucleotides place a special emphasis on identifying the conditions that are compatible with a sustainable oligonucleotide production on longer time scales in a geochemical context relevant to the early Earth. This requires not only sufficiently stable monomers, but also a robust chemistry. As Albert Eschenmoser suggested [20], the robustness of a prebiotic synthetic network lies in the multimodality of catalytically accessible chemical pathways between reactants and final products, because it enables adaptation of the chemical system to the quickly changing chemical environment. In the current paper we provide a brief account of the available knowledge on the polymerization of 30,50 cyclic nucleotides. Since this research has been the focus of general attention for a long time, a number of questions and objections have been raised by the prebiotic chemistry community related to this topic. Our main goal with this paper is to provide an in-depth answer to the most frequent ones and to clarify those disputed points, which emerged after publishing the original studies.

2. How Could 30,50 Cyclic Nucleotides Accumulate in the Prebiotic Pool? Synthesis of 20,30 cyclic nucleotides [5–8] was demonstrated by a number of studies both in aqueous and non-aqueous media. In contrast, no attention has so far been devoted to the prebiotic synthesis of 30,50 cyclic nucleotides. Nevertheless, based on historical studies by Khorana et al., a simple potential prebiotic route can be outlined to these compounds: Reference [21] illustrates that in the presence of carbodiimides nucleoside 50-phosphates can be converted into the 30,50 cyclic form. Since carbodiimides form in large amounts upon thermal treatment of formamide in the presence of meteorites [22] and nucleoside-50-phosphates belong to the main products of phosphorylation reactions conducted in formamide [7], accumulation of 30,50 cyclic nucleotides seems to be plausible in a formamide-rich medium under prebiotic conditions. Work on this topic is in progress.

3. Mechanism of Ring-Opening Polymerization Reactions Involving 30,50 Cyclic Nucleotides Ring-opening polymerization reactions of cyclic phosphate and phosphonate esters in non-aqueous medium are well-known in the literature [23,24]. These reactions are commonly catalyzed by DBU suggesting that a base-catalyzed ring-opening polymerization operates in this case. In the ﬁrst reports on polymerization of 30,50 cGMP the highest reaction yields were observed at pH 9 achieved with Tris-HCl buffering suggesting a base-catalyzed reaction mechanism [14]. Addition of DBU also promoted the reaction [14]. Recent studies on the pH-dependence of the reaction conducted in dry state indicated that the reaction proceeds by an acid- rather than a base-catalyzed mechanism [25]. We note, that this observation does not contradict the original observation made in a pH = 9 Tris-HCl buffer, since a recent study showed that the pH of this buffer exhibits an extraordinarily strong Life 2021, 11, x FOR PEER REVIEW 3 of 14

that this observation does not contradict the original observation made in a pH = 9 Tris- HCl buffer, since a recent study showed that the pH of this buffer exhibits an extraordi- Life 2021, 11, 800 3 of 13 narily strong temperature dependence [26]. Thus, the actual pH of the Tris-HCl buffer could be 3–4 orders of magnitude lower at the temperature of the polymerization experiment (75–85 °C). At this temperature, a large part of the water present in the buffer (commonly only 15 μl is addedtemperature to the dry dependence material) evaporates [26]. Thus, and theactual condenses pH of in the the Tris-HCl tip of the buffer could be 3–4 orders of magnitude lower at the temperature of the polymerization experiment (75–85 ◦C). test tube, thus, the pH-shift towards acidic becomes even more pronounced, making the At this temperature, a large part of the water present in the buffer (commonly only 15 µl acid-catalyzed mechanismis added more toplausible the dry material)[25]. evaporates and condenses in the tip of the test tube, thus, A computational investigationthe pH-shift towardshas shown acidic that becomes a ladder-like even more stacked pronounced, supramolecular making the acid-catalyzed architecture may host bothmechanism the base- more as well plausible as the [25 acid-catalyzed]. reaction routes (see Fig- ure 2a) [16]. The only differenceA computational between the investigation two pathways has is shown that while that a ladder-likein the base-cata- stacked supramolecular lyzed reaction pathwayarchitecture the nucleophile may host is ac bothtivated the base-by deprotonation as well as the of acid-catalyzed the ribose O3’ reaction routes (see position, in the acid-catalyzedFigure2 pathwaya) [ 16]. The the only phosphate difference of the between substrat thee two is activated pathways by is thatpro- while in the base- tonation (see Figure 3). Letcatalyzed us note reaction that use pathway of both the activation nucleophile strategies is activated is widespread by deprotonation in of the ribose O30 position, in the acid-catalyzed pathway the phosphate of the substrate is activated by organic chemistry to facilitate addition-elimination-type substitution reactions [27]. protonation (see Figure3). Let us note that use of both activation strategies is widespread in organic chemistry to facilitate addition-elimination-type substitution reactions [27].

(a)

(b)

Figure 2. Cont.

Life 2021, 11, x FOR PEER REVIEW 4 of 14

Life 2021, 11, 800 4 of 13

Figure 2. Theoretically predictedFigureFigure 2. optimum Theoretically2. Theoretically model predicted for predicted the optimum transphosphorylation optimum model for themodel transphosphorylation of for 3’,5’ the cGMP transphosphorylation [16] of 30,50 cGMP [16](a of) 3’,5’ cGMP [16] (a)as well as crystal structuresas(a well)as of well as3’,5’ crystal cGMPas crystal structures [28] structures(b) ofand 30,5 3’,5’0 cGMP of cAMP 3’,5’ [28 ](cGMP[29]b) and(c). [28] Color 30,50 (cAMPb coding:) and [29 3’,5’ ](C—grey;c). cAMP Color coding: [29] (c). C—grey; Color coding: C—grey; O—red; N—blue; P—orange;O—red;O—red; H—white. N—blue; N—blue; P—orange; P—orange; H—white. H—white.

Figure 3. Mechanistic models for the chain-extension step of the ring-opening polymerization of 3’,5’ cGMP. G‡ refers to the computed activation free energy of the reaction step [16,25]. FigureFigure 3. Mechanistic3. Mechanistic models models for the chain-extension for the chain-extension step of the ring-opening step of polymerizationthe ring-openi ofng 30,5 polymerization0 of ‡ ‡ Amazingly, the stackedcGMP.3’,5’ cGMP. Gstructurerefers G to therefersof computed3’,5’ to thecGMP activationcomputed predicted free activation energy by state-of-the of the free reaction energy art step ofelec- [16 the,25 ].reaction step [16,25]. tronic structure (QM) calculationsAmazingly, is essentially the stacked identical structure ofto 3 0the,50 cGMPstacking predicted pattern by seen state-of-the in art electronic the X-ray structure (seestructure FigureAmazingly, 2b), (QM) reflecting calculations the the stacked is unique essentially structurecapability identical ofof to 3’,5’3’,5’ the stacking cGMPcGMP patternto predicted self- seen in by the state-of-the X-ray art elec- associate by stacking. Thestructuretronic electronic structure (see Figure structure 2(QM)b), reﬂecting calculations the unique in is addition capabilityessentially ofwithout 3 identical0,50 cGMP doubt toto self-associate the stacking by pattern seen in prove that this robust ground-statestacking.the X-ray The structure stacking electronic arrangement structure(see Figure calculations is2b), readily reflecting in additioncapable the withoutto unique support doubt capability a prove that of this3’,5’ cGMP to self- reactive geometry for therobustassociate oligomerization ground-state by stacking. stacking process. arrangementThe In electronicview of is readilythis structure we capable conclude tocalculations support that all a reactive in addition geometry without doubt for the oligomerization process. In view of this we conclude that all claims that 30,50 cGMP claims that 3’,5’ cGMP cannotprove adopt that thisa reactive robust conformation, ground-state etc., stacking should arrangementbe dismissed as is readily capable to support a cannot adopt a reactive conformation, etc., should be dismissed as scientiﬁcally unfounded. scientifically unfounded. Due to its stacking properties 3’,5’ cGMP can clearly approach a Duereactive to its stacking geometry properties for the 30,5 oligomerization0 cGMP can clearly approachprocess. a reactiveIn view conformation of this we for conclude that all reactive conformation forpolymerizationclaims polymerization that 3’,5’ on its cGMP on own its and cannotown does and notadopt does need anot helpreactive need from help anyconformation, otherfrom midwife any etc., molecules should or be dismissed as other midwife moleculesactivatingscientifically or activating agents. unfounded.agents. The ability The ofability Due biopolymers to of its biopolymers stacking and their properties and monomers their mon- 3’,5’ to self-assemble cGMP can into clearly approach a omers to self-assemble into reactive geometries could play a pivotal role at the emergence reactivereactive geometries conformation could play for a pivotalpolymerization role at the emergence on its own of life and on ourdoes planet not [30need]. help from any of life on our planet [30]. Our computations showother that midwife both mechanis moleculesms orhave activating similar activation agents. The energies ability and of biopolymers and their mon- thermodynamics [16,25].omers Thus, bothto self-assemble mechanisms intoare chemically reactive geometries plausible per could see. Never- play a pivotal role at the emergence theless, the acid-catalyzedof pathwaylife on our seems planet to be [30]. more likely in a prebiotic context. This is because the reaction requiresOur formation computations of a well-defined show that supramolecular both mechanis architecturems have similar activation energies and prior to the transphosphorylationthermodynamics reactions [16,25]. which is Thus, more botheasily mechanisms achievable under are chemically acidic plausible per see. Never- conditions. theless, the acid-catalyzed pathway seems to be more likely in a prebiotic context. This is because the reaction requires formation of a well-defined supramolecular architecture prior to the transphosphorylation reactions which is more easily achievable under acidic conditions.

Life 2021, 11, 800 5 of 13

Our computations show that both mechanisms have similar activation energies and thermodynamics [16,25]. Thus, both mechanisms are chemically plausible per see. Nev- ertheless, the acid-catalyzed pathway seems to be more likely in a prebiotic context. This is because the reaction requires formation of a well-deﬁned supramolecular architecture prior to the transphosphorylation reactions which is more easily achievable under acidic conditions. It has been found that the crystal structure [28,31] of free acid- and Na-form cGMP pro- vides optimum steric conditions for the transphosphorylation reactions necessary to form intermolecular phosphodiester linkages between the cyclic monomers (see Figure2b )[16]. Notably, the crystal structure of cAMP [29] (Figure2c) is less suited to support the polymerization chemistry. In line with this, only short (3–4 nt long) oligomers form from 30,50 cAMP in contrast to 30,50 cGMP, which polymerizes to a length up to 15–20 nucleotides. In addition, polymerization of 30,50 cAMP is considerably slower as compared to that of 30,50 cGMP [32]. While the computed thermodynamic and kinetic parameters are very similar for the acid- and base-catalyzed mechanisms, crystallization of the cGMP monomers under acidic and basic conditions is strikingly different [18,25]. While the H-form (i.e., acidic) monomers crystallize on the timescale of minutes, crystallization of the Na-form (i.e., basic salt) material requires several days [31] due to the apparent electrostatic repulsion between the negatively charged cGMP- monomers. Faster crystallization of the acid-form cGMP molecules enables faster formation of the stacked supramolecular architecture that is the prerequisite of the polymerization reaction. This is compatible only with a low pH environment and an acid-catalyzed mechanism [25].

4. Is Detection of Oligonucleotides Formed from 30,50 cGMP by Denaturing Gel Electrophoresis Biased by Noncovalent Adducts? A number of methods have been used for detection of the various products formed in the oligomerization of 30,50 cGMP. Among them, electrophoresis on denaturing gels has been the most frequently applied technique mainly because of its high sensitivity. The method was used in combination with radioactive [13,14,16–18,32] as well as ﬂuorescent labeling [15,25] (for a recent detailed description of methodological details see [18,25]). The radioactive labeling has often been criticized because of the use of enzymes and the necessity of a precipitation step that is used to remove the excess 32P-ATP prior to loading the labeled material on the gel. Potentially, this could lead to the formation of non- covalent aggregates. This possibility has been tested with a simple experiment, in which guanosine has been treated in the same way as 30,50 cGMP [18]. Aggregation properties of H-form 30,50 cGMP are very similar to those of guanosine. Thus, it is reasonable to expect, that if aggregates formed from guanosine withstand the precipitation and labeling procedure, they would have to produce multiple band structures on the denaturing gel, similar to those of the oligomerization products. As Figure4 illustrates, this has not been observed. Further, this experiment excludes the possibility that the polynucleotide kinase used for phosphorylation could induce oligomer formation. Denaturation PAGE is standardly run in the presence of 50% urea at elevated temperatures which effectively disrupts intermolecular bonds. Together, detection of non-covalent aggregates can be excluded. The fact that the oligomers are efﬁciently labeled in a kinase reaction suggests that a considerable portion of the polymerization products contains free 50-OH groups. LifeLife 20212021, 11, 11, x, 800FOR PEER REVIEW 6 of 13 6 of 14

FigureFigure 4. 4.Control Control experiments experiments with variouslywith variously treated 3treated0,50 cGMP 3’,5’ as wellcGMP as withas well guanosine as with and guanosine and guanosineguanosine 50-triphosphate 5′-triphosphate (GTP) (GTP) adopted adopted from [18 ].from Lane [18]. 1: 150 Laneµl 1 mM 1: 150 cGMP-H μl 1 freeze-driedmM cGMP-H in an freeze-dried in Eppendorfan Eppendorf tube (for tube ~1 (for h) without ~1 h) with subsequentout subsequent thermal treatment thermal at treatment elevated temperatures. at elevated Lane temperatures. 2: Lane 1502: 150µl 1 mMμl 1 cGMP-HmM cGMP-H freeze-dried freeze-dried in an Eppendorf in an Eppendorf tube (for ~1 tu h)be and (for treated ~1 h) atand 80 treated◦C for 5 at min. 80 °C for 5 min. LanesLanes 3 and 3 and 4: the 4: samethe same as lane as 2, exceptlane 2, that except the thermal that th treatmente thermal at 80treatment◦C was performed at 80 °C for was 15 andperformed for 15 60and min, 60 respectively. min, respectively. Lane 5: 66 Lane× 3 µ 5:l of66 1 × mM 3 μ cGMP-Hl of 1 mM solution cGMP-H deposited solution in a deposited dropwise manner in a dropwise man- onner the on bottom the bottom of a preheated of a preheated (80 ◦C) borosilicate (80 °C) beaker.borosili Thecate next beaker. drop was The always next depositeddrop was on always the deposited ◦ sameon the place same after place drying after of the drying previous of one. the Theprevious dry material one. The was treateddry material at 80 C was for additional treated at 5 h.80 °C for addi- Lanetional 6: 65× h.7.5 Laneµl of 6: 1 6 mM × 7.5 cGMP-H μl of 1 solution mM cGMP-H was deposited solution in a was dropwise deposited manner in into a dropwise 5 spots on manner into 5 ◦ thespots bottom on the of a bottom preheated of (80 a preheatedC) borosilicate (80 °C) beaker. borosili Thecate next dropbeaker. was The always next deposited drop was on always the deposited ◦ sameon the place same after place drying after of the drying previous of the one. previous The dry material one. The was dry treated material at 100 wasC fortreated additional at 100 °C for addi- ◦ 5tional h. Note 5 thath. Note the optimumthat the optimum temperature temperature for the polymerization for the polymerization reaction is ~75–85 reactionC, according is ~75–85 °C, accord- toing [14 ]to and [14] [16 and], i.e., [16], this experimenti.e., this experiment was performed was outside performed the optimum outside reaction the optimum temperature reaction range. temperature Lane 7: 30 × 6.6 µl of 1 mM cGMP-H solution deposited in a dropwise manner on the bottom of a range. Lane◦ 7: 30 × 6.6 μl of 1 mM cGMP-H solution deposited in a dropwise manner on the bottom preheatedof a preheated (80 C) borosilicate(80 °C) borosilicate beaker. The beaker. next drop The was ne alwaysxt drop deposited was always on the deposited same place afteron the same place drying of the previous one. The dry material was treated at 80 ◦C for additional 5 h. Lane 8: 150 µl of after drying of the previous one. The dry material was treated at 80 °C for additional 5 h. Lane 8: 2 mM (~300 ng) G3 oligomer (from Biomers) was freeze-dried in an Eppendorf tube, followed by a 150 μl of 2 mM (~ 300 ng) G3 oligomer (from Biomers) was freeze-dried in an Eppendorf tube, fol- heat-treatment at 80 ◦C for 3 h. Lane 9 (almost empty): 150 µl of 1 mM guanosine freeze-dried in an lowed by a heat-treatment at 80 °C for 3 h. Lane 9 (almost empty): 150 μl of 1 mM guanosine freeze- Eppendorf tube and subsequently treated at 80 ◦C for 5 h. Since guanosine has the same aggregation dried in an Eppendorf tube and subsequently treated at 80 °C for 5 h. Since guanosine has the same properties as 30,50 cGMP-H, absence of oligomerization products shows that the kinase used for aggregation properties as 3’,5’ cGMP-H, absence of oligomerization products shows that the kinase phosphorylation does not recognize non-covalent aggregates of monomers. Lane 10: 150 µl of 1 mM GTPused prepared for phosphorylation and treated as the does guanosine not recognize sample (diffuse non-covalent smeared aggregates bands towards of themonomers. bottom are Lane 10: 150 μl likelyof 1 causedmM GTP by the prepared highly charged and treated character as of the the guanos molecule).ine Samples sample were (diffuse precipitated smeared using bands the towards the procedurebottom are given likely in [18 caused]. M1 = G3by (Biomers)the highly marker. charged M2 = ch G9aracter (Biomers) of marker.the molecule). Samples were precipitated using the procedure given in [18]. M1 = G3 (Biomers) marker. M2 = G9 (Biomers) marker. 5. Is Mass Spectrometric Detection of Oligonucleotides Formed from Cyclic Nucleotide Precursors Biased by Non-Covalent Adducts? 5. Is Mass Spectrometric Detection of Oligonucleotides Formed from Cyclic Nucleo- tideA Precursors frequently used Biased methodology by Non-Covalent for oligonucleotide Adducts? detection is MALDI-ToF mass spectrometry. The approach was used for detection of oligonucleotides formed from cyclic nucleotideA frequently precursors asused well. methodology Nevertheless, soonfor olig afteronucleotide its ﬁrst use it detection has been shown is MALDI-ToF that mass spectrometry. The approach was used for detection of oligonucleotides formed from cyclic nucleotide precursors as well. Nevertheless, soon after its first use it has been shown that this method is often biased by non-covalent adduct formation, since the molar weight of adducts consisting of n cyclic monomers equals to that of an n-mer oligomer having a 2’,3’ cyclic end [33]. Likewise, molecular weight of an n-mer oligomer is the same as that of n-monomers plus a water molecule.

Life 2021, 11, 800 7 of 13

this method is often biased by non-covalent adduct formation, since the molar weight of adducts consisting of n cyclic monomers equals to that of an n-mer oligomer having a 20,30 cyclic end [33]. Likewise, molecular weight of an n-mer oligomer is the same as that of n-monomers plus a water molecule. Studying the fragmentation spectra of the species detected with MALDI MS could be one of the potential remedies for this problem. For example, inspection of the polymerization products formed from 30,50 cAMP allowed a successful identiﬁcation of pApA dimers. The fragmentation spectra clearly revealed signals corresponding to pA as well as pAp fragments, which exhibited roughly the same intensities as those found in the fragmentation spectra of pApA standards [32]. Unfortunately, the same method cannot be used for identifying pGpG products, since intensity of the pG signal, which also corresponds to the hydrolyzed monomer, is an order of magnitude higher than that of the more important pGp fragment. In our studies dealing with the polymerization of 30,50 cCMP [34], we used ESI-MS detection, which (at least for this monomer) was less biased by formation of non-covalent adducts. In particular, in the spectrum of the heat-treated sample, we did not observe the peak shifted by +18 Da, corresponding to the adduct of the monomer with water or hydrolyzed monomer, whereas the signal corresponding to the pCpC covalent product was observed. Fragmentation of this signal produced the expected pCp and pC species, verifying the covalent character of the pCpC dimer. To assess the extent of non-covalent adduct formation in the MALDI-ToF MS and ESI-MS spectra of polymerization products formed from guanine-containing monomers, we investigated the spectra of guanosine, a molecule with remarkably similar aggregation properties as that of 30,50 cGMP. Figure5 shows that in contrast to the ESI-MS detection of oligoC sequences discussed in [34], non-covalent guanosine aggregates up to tetramer are observed both by MALDI-ToF as well as by ESI mass spectrometry. This is likely caused by the outstandingly high propensity of guanine-derivatives to form stacked aggregates independently on the medium. On the other hand, as expected, in lack of covalent oligomer Life 2021, 11, x FOR PEER REVIEW formation, there is almost no difference between the spectra of the untreated and8 of heat- 14

treated (conducted at 80 ◦C for 5 h in dry form) materials.

Figure 5. Cont.

Figure 5. MALDI-ToF MS (measured in reflectron positive detection mode, panel A) and ESI-MS (measured in negative ion mode, panel B) spectra of untreated and heat-treated guanosine samples. The heat treatment was performed in dry state at 80 °C for 5 hours. Signals of guanosine and its adducts are indicated with an arrow.

6. What Kind of Other Detection Methods Support Oligonucleotide Formation from 3’,5’ cGMP? In addition to detection by electrophoresis and mass spectrometry, the covalent nature of the oligomers formed in the polymerization of 3’,5’ cGMP has been studied by a number of other methods (for a summary of methods applied for characterization of products formed from 3’,5’ cyclic nucleotides, see Table 1).

Table 1. Summary of the methodologies used for detection of oligonucleotides formed from 3’,5’ cyclic nucleotide precursors.

Method Monomer Reference Denaturing gel-electrophoresis with 32P-labelling 3‘,5‘ cGMP [13,14,16–18]

Life 2021, 11, x FOR PEER REVIEW 8 of 14

Life 2021, 11, 800 8 of 13

Figure 5. MALDI-ToF MS (measuredFigure 5. inMALDI-ToF reﬂectron positiveMS (measured detection in reflectron mode, panel posi (tiveA)) anddetection ESI-MS mode, (measured panel A) in and negative ESI-MS (measured in negative ion mode, panel B) spectra of untreated and heat-treated guanosine samples. ion mode, panel (B)) spectra of untreated and heat-treated guanosine samples. The heat treatment was performed in dry ◦ The heat treatment was performed in dry state at 80 °C for 5 hours. Signals of guanosine and its state at 80 C for 5 h. Signalsadducts of guanosine are indicated and its adductswith an arrow. are indicated with an arrow.

6.6. WhatWhat KindKind ofof Other Detection Method Methodss Support Support Oligonucleotide Oligonucleotide Formation Formation from from 33’,5’0,50 cGMP?cGMP? InIn addition addition toto detection detection byby electrophoresiselectrophoresis andand massmass spectrometry,spectrometry, the the covalent covalent nature nature of the oligomers formed in the polymerization of 3’,5’ cGMP has been studied by a of the oligomers formed in the polymerization of 30,50 cGMP has been studied by a number number of other methods (for a summary of methods applied for characterization of prod- of other methods (for a summary of methods applied for characterization of products ucts formed from 3’,5’ cyclic nucleotides, see Table 1). formed from 30,50 cyclic nucleotides, see Table1).

Table 1. Summary of the methodologies used for detection of oligonucleotides formed from 3’,5’0 0 Tablecyclic1. nucleotideSummary precursors. of the methodologies used for detection of oligonucleotides formed from 3 ,5 cyclic nucleotide precursors. Method Monomer Reference Denaturing gel-electrophoresisMethod with 32P-label- Monomer References ling 3‘,5‘ cGMP [13,14,16–18][13,14,16–18 ] Denaturing gel-electrophoresis with 30,50 cGMP [32] 32P-labelling 30,50 cAMP

Denaturing gel-electrophoresis with 30,50 cGMP [15,25] intercalating ﬂuorescent labelling 30,50 cGMP [14–16] MALDI-ToF 30,50 cAMP [32]* 30,50 cCMP [34] ESI-MS 30,50 cCMP [34]* HPLC-ESI-MS 30,50 cGMP [25] 31P-NMR 30,50 cGMP [14] Infrared nanospectroscopy 30,50 cGMP [17] * Including MS/MS fragmentation analysis.

In [14] 31P-NMR spectroscopy showed that upon heat treatment of 30,50 cGMP new signals appear in the 0 to −2.6 chemical shift range, characteristic to RNA oligonucleotides. Since the signal of phosphorus of acyclic or 20,30 cyclic monomers falls into the positive chemical shift range, the new signals resulting from sample treatment have been attributed to RNA-oligomers formed from 30,50 cyclic precursors. Life 2021, 11, 800 9 of 13

Infrared nanospectroscopy has been used to follow the change of the IR spectrum of a 30,50 cGMP sample deposited on gold surface upon heat-treatment at 80 ◦C[17]. The analysis has shown that a new band appears at 970 cm−1, i.e., in the ﬁngerprint region characteristic to RNA-backbones, in the spectrum of the heat-treated material. A recent study combines HPLC-separation and mass spectrometry to identify oligonucleotide products formed when polymerizing 30,50 cGMP [25]. This investigation has shown that besides non-covalent aggregates a noticeable amount of covalently bound oligomers is formed. In summary, although the individual experimental methods used to monitor the oligomerization of 30,50 cNMP compounds are complicated by some uncertainties, on aggregate, the 30,50 cNMP polymerization reactions have been characterized by multiple independent methods and are thus more thoroughly scrutinized than many other oligomerization processes discussed in contemporary prebiotic chemistry literature.

7. Does the Oligomerization Reaction of 30,50 cGMP Proceed in the Presence of Water? The reaction was studied under a multitude of experimental conditions. Its temperature optimum was found to be ca. 80 ◦C[13,14,16]. It was found that at least one drying step is necessary to get oligonucleotides longer than trimers [15,17]. Oligomer formation is regularly observed after ca. 1 h of reaction time. The reaction yields exhibit an oscillatory behavior on longer (several days) time scales, regardless of whether the reaction is conducted in dry state or in a nearly saturated solution [18]. The oscillatory behavior of the reaction has pointed at the importance of phase-separation processes in this chemistry. Nonetheless, oligomer (>trimer) formation is detectable on timescales of several weeks, assumed that at least partial drying of the monomers occurs in the course of sample treatment. Thus, in summary, our investigations show that the presence of water in the reaction mixture is not poisonous for the polymer formation if part of the material is at least temporarily present in dry form during the sample treatment. Nonetheless, only short (≤3 nt long) oligomers form if drying of monomers is prevented [25,35].

8. Technical Requirements of the Oligomerization Experiment When studying the polymerization of 30,50 cGMP, it was found that the drying process itself induces oligomer formation. Thus, conducting the polymerization experiment in a straightforward manner requires the use of a specially puriﬁed material that was never precipitated or dried during the sample preparation procedure. Using material purchased in dry form, due to its oligonucleotide content, produces false positive results. When working with minerals, the specimens must be thoroughly cleaned by using H2O2, ethanol to remove their organic material content, which may degrade the oligomers formed in the polymerization reaction.

9. How Much is the Polymerization of 30,50 cGMP Sensitive to the Presence of Cations? May It Be Relevant in a Geological Context? Earlier reports interpreted the absence of polymer formation from an Na-form salt of 30,50-cGMP by the fact that cations block the anionic ring-opening polymerization mechanism [16]. A recent study revisited this issue, and showed that presence of cations in the reaction mixture is not as restrictive as thought before [25]. As long as the polymerization is performed in an acidic environment, i.e., the monomers are neutral molecules, cations start inhibiting the reaction at very large (ca. 50×) excess. This means that basic pH rather than cations exert an inhibitory effect on the reaction [25]. The argument that the polymerization reaction is sensitive to cations logically suggested that such chemistry might be irrelevant to a real prebiotic environment, where free cations were likely available in large amounts. To project laboratory experiments into more geologically plausible models, we recently tried deposition of the cGMP monomers from a dropping solution on various heated mineral surfaces relevant to an early Earth Life 2021, 11, 800 10 of 13

scenario. The minerals chosen for our experiments were all included in Hazen’s “preliminary species list” [36]. We found that minerals, like amorphous and crystalline silica forms, mica, amphibole and other silicates, like andalusite, may host the reaction. On the other hand, minerals that chemically react with the acid-form of 30,50 cGMP leading to its deprotonation (like carbonate- and serpentine-group minerals or olivine) are incompatible with the chemistry [18]. This way the free acid form of 30,50 cGMP is converted into a salt-form material, which does not crystallize on the time scales of the experiment. The lack of crystal order in the dry deposited material then interferes with the oligonucleotide Life 2021, 11, x FOR PEER REVIEW 11 of 14 formation (see Figure6). In summary, the ions per se do not prevent the oligomerization process.

Figure 6. 30,50 cGMP crystallizesFigure and 6. 3’,5’ polymerizes cGMP crystallizes on muscovite and micapolymerizes (M). No crystallizationon muscovite mica and polymer(M). No formationcrystallization are and observed on calcite (C). Frompolymer left to right: formation electrophoretic are observed analysis on calcite of the polymerization(C). From left productsto right: electrophoretic formed by heat-treatment analysis of the polymerization products formed by heat-treatment (80 °C, 5 h) of the dry material deposited in a (80 ◦C, 5 h) of the dry material deposited in a dropwise manner on the mineral surface; densitometric analysis of the dropwise manner on the mineral surface; densitometric analysis of the electrophoretic images; elec- electrophoretic images; electron microscopic images of the dry materials deposited on the mineral surfaces. Adopted tron microscopic images of the dry materials deposited on the mineral surfaces. Adopted from [18]. from [18]. 10. Does Depurination Hamper Formation of Short Oligonucleotide Sequences from 3’,5’10. Does cGMP? Depurination Hamper Formation of Short Oligonucleotide Sequences from 30,50 HighercGMP? propensity of guanine-containing nucleotides to depurinate in acidic conditions Higheris well propensityknown [37]. of Nevertheless, guanine-containing depurination nucleotides reactions to depurinate also require in the acidic presence condi- oftions water is well in the known reaction [37]. Nevertheless,mixture. Under depurination drying conditions, reactions where also require oligonucleotide the presence for- of mationwater in has the been reaction demonstrated, mixture. Under the depurination drying conditions, process where is not oligonucleotide significant and formation is unable tohas outcompete been demonstrated, the polymerization the depurination reaction. process It does is not signiﬁcantinterfere with and the is unable oligomerization. to outcom- Ourpete recent the polymerization mass spectrometric reaction. investigation It does not interfere(using both with ESI- the oligomerization.and MALDI-ToF OurMS recentmeth- ods)mass focused spectrometric on the investigationlow molecular (using weight both fraction ESI- and products MALDI-ToF present MS in methods)the reaction focused mix- tureon the and low has molecular shown that weight sample fraction treatment products in dry present form at in 80 the °C reaction for 5 h mixtureresults only and in has a negligibleshown that depurination sample treatment of the in cyclic dry form monome at 80 ◦rsC [18]. for 5 hIn results contrast, only under in a negligible dry-wet cycling depuri- conditionsnation of the used cyclic by monomersRajamani et [ 18al.]. [38], In contrast, half-life underof acyclic dry-wet 5’ AMP cycling was conditionsestimated to used be 6.5 by hRajamani only. This et study al. [38 ],suggests half-life that of acyclic repeated 50 AMP dry-wet was cycles estimated may tointerfere be 6.5 hwith only. formation This study of oligonucleotidessuggests that repeated from purine dry-wet nucleotide cycles may precursors. interfere with formation of oligonucleotides fromWe purine note, nucleotide that depurinated precursors. short oligonucleotide sequences have been detected amongWe the note, polymerization that depurinated produc shortts oligonucleotideby gel-electrophoresis sequences [18]. have Their been origin detected is not amongneces- sarilythe polymerization associated with products the polymerization by gel-electrophoresis process, since [18]. they Their could origin be is the not result necessarily of the sampleassociated treatment with the required polymerization by the radioactive process, sinceend labeling they could and bethe the denaturing result of thegel electro- sample phoresis.treatment Signals required of bydepurinated the radioactive oligomer end labelingproducts and were the not denaturing observed gelin the electrophoresis. MALDI-ToF MSSignals analysis of depurinated presented oligomerin [16]. products were not observed in the MALDI-ToF MS analysis presented in [16]. 11. Conclusions: What are the Potential Reasons for the Erratic Reproducibility of 3’,5’ cGMP Polymerization Experiments? Lane et al. [39] rightfully designated reproducibility of the 3’,5’ cNMP oligomerization reactions as “erratic”. This means that the reaction requires suitable physical and chemical conditions to proceed, and if these conditions are not fulfilled, the reaction is inhibited, which can lead to the erroneous claim that the experiments could not be reproduced. However, erratic behaviors of chemical processes are not uncommon (on example could be crystallization processes which often require many empirical trial and error experiments to find the right condition) and have nothing to do with reproducibility. In fact, some degree of erratic behavior may be advantageous in chemical evolution, as it indicates the reaction is variable and sensitive to external conditions. Therefore, it can be regulated and optimized by external factors, and coupled with other reactions. In other words, while we certainly found some suitable conditions upon which the 3’,5’ cGMP

Life 2021, 11, 800 11 of 13

11. Conclusions: What are the Potential Reasons for the Erratic Reproducibility of 30,50 cGMP Polymerization Experiments? Lane et al. [39] rightfully designated reproducibility of the 30,50 cNMP oligomerization reactions as “erratic”. This means that the reaction requires suitable physical and chemical conditions to proceed, and if these conditions are not fulfilled, the reaction is inhibited, which can lead to the erroneous claim that the experiments could not be reproduced. However, erratic behaviors of chemical processes are not uncommon (on example could be crystallization processes which often require many empirical trial and error experiments to find the right condition) and have nothing to do with reproducibility. In fact, some degree of erratic behavior may be advantageous in chemical evolution, as it indicates the reaction is variable and sensitive to external conditions. Therefore, it can be regulated and optimized by external factors, and coupled with other reactions. In other words, while we certainly found some suitable conditions upon which the 30,50 cGMP oligomerization occurs, we almost certainly did not find the optimal conditions. Importantly, this polymerization reaction does not seem to need a sophistically tailored chemical environment with other supportive chemical species; it appears to be responsive to straightforward changes in common physical-chemistry conditions. Apart from trivial reasons (like not ensuring conditions in which relatively fragile short RNA sequences are capable of surviving; or using a salt form of the cyclic nucleotide monomer), over the years we identified several other reasons that might potentially hamper detection of oligonucleotides formed from 30,50 cGMP in specific experimental conditions. We showed that the reaction itself exhibits an oscillatory behavior on longer time scales [17]. This feature originates in the highly non-equilibrium nature of this chemistry and manifests itself as the strong time dependence of the reaction yields. For this reason, if the reaction time is improperly selected for a given experiment, only low amounts of short (≤3 nt long) oligomers may be detected. Thus, it is advisable to study the reaction on longer time scales at several time points for a successful detection. Another very important feature of the reaction is that at least one drying step is needed for a continued oligonucleotide production. This means that to a lesser extent, longer (>3 nt) oligonucleotides do form even in the presence of water, if the experimental setup allows for partial drying of the sample. In our previous studies we showed that this can be achieved by rewetting of the dry material, or by partial drying of the sample during the heat treatment [17,25]. The method used for drying may also influence the outcome of detection for two reasons. Prolonged (several hours long) drying in a low-performance vacuum evaporator may cause depurination of the nucleotide monomers. Further, too fast drying at elevated temperatures may occasionally lead to the formation of a glassy, non-crystalline material with randomly placed monomers even when using an H-form material. Obviously, in this case the lack of a properly organized supramolecular architecture interferes with the oligonucleotide formation. Many doubts about the reproducibility of the cGMP oligomerization reaction may stem from variability between experiments leading to different amounts and quality of crystalline materials.

Author Contributions: All authors contributed to writing—original draft preparation. All authors have read and agreed to the published version of the manuscript. Funding: This research was funded by the Czech Science Foundation (grant 19-03442S). G.C. and E.D.M. were supported by the Italian Space Agency (ASI) project N. 2019-3-U.0, “Vita nello spazio—Origine, Presenza, Persistenza della vita nello Spazio, dalle molecole agli estremofili” (Space life—OPPS). This work was supported by the project SYMBIT reg. number: CZ.02.1.01/0.0/0.0/ 15_003/0000477 financed by the ERDF. CIISB, Instruct-CZ Centre of Instruct-ERIC EU consortium, funded by MEYS CR infrastructure project LM2018127, is gratefully acknowledged for the financial support of the measurements at the CEITEC Proteomics Core Facility. Institutional Review Board Statement: Not applicable. Informed Consent Statement: Not applicable. Life 2021, 11, 800 12 of 13

Data Availability Statement: Data reported in this paper can be obtained from the authors upon request. Acknowledgments: The authors thank Dieter Braun for his comments on the manuscript. Conﬂicts of Interest: The authors declare no conﬂict of interest.

Abbreviations

30,50 cGMP 30,50 cyclic guanosine monophosphate 30,50 cGMP-H free adic form of 30,50 cyclic guanosine monophosphate 30,50 cAMP 30,50 cyclic adenosine monophosphate 30,50 cCMP 30,50 cyclic cytosine monophosphate 50 AMP adenosine 50 monophosphate 30,50 cNMP 30,50 cyclic nucleotide nt nucleotide pG 50 guanylic acid pGpG 50 diguanylic acid pGp 50 guanylic acid 30 phosphate pC 50 cytidylic acid pCpC 50 dicytidylic acid pCp 50 cytidylic acid 30 phosphate DBU 1,8-diazabicyclo(5.4.0)undec-7-ene MALDI matrix-assisted laser desorption/ionization ToF time of ﬂight ESI electrospray ionization MS mass spectrometry HPLC high-performance liquid chromatography

References 1. Verlander, M.S.; Lohrmann, R.; Orgel, L.E. Catalysts for self-polymerization of adenosine cyclic 2', 3'-phosphate. J. Mol. Evol. 1973, 2, 303–316. [CrossRef] 2. Usher, D.A.; Yee, D. Geometry of the dry-state oligomerization of 2',3'-cyclic phosphates. J. Mol. Evol. 1979, 13, 287–293. [CrossRef] 3. Tapiero, C.M.; Nagyvary, J. Prebiotic formation of cytidine nucleotides. Nature 1971, 231, 42–43. [CrossRef] 4. Morávek, J.; Kopecký, J.; Škoda, J. Thermic phosphorylations. 6. Formation of oligonucleotides from uridine 2'(3')-phosphate. Collect. Czech. Chem. Commun. 1968, 33, 4120–4124. [CrossRef] 5. Powner, M.W.; Gerland, B.; Sutherland, J.D. Synthesis of activated pyrimidine ribonucleotides in prebiotically plausible conditions. Nature 2009, 459, 239–242. [CrossRef][PubMed] 6. Powner, M.W.; Sutherland, J.D.; Szostak, J.W. Chemoselective multicomponent one-pot assembly of purine precursors in water. J. Am. Chem. Soc. 2010, 132, 16677–16688. [CrossRef][PubMed] 7. Costanzo, G.; Saladino, R.; Crestini, C.; Ciciriello, F.; Di Mauro, E. Nucleoside phosphorylation by phosphate minerals. J. Biol. Chem. 2007, 282, 16729–16735. [CrossRef][PubMed] 8. Crowe, M.A.; Sutherland, J.D. Reaction of cytidine nucleotides with cyanoacetylene: Support for the intermediacy of nucleoside- 20,30-cyclic phosphates in the prebiotic synthesis of RNA. Chembiochem 2006, 7, 951–956. [CrossRef][PubMed] 9. Shabarova, Z.; Bogdanov, A. Advanced Organic Chemistry of Nucleic Acids; VCH Verlagsgesellshaft mbH: Weinheim, Germany, 1994. 10. Mulkidjanian, A.Y.; Bychkov, A.Y.; Dibrova, D.V.; Galperin, M.Y.; Koonin, E.V. Origin of ﬁrst cells at terrestrial, anoxic geothermal ﬁelds. Proc. Natl. Acad. Sci. USA 2012, 109, E821–E830. [CrossRef] 11. Kompanichenko, V.N. Exploring the Kamchatka geothermal region in the context of life's beginning. Life 2019, 9, 41. [CrossRef] 12. Newton, A.C.; Bootman, M.D.; Scott, J.D. Second messengers. Cold Spring Harb. Perspect. Biol. 2016, 8, a005926. [CrossRef] [PubMed] 13. Costanzo, G.; Pino, S.; Ciciriello, F.; Di Mauro, E. Generation of long RNA chains in water. J. Biol. Chem. 2009, 284, 33206–33216. [CrossRef] 14. 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. [CrossRef][PubMed] 15. Morasch, M.; Mast, C.B.; Langer, J.K.; Schilcher, P.; Braun, D. Dry polymerization of 30,50-cyclic GMP to long strands of RNA. ChemBioChem 2014, 15, 879–883. [CrossRef][PubMed] 16. Šponer, J.E.; Šponer, J.; Giorgi, A.; Di Mauro, E.; Pino, S.; Costanzo, G. Untemplated nonenzymatic polymerization of 30,50 cGMP: A plausible route to 30,50-linked oligonucleotides in primordia. J. Phys. Chem. B 2015, 119, 2979–2989. [CrossRef][PubMed] Life 2021, 11, 800 13 of 13

17. Costanzo, G.; Šponer, J.E.; Šponer, J.; Cirigliano, A.; Benedetti, P.; Giliberti, V.; Polito, R.; Di Mauro, E. Sustainability and chaos in the abiotic polymerization of 30,50 cyclic guanosine monophosphate: The role of aggregation. ChemSystemsChem 2021, 3, e2000056. [CrossRef] 18. Šponer, J.E.; Šponer, J.; Výravský, J.; Šedo, O.; Zdráhal, Z.; Costanzo, G.; Di Mauro, E.; Wunnava, S.; Braun, D.; Matyášek, R.; et al. Non-enzymatic, template-free polymerization of 30,50 cyclic guanosine monophosphate on mineral surfaces. ChemSystemsChem 2021. Available online: https://chemistry-europe.onlinelibrary.wiley.com/doi/full/10.1002/syst.202100017 (accessed on 4 August 2021). [CrossRef] 19. Dorr, M.; Loffler, P.M.G.L.; Monnard, P.A. Non-enzymatic polymerization of nucleic acids from monomers: Monomer self- condensation and template-directed reactions. Curr. Org. Synth. 2012, 9, 735–763. [CrossRef] 20. Eschenmoser, A. Question 1: Commentary referring to the statement “The origin of life can be traced back to the origin of kinetic control” and the question “Do you agree with this statement; and how would you envisage the prebiotic evolutionary bridge between thermodynamic and kinetic control?” stated in section 1.1. Orig. Life Evol. Biosph. 2007, 37, 309–314. 21. Smith, M.; Drummond, G.I.; Khorana, H.G. Cyclic phosphates. IV.1 Ribonucleoside-30,50 cyclic phosphates. A general method of synthesis and some properties. J. Am. Chem. Soc. 1961, 83, 698–706. [CrossRef] 22. Saladino, R.; Botta, G.; Delfino, M.; Di Mauro, E. Meteorites as catalysts for prebiotic chemistry. Chem.-Eur. J. 2013, 19, 16916–16922. [CrossRef] 23. Iwasaki, Y.; Yamaguchi, E. Synthesis of well-defined thermoresponsive polyphosphoester macroinitiators using organocatalysts. Macromolecules 2010, 43, 2664–2666. [CrossRef] 24. Steinbach, T.; Ritz, S.; Wurm, F.R. Water-soluble poly(phosphonate)s via living ring-opening polymerization. ACS Macro Lett. 2014, 3, 244–248. [CrossRef] 25. Wunnava, S.; Dirscherl, C.F.; Výravský, J.; Kovaˇrík, A.; Matyášek, R.; Šponer, J.; Braun, D.; Šponer, J.E. Acid-catalyzed RNA- oligomerization from 3,5-cGMP at an air-water interface. Chem. Eur. J. 2021, in press. 26. Reineke, K.; Mathys, A.; Knorr, D. Shift of pH-value during thermal treatments in buffer solutions and selected foods. Int. J. Food Prop. 2011, 14, 870–881. [CrossRef] 27. Carey, F.A.; Sundberg, R.J. Advanced Organic Chemistry. Part A: Structure and Mechanism; Springer Science+Business Media, LLC: New York, NY, USA, 2007; p. 325. 28. Chwang, A.K.; Sundaralingam, M. The crystal and molecular structure of guanosine 30,50-cyclic monophosphate (cyclic GMP) sodium tetrahydrate. Acta Crystallogr. B 1974, 30, 1233–1240. [CrossRef] 29. Varughese, K.I.; Lu, C.T.; Kartha, G. Crystal and molecular structure of cyclic adenosine 30,50-monophosphate sodium salt, monoclinic form. J. Am. Chem. Soc. 1982, 104, 3398–3401. [CrossRef] 30. Scognamiglio, P.L.; Platella, C.; Napolitano, E.; Musumeci, D.; Roviello, G.N. From prebiotic chemistry to supramolecular biomedical materials: Exploring the properties of self-assembling nucleobase-containing peptides. Molecules 2021, 26, 3558. [CrossRef] 31. Slepokura, K. Purine 30:50-cyclic nucleotides with the nucleobase in a syn orientation: cAMP, cGMP and cIMP. Acta Crystallogr. C 2016, 72, 465–479. [CrossRef] 32. 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 30,50 cyclic AMP. PLoS ONE 2016, 11, e0165723. [CrossRef] 33. 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. [CrossRef][PubMed] 34. Costanzo, G.; Giorgi, A.; Scipioni, A.; Timperio, A.M.; Mancone, C.; Tripodi, M.; Kapralov, M.; Krasavin, E.; Kruse, H.; Šponer, J.; et al. Nonenzymatic oligomerization of 30,50-cyclic CMP induced by proton and UV irradiation hints at a nonfastidious origin of RNA. ChemBioChem 2017, 18, 1535–1543. [CrossRef][PubMed] 35. Šponer, J.E.; Šponer, J.; Nováková, O.; Brabec, V.; Šedo, O.; Zdráhal, Z.; Costanzo, G.; Pino, S.; Saladino, R.; Di Mauro, E. Emergence of the first catalytic oligonucleotides in a formamide-based origin scenario. Chem.-Eur. J. 2016, 22, 3572–3586. [CrossRef][PubMed] 36. Hazen, R.M. Paleomineralogy of the Hadean eon: A preliminary species list. Am. J. Sci. 2013, 313, 807–843. [CrossRef] 37. Kunkel, T.A. Mutational specificity of depurination. Proc. Natl. Acad. Sci. USA 1984, 81, 1494–1498. [CrossRef] 38. Mungi, C.V.; Bapat, N.V.; Hongo, Y.; Rajamani, S. Formation of abasic oligomers in nonenzymatic polymerization of canonical nucleotides. Life 2019, 9, 57. [CrossRef][PubMed] 39. Whicher, A.; Camprubi, E.; Pinna, S.; Herschy, B.; Lane, N. Acetyl phosphate as a primordial energy currency at the origin of life. Orig. Life Evol. Biosph. 2018, 48, 159–179. [CrossRef][PubMed]