6.1 RNA, A Molecule with Skills
Geologists like to divide the earth’s history into time periods during which conditions were constant over a long period of time. A somewhat fuzzy but more complementary division is made with the help of living beings, which at certain times constituted the predominant groups. The best known is the era of the dinosaurs, which began about 230 million years ago and lasted until the Cretaceous–Paleogene (K–Pg) boundary 66 million years ago. The era of the mammals followed. Prior to the dinosaurs, there were times when, for instance, amphibians, fish, or trilobites existed. There was a suspicion relatively early for the beginning of life that the complex information store DNA represents an evolutionarily higher development than RNA, which can also store information in a way. The considerations originated from the US microbiologist Carl Richard Woese. Building on this, Walter Gilbert, an American biochemist, proposed the term “‘RNA world”’ in the mid-1980s. The era of RNA was born, the exact beginning and end of which are still unclear. At least since then, many scientists who research the origin of life have been advocating the RNA world hypothesis.
And what makes the RNA world hypothesis so attractive? It was the discovery of a special RNA that is capable of developing catalytic properties [1]. It was found in the ribosome of a eukaryotic unicellular organism, in the molecular tool responsible for the assembly of the proteins. In other words, this RNA can be something that enzymes normally do elsewhere. However, the speed of the catalytic processes is much slower than that of enzymes today. The idea quickly developed that, at the beginning of life, primarily an RNA with catalytic properties existed, and only gradually did the functions switch to the more efficient enzymes. And that is not everything. At the same time, the sequence of the base triplets means that RNA is an information store that only had to be brought into the right relationship with the other molecules. The RNA may even have started the amino acid chaining to peptides, which continues to this day in the ribosomes. In addition, under certain circumstances, the RNA can copy itself, which represents the ideal precondition linking the protein world to the RNA world.
Laboratory tests have shown that ribose, the sugar from RNA, is much easier to obtain from prebiotic starting materials available than deoxyribose, the sugar from DNA. In cells today, deoxyribose is produced with the help of an enzyme from ribose. In addition, the DNA is generated from RNA building blocks, which therefore have to be present first in the cell. These are indications of the life history of molecules, which suggest that RNA existed before DNA. The transition from the RNA to the DNA world apparently had its advantages due to the significantly shorter lifespan of RNA owing to its chemical instability. The reason for this is the slightly different composition of the sugar ribose in RNA compared to the deoxyribose in DNA. While the ribose has an OH molecule at one point in the ring-shaped sugar, at the same point in the DNA sugar there is only one hydrogen molecule that lacks oxygen (deoxy = without oxygen). If OH molecules occur at higher pH values due to the presence of bases, they will extract the hydrogen (H) from the OH molecule in the ribose in order to form a water molecule. This does not happen with the single hydrogen atom in deoxyribose since it is too tightly bound to the sugar ring. With the loss of the hydrogen atom in the ribose, the remaining oxygen immediately combines with the phosphate in the backbone of the RNA. This simultaneously gives up the bond to the next ribose molecule, and the RNA disintegrates [2]. It is hydrolyzed. The reaction also shows that the RNA is unstable at higher pH values. And from a biochemical point of view, this was exactly one of the arguments against white smokers. Waters with high pH values appear on them, which give RNA no chance of survival. In cells today, the lifespan of the various RNA molecules is limited to a few minutes. Afterward, they have fulfilled their function, are separated into individual components, and become available to assemble new RNA with new tasks. DNA, on the other hand, has a high level of stability with the same storage principle, which is evidenced solely by findings in the bones of people from prehistoric times such as the Neanderthals.
The comparison of RNA and DNA makes it probable that the RNA formed at the beginning of life and, as evolution progressed, the more stable long-term storage DNA developed from it. The RNA was retained and developed into the bearer of a wide variety of functions. But now the question arises as to what the information store looked like or how it functioned before the more stable DNA replaced the RNA.
And so, we touch one of the main difficulties in the RNA world. When DNA took to the stage in the course of evolution, the processes of storing and the information content in the base triplets were very likely already long established. This means that the function the DNA had taken over must have been performed previously by the less stable form of the RNA. Assuming that at the beginning nucleotides had been assembled into an RNA in a favorable environment—without “building instructions,” a random sequence of bases resulted with no information content for arranging amino acids in a protein. The matrix that gave the base sequence a logic was missing. The situation is comparable to a long column of the numbers 1–4 in any order without spaces, which fills an entire book. Three of the four numbers should always represent an information unit. Without an interface to a selection system that assigns a precisely defined content to a block of three and at the same time specifies from which of the first three numbers the selection process should start, no information can be selected. However, the strand itself represents information that is passed on each time copying takes place. If a randomly compiled sequence of bases in an RNA receives a certain three-dimensional structure that is, for example, catalytically active, this property is retained by the copying.
An interesting consideration can be made regarding the assignment of the codons from an RNA to a specific amino acid sequence in a chain. Assuming that RNA strands from the very early stages of chemical evolution were found today, then, as with today’s RNA strands, three adjacent bases could be defined to form a triplet. The difficulty here is determining a start. The bases in an RNA are all adjacent to one another, which means that the first group of three can begin at three different points. Each time the start is “moved forward” by one base, all the triplets move to a new three-group sort. As a consequence, three completely different assignments exist in the triplet definition for reading an RNA. This does not even matter at first; it only increases the possibilities for subsequent consideration: complementary anticodons of today’s tRNAs would certainly be found for the random sequence of bases within each triplet, each specific for one amino acid (with the four exceptions of the start and stop triplets). More than 4 billion years ago, however, today’s specifically loadable tRNAs did not yet exist. That means that with the help of modern tRNAs, we could test the RNA from the early days and determine which amino acid chain results from its supposed store. The result is predictable. Completely nonfunctional amino acid chains would emerge each time. In other words, the information content of a random base sequence in an RNA cannot be used as long as a connection to an information system does not exist. This system requires an exact information assignment between RNA, tRNAs, and associated synthetases, which load the tRNAs specifically. At this point we have an egg, but we don’t know how it was laid, because we don’t have a hen.
6.2 Problems in the RNA World
Previous experiments on the generation of long RNA strands have shown that the degradation proceeds faster than the assembly owing to its own catalytic activity [1, 3]. This means that a randomly formed longer RNA strand has a very short lifespan, since it is broken up into short sections immediately. An interesting discovery in this context was that RNA molecules are linked into longer strands when they circulate in a cell with a thermal gradient [4]. In a laboratory experiment, one side of a capillary a few millimeters thick was heated to over 70 °C while the opposite side was cooled, so that a temperature difference of over 30 °C occurred. While the strands on the warmer side disintegrated, chains were formed on the cooler side [5].
Connecting the concept for the model to existing natural systems is, however, difficult. The conditions in the microchannels in the white smokers were given as an example, from which the higher temperature waters from the oceanic crust reach the seawater. However, the solution load in the water is so high that crystallizing minerals constantly seal the tubules, so that the pathways constantly shift. In addition, the pH values for the solutions that emerge are very high (see Sect. 5.6), so that possible RNA molecules are quickly hydrolyzed.
It is therefore necessary to search for a realistic geological environment for the early days of the earth, which could provide constant conditions over very long periods. Ribozymes (catalytically active RNA molecules) previously developed in experiments in the laboratory demonstrated a relatively high error rate in reproduction and only very short sections could be reproduced [3].
But apart from that—even if a randomly formed RNA copies itself any number of times, suffers errors while copying, and is constantly being enhanced (in the direction of improved catalysis)—it is impossible for it to randomly catalyze enzymes that simultaneously form up to 20 different synthetases, which in turn load tRNAs so specifically that the peptide machinery can develop from them. This would be chicken that hatches from the egg that it laid itself.