Chemist Shows How RNA Can Be the Starting Point for Life May 14, 2009
The Tree of Life: Tangled Roots and Sexy Shoots: Tracing the genetic pathway from the first eucaryotes to Homo sapiens Chris King Jan 2009
Biocosmology/2: Central Polymerization Pathways
Back to Biocosmology Part 1 Symmetry-breaking and Molecular Evolution?
We now examine in detail how the quantum bifurcations between the orbital structures of the elements lead to the large complex biomolecules, which in interaction form the endless replicatinve processes which constitute life on Earth today. Although this problem has for many years seemed a quite intractible mystery, recently several of the pieces have fallen into place which give a significant degree of confidence to a very straightforward description of this process centering on the unique properties of RNA and the dehydration and phosphorylation reactions associated with its polymerization.
The first stage of this path of increasing molecular complexity is the generation of organic molecules from simple precursors such as the primitve gasses that may have constituted the primal atmosphere. Although there has been some debate whether the primal atmosphere was actually as reducing as the original Miller-Urey experiments, there is likely to also have been a vast amount of organic matter deposited directly on the earth from astronomical impact during the earlier more active phase ofthe solar system. Recently millions of tons of buckyballs have been found deposited intact from space suggesting that such impacts could leave organic molecules realtively unscathed (see below).
The fact that a variety of energy sources from heat through spark chemical exudates from ocean vents, to solar radiation are all capable of generating the key monomers of the biosynthetic polymerization pathways lends weight further to the centrality of these pathways to the sturctural interaction of the four forces.
The first section has already discussed how a variety of energy sources can give rise to organic molecules of a wide variety of types. Central to these polymerizations is a process where the high energy favours the formation of the multiple bonded forms - CC-, - CN, and > C = O because they are the strongest and hence most likely to survive high energies. These in turn become capable of further polymerization, because at low energies their multiple bonds are energetically liable to open to form chain and ring interactions. The wide variety of products of these types is illustrated above for HCN and HCCH and below fro HCHO. The products include both the pyrimidine and purine bases of nucleic acids, a variety of amino acids often joined as polypeptides, porphyrins and a wide variety of other organic molecules including many capable of performing further condensations.
These reactions are also capable of producing larger structures such as microcells (see top of article) which sometimes display the bilayer structure of lipid membranes in living cells. HCN can also aggregate to a less diverse 'black polymer', although the occurrence of this will depend on the reaction conditions. Understanding the products of these polymerizations is complicated by the quantum information paradox they present. The initial conditions consist of only a few simple molecules and the final conditions are a diverse array of increasingly complex polymers.
The simplified and highly ordered conditions of traditional chemical laboratory reactions are not well-attuned to handling such complexity and the great potentialities for feed back they present.
Stanly Miller in Nature June 95 also reported that they had synthesized copious ammounts of cytosine and uracil the two pyrimidine bases that had remained difficult under plausible prebiotic conditions from cyanoacetaldehyde and urea under conditions which simulated a warm tidal pool. This comes 40 years after Miller as a 23-year old graduate student first synthesized peptides and large amounts of the purine bases adenine and guanine by spark discharge of ammonia, hydrogen, water vapour and methane.
Although people have since suggested that this mixture did not occur on the primitive earth, which would rather have had a high CO2 atmosphere, the discovery by Jeffery Bada of "mother lodes of undestroyede buckyballs - soccer-ball shaped carbon polymers containing galactic helium arrived unburned in an early meterioid - confirms that large quantities of complex organic molecules would have reached the earth's surface.
Structural Features of the - CC-, - CN, and > C = O polymerizations.
At least three distinct factors are capable of influencing the products of the polymerizations of multiply-bonded forms:
Free Energies and Resonance: The lower energy configuration of key stable products such as adenine leads to their formation based on free energy considerations.
Stochastic Kinetics: Accidental kinetic association between initial molecular species may form an organizing centre for subsequent structural evolution. For example, the HCN dimer is a key bifurcation point in the reaction. Stochastic kinetics ultimately derives its indeterminacy from quantum uncertainty.
Autocatalytic Bifurcations: Products of increasing complexity such as polypeptides and polynucleotides may generate autocatalytic pathways which alter the structural-stability of the polymerization to favour certain types of product. Polypeptides and polyribonucleotides both provide a rich variety of possibilities for autocatalysis through non-random association factors during polymerization.
Cyclic terminators: Both the HCN and HCHO polymerizations have prominent cyclic products which act as spontaneous terminators of polymerization, because the self-interaction of cyclization terminates oligomerization by removing the principal reactive moieties. The purines, pyrimidines, ribose and porphyrins all display structure consistent with being cyclic terminators. Eschenmoser (1992) has discovered that the phosphorylated oligo-aldehydes have a selective propensity to form ribose. These conditions coincide precisely with those we would expect to occur during nucleotide formation and oligomerization as a result of phosphate dehydration.
Ribonucleotides as Universal Stability Structures
Adenine is one of the principal thermodynamic products of HCN polymerization. Guanine is also formed from the same pathway. The cross-reaction of HCCH with HCN leads to a direct synthesis of the pyrimidines. The synthesis of pyrimidines has recently been found by Miller to be strongly facilitated by the presence of urea, another component of the polymerization pathway. These stages are illustrated above.
Ribose is produced in HCHO polymerization in concentrations around 2%, but the polymerization of phospho-glyceraldehyde is selective for ribose, supporting the conclusion that ribose is itself a product of the same phosphate environment that facilitates nucleotide polymerization. The particular conformation of ribose as opposed to arabinose or the other sugars appears to be important in providing the free rotation of the base and phosphate moieties and the chirality of the resulting polymer.
The nucleotide unit, as exemplified in ATP, is a quantum stability structure linking cyclic oligomers of HCN and HCHO, [adenine and ribose are pentamers of each] linked via dehydration to the dehydration-mediating phosphate group which appears to be responsible for their linkage in the first place. This structure is further stabilized by water and Mg2+. In combination with the cosmic occurrence of HCN and HCHO, this gives RNA the potential status of a generic structure in cosmology, taking the form of a non-periodic linear polymer.
The fact that polymerizations of nucleotides, amino acids and sugars alike involve a common dehydration step similarly emphasizes the direct relationship between polynucleotides, polypeptides, polysaccharides and their monomers in the phase transition from aqueous to dehydrated.
There has been a great deal of debate about whether life could have started from RNA because it is relatively difficult to polymerize under ordered laboratory conditions and has types of self-affinity which can hinder replication. This has led to a variety of suggestions from genetic takeover, the idea that some other replicative process, for example replicating crystal defects in clays might have preceded and aided RNA replication, resulting in an RNA takeover. Other people have suggested that another type of polymer might have preceded RNA. Alternatives such as Orgel's peptide nucleic acids have been suggested as a potential basis of such thinking.
However many of these arguments stem from the very difficulties experimentalists place in the way of their own understanding, by reducing their model systems to simplified controlled conditions which cannot then display the more convoluted feedback responses displayed by the wider environment, which thmselves can be very selective, as evidenced by natural separation processes such as chromatography. The fact that it has taken so long to discover the the role of the mutual interaction of clays in stabilizing ribonucleotide polymerisation emphasizes this point.
The real lesson of the evolutionary behaviour of ribozymes devised y Szostack and his co-workers, which we discuss next, despite depending on clonal selection techniques is that RNAs are very capable of strongly adaptive responses, when allowed wider behavioural interaction than simple liear polymerizations.
The key idea about the development of replicative life is that it is a fractal negentropic phase transition. We have seen that the central biological polymerizations involve dehydration. The energy currency for nucleotide polymerzation is the phosphate energy of ATP. Usually prebiotic reserchers look for an energy metabolism to support life, generally a catalytically complicated and indirect heterotrophic chemical conversion.
However it is much more likely that the initial emergence of genetic replication arose directly from an informational phase transition, rather than indirectly from a metabolism. Even today, a virus outside a cellular metabolism functions only as information. Certainly viral replication requires energetic cellular enzymes and substrates. Nevertheless the role played by the virus is precisely to produce an informational phase transition in the cell.
The essential dilemma of RNA polymerization is how information should increase (and entropy decrease) by a dehydration polymerization in an aqueous medium. RNA is energetically prone to hydrolysis, because of the free energy of dissociation of its monomers. The answer to this problem is that the aqueous medium has to be in repeated phase transition from aqueous to dehydrated. If we combine a medium in which the primal polymerizations are producing reasonable quantities of the purine and pyrimidine bases and ribose (which itself requires a high-phosphate milieu) we are led to a high phosphate dehydrating environment typified by the margins of evaporating ponds, the 'salinated' ocean edge etc. These could lead directly to the formation of oligophosphates and hence high-energy pyro-phosphate bonds typified by ADP and ATP.
It has recently been found that RNAs can be polymerized to large enough oligomers to support the replicative process by forming a binding association with silicate clays, because of the interaction of the positively charged metal groups in the clay with the phosphate groups in the oligonucleotides. This allows a geometrical stability to the polymerization process as well.
A natural model for fratal phase transition thus consists of the following four components:
The RNA Era
Like proteins, RNA is capable of forming tertiary structures as illustrated for tRNA, partly through H-bonding to the free OH group in ribose. Catalytic activity of polynucleotides, including transesterification, hinges on proton transfer . A popular concept concerning the development of genetic specificity is that the combined roles of RNA as a genetic replicator and catalyst through its tertiary structure solves a fundamental problem concerning the order of appearance of nucleic acids and coded proteins. In this model an RNA era preceded coded enzymes, in which simple replicative and enzymatic process based purely on RNA catalysis maintained a simple form of evolutionary biochemistry.
RNAs which can partially replicate
A new perspective has developed from the discovery of spontaneous splicing of RNAs in living systems and the capacity of such RNAs to function as catalysts in RNA-RNA reactions. The experimental demonstration using the G-rich template sequence of the Tetrahymena rRNA intron core to act as a C polymerase, converting C5 is into C4 and C6 has made the idea of the RNA world before proteins a natural hypothesis. The model has been extended to others for RNA-based error-correction, synthetases and the ribosome.
The ribosome consists of three types of RNA subunit the mRNA which codes the message the large and small rRNA subunits and the short tRNAs which code each amino acid to a particular triplet code of nucleotides. The ribosome, itself one of the most complex pieces of molecular machinery in the cell, containing over 50 protein units in its two-component structure, has proved capable of carrying out the essential act of translation even when virtually all of the proteins are stripped off indicating that the RNA components are not a mere scaffolding used by proteins, but the catalytic core of the process. This is consistent with the idea that the ribosome was originally a way that RNAs instructed and made proteins directly and autonomously.
Modified ribozymes have proved capable of acting as polymerases which can replicate complements to subsections of themselves. Experiments from Szostak's group give the clearest indication to date of how RNA-based replication might occur. Experimental cloning and mutation of a variety of RNAs has successfully evolved RNAs with extensive catalytic powers including partical self-assembly..
Replication in a Fractal Phase Transition
The sunY polymerase illustrates fractal RNA dynamics which could both explain the difficulties facing non-enzymic syntheses and illustrate how RNA replication developed prebiotically. The polymerization is structurally a three-level fractal process:
(a) The catalytic RNA is itself composed of separate subunits each of whose structures is simpler and shorter than the assembled enzyme, both permitting higher error rates and providing less competing secondary structure.
(b) On a second fractal level the subunits have as complements a collection of small oligomers which are small enough for any variant to exist in acceptable concentrations but long enough to provide specific binding regardless of predominant base type.
(c) Finally the oligomers require synthesis from individual nucleotides. For oligomers of length up to 4 or 5 this could be random single-stranded polymerization without reducing concentrations by more than 3 orders of magnitude. For longer oligomers a catalysed reaction using oligomer templates could maintain a non-random population of suboligomers of a multi-unit catalytic RNA. The onset of replication is then naturally modelled as a phase transition in the fractal dynamics.
RNAs polymerize proteins:
The major discovery that RNA appears to be the agent of peptide-bond synthesis in the modern ribosome and the capacity of modified ribozymes to act as amino-acyl esterases (Picarilli et.al. 1992) [the first step of ribosomal action] establish RNA can act as synthetase as well as transfer, messenger and ribosome. This gives RNA the capacity to act on its own to catalyse both its own replication and the ordered polymerization of proteins. Simpler model systems have also been advanced of the stereospecific capacity of D-nucleotides to act as a catalyst of L-amino acid polymerization. These results pinpoint RNA as the key prebiotic molecule generating ordered polynucleotide and polypeptide structures.
The development of RNA replication is modelled as a fractal phase transition. A central bifurcation pathway is coutlined, which could be capable of generating the major structural features of molecular biological evolution, including protein and nucleic acid structures, glycolysis, the tricarboxylic acid cycle, electron-transport, ion-pumping and the excitable membrane. These aspects of molecular evolution could thus be cosmologically general.
Following Thomas Cech's discovery of ribozymes, Jack Szostak and Charles Wilson revealed in Nature April 95 that they had made ribozymes capable of a broad class of catalytic reactions, not simply confined to the sugar-phosphate backbone of RNA, but including the peptide bonds of proteins and between carbon and nitrogen. They took between 100 and 1000 million 200 unit nucleotides and selected them for catalytic activity mutating and re-cloning the most successful candidates. Although the transesterficiations are as likely to snip RNA as join it David Bartel has developed ribozymes which can stitch together RNA oligomers without breaking the larger molecules.
Szostak and Eric Ekland and David Bartel argue in Science July 95 that although they have selected such ribozymes out of trillions in lab selection experiments, the ease with which they were generated suggests they are almost certainly part of a vastly larger class of similar molecules which nature is capable of producing.
Recently the situation has moved a step forwards with the active regions of RNAs capable of polymerising amono acids displaying features similar to those on biological ribosomes suggesting the biological solution may be a universal chemically-optimal sequencing arrangement.
It remains possible RNA has had a more primitive precursor. Leslie Orgel has also in Nature announced the formation of peptide nucleic acid PNA, which can serve as a template for its own replication and for formation of RNA from its subcomponents. However Jim Ferris reported in Nature May 95 that he had overcome basic problems in the polymerization of short RNA oligomers by making adenine oligomers 10-15 nucleotides long on positively charged motmorillonite clays which grew to 55 units on repeated washing with nucleotides. This bridges the gap by providing a potential source for large quantities of the oligomers similar to those used in Szostak's experiments.
Carl Woese doubts that RNA copying was the central mechanism because a study of RNA-copying genes from the diverse branches of the evolutionary tree display different solutions to this process. However the evidence of the RNA world is diverse.
Summary of evidence for the cosmoligical status of RNA
UNIVERSAL STABILITY STRUCTURES IN MOLECULAR BIOLOGY
The previous discussion of the RNA era highlights a problem that is central to the form of molecular biology - how did the central molecular biological structures become generated, starting from a simple RNA-based genetic system? The traditional viewpoint is that they were successively created during evolution through mutations building one-by-one the protein components necessary to make a working whole. This however does not explain how systems as electron transport and the citric acid cycle could have functioned with only a partial complement of enzymes. A further problem is how such enzymes would be advantageous and evolve selectively if the system they were supporting did not function in some form without coded enzymes.
The alternative thesis is that many of the major features of molecular biology have arisen in parallel as generic structures through bifurcation, independently of the emergence of RNA, and were subsequently captured by genetic takeover as genetic complexity permitted. The candidates for this primal status as stability structures include the following : The polymeric structure of proteins and RNA, the form and function of nucleotide coenzymes, bilayer membrane structure and the topological closure of the cellular environment, ion transport, concentration gradients in the cytoplasm and excitability, membrane-bound electron transport, glycolysis and the citric acid cycle.
Such a parallel model requires mutational evolution as a takeover process in fixing these stability structures into the biological scheme, but also has far-reaching conclusions concerning the generality of molecular biology in cosmological terms, for while the details of mutational evolution would be unique to each environment, the major features underlying biology would be universal.
Nucleotides and the Nucleotide Coenzymes
In addition to the key role of ADP and ATP as energy currency in the bio-metabolism, the other nucleotides have generic roles which may predate the development of coded proteins. GTP for example is used selectively in protein elongation, in the ribosome, and the nucleotides UDP and CDP are generically selective as carriers of glucose and choline and other membrane components respectively, suggesting an RNA-based selectivity for each of these classes of molecule during the RNA era. Model prebiotic reactions have successfully coupled UDP and CDP to glucose and choline (Mar et. al. 1986). Similarly other nucleotide coenzymes have generic roles consistent with a primal function. Both NAD, fig 11(a), and FAD function as carriers of redox energy through ring bond transformations, coupling H on the nicotine and flavin bases. Coenzyme A consists of adenosine coupled to pantothenic acid and functions as a carrier of acyl and other groups via the terminal SH bond (Reanney 1977). Although CoA is currently used in different processes, its structure is consistent with an initial role in pre-translatory protein synthesis. The pantothenic acid moiety appears to be a molecular fossil of two such polymerized amino acids. Vitamin B12 also illustrates how a dinucleotide could bind an Fe-porphyrin ring, lowering its Fe2+- Fe3+ activation energy and thus form a carrier of electrons. Such coenzymes would extend the nature of phosphorylation energy by linking it to H+ and e- transfer reactions, hydride ion, and peptide transfer, consistent with a model for RNA-based electron transport involving Fe-porphyrins, FeS groups, FAD & NAD.
The Form of Translation
The discovery that ribosomal, synthetase, messenger and transfer functions of protein synthesis can all in principle be carried out by RNAs without proteins leads to a natural interpretation of the development of the genetic code from a protein-free translation system. The major partitions of the genetic code have structural features consistent with an origin in underlying chemical bifurcations. The fundamental bifurcation sequence, fig 9 which should be read in conjunction with the bifurcation scheme for the amino acids in 5.1 is as follows:
The Membrane, Excitability and Ion Transport
The structure of the bilayer membrane is a direct consequence of the polarity bifurcation. The formation of amphophilic lipid-like molecules, based on a linear hydrocarbon non-polar section combined with an ionic or H-bonding polar terminal, leaves 2 degrees of freedom for layer formation. Backing of the non-polar ends completes the bilayer. Cell structure then arises directly from budding of the bilayer, as illustrated in budding in several types of prebiotic reaction medium. The use of CDP associated with choline, inisotol & lipids in membrane construction is consistent with membrane formation in the RNA era. The structure of typical biological lipids such as phosphatidyl choline display a modular structure similar to ATP, consisting of fatty acid, glycerol, and substituted amine again linked by dehydration and involving phosphateThe existence of the membrane as a non-polar structure leads to segregation into ionic and non-polar regimes. Ion transport is essential in maintaining the concentration gradients that distinguish the cytoplasm from the external environment and thus must develop in the earliest cellular systems (MacElroy et. al. 1989). Ion transport is a source of significant electronic effects, because the membrane under polarization is piezo-electric and is capable of excitation in the presence of suitable ions. Model systems using the simple 19 unit oligopeptide Na-ionopore alamethicin and artificial membranes display action potentials (Mueller and Rudin 1968). Similar results have been reported for microcells produced by prebiotic techniques containing light irradiated chromophores (Przybylski and Fox 1986), demonstrating that such effects are fundamental to the quantum architecture of lipid membranes (King 1990). Four groups of non-polypeptide neurotransmitters : acetyl-choline, catecholamines, serotonin and histamine are amines, the latter three being derived from amino acids tyrosine, tryptophan and histidine by decarboxylation. Two others are amino acids and thus also contain amine groups. Notably alamethicin also has glutamine amides located in the core of the pore (Fox & Richards 1982) consistent with a primal role for amine neurotransmitters in moderating ion flow through the membrane. The catecholamines are linked to indoles such as serotonin by a prebiotic pathway.
Electron Transport H+, e- & H2O
The fact that the proton is soluble in water to form the hydrogen ion, but the electron is not, unless attached to another group such as a quinone through reduction, causes a physical linkage to exist between the polarity bifurcation and the charge bifurcations associated with electron and proton transfer, fig 10(b). Despite the complexity of modern electron transport in photosynthesis and respiration, there is considerable evidence that membrane electrochemistry could have arisen before translation could produce coded enzymes. Firstly there is a consistent basis for the existence of many of the components of electron transport during the RNA era, since the nucleotide coenzymes NAD, FAD, a nucleotide-bound Mg & Fe-porphyrin ring similar to B12, a cysteine-bound FeS group (Hall et. al. 1974), possibly based on glutathione (g-glutamyl-cysteinyl-glycine) and quinones provide all the key components of electron transport in an RNA dependent but protein-free form, fig 10(e) (King 1990). Both porphyrins and quinones have obvious prebiotic syntheses and the primal role of nucleotide coenzymes has already been mentioned. Secondly, membrane structure and the solubility differences between the electron and proton guarantee a link between electron and hydrogen ion transport. Electron transfer does not require the coded active sites catalysing specific molecular transformations. Model systems using Fe-porphyrins and imidazole can couple oxidative electron transport to phosphorylation (Brinigar et. al. 1966) and photo activated Mg-porphyrin to phosphate link (Goncharova and Goldfelt 1990, Lozovaya et. al. 1990).
Glycolysis and the Involvement of Phosphate in Sugar Metabolism
Glycolysis forms a bridge between six and three carbon sugars, reversing the structural pathway from H2CO to the cyclic sugars, (see below). This is made energetically possible by phosphorylation, and releases high energy phosphate capable of driving other phosphorylations (Hermes-Lima and Vieyra 1989), fig 11(a). It is notable that the di-phosphorylation of fructose in glycolysis is homologous with the model route for nucleotide formation of fig 6(c). The high phosphate environment leading to RNAs would then naturally lead to similar phosphorylation of other sugars, and release of the high-energy phosphate bond through cleavage of the sugar. Mineral catalysis associated with phosphate gives the glycolytic pathway a natural basis for lysis of sugars as a dissipative structure. UDP-glucose coupling is also consistent with the involvement of glycolysis in the RNA era.
The Tricarboxylic Acid Cycle
The tricarboxylic acid cycle forms a pool of multiply carboxylated molecules which carry CO2 in various states of energy, and result in reducing energy via nucleotide coenzymes NAD and FAD, which coupled with the use coenzyme A provide a coenzyme basis for the tricarboxylic acid cycle in the RNA era. This could have thus existed as a limit cycle of di- and tri-carboxylated molecules acting both as an acceptor of acetate (a carbohydrate-equivalent i.e. (H2CO)2) and as an emitter of molecular CO2 and reducing H, thus bifurcating carbohydrate level redox potential into reduced and oxidized components. The linkage to nucleotide coenzymes such as NAD would have served to create a bifurcation of redox potential in the molecular milieu contributing to the diversity of reacting species. This gives at least one possible role for Eigen's hypercycle concept however the process could have also been more chaotic, consisting of a population of molecules undergoing various generic transformations with net inflow of carboxylic acids and net emission of CO2 and transfer of H, due to generic transformations as illustrated in fig 11(b). Isomerization would have been catalysed by Fe2+. Several steps may have been driven by sunlight photolysis.
Genetic Takeover of the Generic Systems
The probability that the the central structures of molecular biology existed in the RNA era is consistent with their being chemical stability structures utilized by catalytic RNAs. The small genomes during the RNA era and limited catalytic capacity of RNAs by comparison with protein makes it likely that an RNA-based system had to capitalize on existing chemical stability structures without requiring enzyme-based biosynthetic pathways. Genetic takeover of the major features illustrated in fig 12(b) is consistent with such a limited role for RNA catalysis. However it also places these stability structures clearly in a category determined by cosmological symmetry-breaking, thus giving evolutionary biology a common pattern of inheritance on a cosmological footing. The model thus gives a more plausible account of the RNA era and makes specific predictions about the aspects of biology likely to be common to the universe at large.
The Terrestrial Record
Evidence fo life has been found in the earlist rocks leaving only a few hundred million years for life to form. The prebiotic syntheses of uracil and cytosine have been established by Miller himself, a prebiotically-plausible synthsis for RNA is emerging from Ferris's work and the selection of RNAs with catalytic activity has been amply demonstrated by Szostak and others. What was once a major impenetrable mystery is rapidly becoming a straightforward process.
First life on Earth survived battering by meteors New Scientist 9 Nov 96
In Nature (384 p 55) Gustaf Arrhenius studying tiny apatite grains in the Isua formation of Greenland has found carbon 12 to 13 ratios consistent with the grains originating from living matter.
The Isua rocks date from 3.85 billion years ago. Although indications from zircon crystals indicate a solid crust 4.2 billion years ago, no intact rocks have been discovered older than 3.96 billion years. The moon and probably the Earth likewise was heavily bombarded with meteors up to 3.8 billion years ago. This suggests that life evolved on earth as soon as environmental conditions allowed.