Why Life Must Start with RNA and not Something Else.

In the previous post, I argued that reproduction had to start with RNA, but that leaves the obvious question, why not something else? The use of purines and pyrimidines to transfer energy arises simply because the purines and pyrimidines are the easiest to form, given the earliest atmosphere almost certainly was rich in ammonia, hydrogen cyanide, cyanocetylene, and urea would soon be formed. Some may argue with the “easily formed”, however leaving a sample of ammonium cyanide and urea to its own devices will get nucleobases. Cytosine is a little more difficult, but with available cyanoacetylene, it is reasonably likely. The important point is that if you accept my mechanism for how rocky planets form, these chemicals are going to be prolific. I shall justify that later. The important thing about these chemicals is that they lead to the formation of multiple hydrogen bonds only with their partners. As explained in the last post, there is no alternative to hydrogen bonds for transferring information, and these are the only chemicals that can provide accuracy under abiogenic conditions.

The polymer linking agent is phosphate, so why phosphate? Phosphoric acid has three hydrogen atoms that are available for substitution, i.e.it can form three functions. Two are to form esters and as I noted previously, the third is to provide solubility. The solubility is important because if there was not anionic repulsion, the strands would bundle together and reproduction would not work. The strands would also not provide catalysis, which occurs because a strand can fold around a cation like magnesium and form the shapes that seem to be needed. The good news is that unlike in enzymes, it can rearrange the magnesium and thus get different effects. Of course, enzymes are hugely more effective, but an enzyme generally only does one thing.

The polymer forms esters by phosphate bonding to a sugar. Think of the reaction as

P – OH  + HO – C    ->  P – O – C  + H2O       (1)

where P is the phosphorus of a phosphate or phosphoric acid, and C is the carbon atom of a sugar. Note that this reaction is reversible, but at room temperature the bonds are quite stable. These ester bonds are very strong, which is important because you do not want your carefully prepared polymer to randomly fall to bits. On the other hand, it must be able to be disrupted or substituted and not be essentially fixed, as would happen if proteins were used for information transfer. The reason is, life is evolving by random trials, and it is important that since many of these trials will be unproductive, there has to be a way to recover an many of the valuable chemicals as possible for further trials, and also to unclutter the system so that something that conveys advantages does not get lost in the morass of failures or otherwise useless stuff. Only phosphate offers these properties. In principle, you might argue for arsenate, but its bonds are weaker, thus less reliable, and worse, arsenic reacts with hydrogen sulphide (common around fumaroles which as we shall see are necessary sites) to form insoluble sulphides. These are the very pretty yellow layers in geothermal areas. No other element will do.

There are a variety of other sugars that if used to link nucleobases to phosphate will form duplexes, so the question then is, why weren’t they used? The ability to catalyse its own scission is the first of two reasons why ribose is so important. Once the strands get long enough to fold around themselves, catalysis starts, and one of the possible catalytic reactions is the promotion of the remaining OH group on the ribose to help water send the reaction (1) into reverse, which would break a link in the polymer chain. Deoxyribose does not have such a free hydroxyl and hence does not have this option, which is why DNA ended up being the information transfer chemical once a life form that had something worth keeping had emerged. What this means is that RNA has the opportunity to mutate, which is a big help in getting evolution going, and when it is broken, the bits remain available for further tries in some rearranged form.

The question then is, how do you form the phosphate ester? You mix phosphate and the sugar in solution and – oops, nothing happens. Reaction (1) is so slow at ambient temperature that you could sit there indefinitely, however, if you heat it, it does proceed. However, the rate of a reaction like this depends on the product of the concentrations on each side, with such a product on the right-hand side determining the rate of the reaction going from right to left. If you look at (1), it probably occurs to you that in aqueous solution, the concentration of water is far greater than the concentration of sugar. You will see people say that life could start around black smokers, but when you check, at the temperatures they require for the forward reaction to go it requires the concentration of water to be less than about 2%. Good luck getting that at the bottom of the ocean. You may protest that nevertheless there is life there, devouring emerging nutrients. True, although the ocean acts as a cooling bath, and the life forms have evolved protective systems. There are no such things when life is getting started. Life has moved to be close to black smokers but it did not start there.

What we need is a more precise way of delivering the required energy to the reaction site. So far, one and only one method has been found to make such initial linkages, and that is photochemical. If adenine is irradiated with light in the presence of ribose and phosphate, you get AMP, and even ATP. We now see why only ribose was chosen. AMP, and for that matter, RNA, link the nucleobases and phosphate through the ribofuranose form. Such sugars can exist in two forms: a furanose (a five-membered ring involving an oxygen atom linked to C1 of the sugar) and a pyranose form (the six-membered equivalent.). Now the first important point about a sugar is it cannot transmit electronic effects arising from the nucleobase absorbing a photon. However, it can transmit mechanical vibrational energy, and this is where the furanose becomes important. While the pyranose form is always rigid, the furanose form is flexible. The reason ribofuranose can form the links, in my opinion, is it can transmit and focus the mechanical energy to the free C-5 which will vibrate vigorously like the end of a whip and form the phosphate ester. Ribose is important because it is the only sugar with a reasonable amount of furanose form in aqueous solution. It is also worth noting that in the original experiments, no phosphate ester was formed from the pyranose form. As the furanose is used, the equilibrium ensures pyranose maintains the furanose/pyranose ratio.

That leaves open the question, how are the polymers formed? It appears that provided you can get the mers embedded in a lipid micelle or vesicle (the most primitive form of the cell wall), leaving these in the sun on a hot rock to dry them out leads to polymers of about 80 units in an hour. This is the first reason why life probably started around geothermal vents on land. Plenty of hot rocks around, with water splashes to replenish the supply of mers, and sunlight to form them. The second reason will be in the following post.

The title statement can now be answered. Life must start with RNA because it is the only agent that can lead to biological reproduction without external assistance. I started the last post indicating I would show what sort of planets might harbour life. The series is nearly there, but some might like to try the last step for themselves.

What do we need for life?

One question that intrigues man people is, is there life in the Universe besides what we see? Logic would say, almost certainly yes. The reason is we know there is a non-zero probability that it can form elswhere because it did here. That probability may be small but there is an enormous number of stars in the Universe (something in excess of 10^22 that we can see) that the conditions that led to life here must be reproduced a very large number of times. Of course, while there are such a large number of stars, by far the most are at such an extraordinarily large distance from us that they are essentially irrelevant. For the bulk of them, it took the light more than ten billion years to get here. But if we were to look for life nearby, where would we look? To answer that, we need to ask ourselves, what conditions are needed to get life started? The issue is NOT where can life exist, but rather where can it form. We know life can be found now on Earth in a wide range of environments, but that does not mean it can form there. It can migrate from somewhere else, gradually evolving systems needed to stabilize it to the new environment. The most obvious example is life emerging from the water to live on land. Nobody suggests life did not start in water because you need a solution to move nutrients around.

The first question to ask is, what are the most difficult things to achieve for life to get started? I think the four hardest things to get started are reproduction, energy transport, solubilization, and catalysis. Catalysis is required to make the chemical reactions that are desirable to go faster, and thus get more of the available resources going in the desired direction. (In this, when I use the word “desired” I mean to get on a right track to where life can get going and then support that choice during subsequent evolution. I do not mean to imply some sort of planning or directing.) Solubilization is required because many of the chemicals with functions that will be needed are not soluble in water, and hence they would simply settle out as a layer of brown gunk, which, as an aside, is what happens in many experiments designed to simulate the origin of life. What needs to happen is that something joins on at a place that does not spoil the function and then conveys solubility. Energy transport is a critical problem: if you do not have something that stores energy, functionality is restricted to microdistances from energy inputs.

Each of these critical functions, as well as reproduction, as I shall show below, depend on forming phosphate esters. Thus energy transport is mediated by adenosine tripolyphosphate (ATP), solubilisation of many of the most primitive cofactors that do not contain a lot of nitrogen or hydroxyl groups is aided by an attached adenosine monophosphate (AMP), initial catalysts came from ribozymes, RNA would be the initial source of reproduction, and both ribozymes and RNA (the latter is effectively just far longer strands of the former) are both constructed of AMP or the equivalent with different nucleobases. The commonality is the ribose and phosphate ester.

Catalysis is an interesting problem. Currently, enzymes are used, but life could not have started that way. The reason lies in the complexity of enzymes. The enzyme that will digest other protein, and hence make chemicals available from failed attempts at guessing the structure of a useful enzyme, has a precise sequence of three hundred and fifteen amino acids. There are twenty different common amino acids used (and in abiogenic situations, a lot more available) and these occur in D- and L- configurations, except for glycine, which means the probability of getting this enzyme is two in 39^315. That number is incredibly improbable. It makes selecting a specific proton in the entire Universe trivial in comparison. Worse, that catalyses ONE reaction only. That is not how initial catalysis happened.

Now, look at the problem of reproduction. Once a polymer is formed that can generate some of whatever requirements life needs, if it cannot copy itself, then it is a one-off wonder, and eventually it will degrade and be lost without a trace. Reproduction involves the need to transfer information, which in this case is some sort of a pattern. The problem here is the transfer must be accurate, but not too accurate initially, and we need different entities. By that I mean, if you just reproduced the same entity, such as in polyethene, you have two units of information: what it is and how long it is, but that second one is rather useless because life has no way of measuring the length without having a very large set of reference molecules. What life here chose appears to have been RNA, at least to start with. RNA has two purines and two pyrimidines, and it pairs them in a double helix. When reproduction occurs, one strand is the negative of the other, but if the negative pairs, we now have two strands that are equivalent to each original strand. (you retain the original.) There are four variations possible from the canonical units at any given position, and once you have many millions of units, a lot of information can be coded.

Why ribonucleic acid? The requirement is to be able to transfer information reliably, but not too accurately (I shall explain why not in a later post.) To do that, the polymer strands have to bind, and this occurs through what we call hydrogen bonds, which each give a binding energy of about 13 kJ/mol. These are chosen because they are weak enough to be ruptured, but strong enough you can get preferences. Thus adenine binds with uracil through two hydrogen bonds, which generates a little over 26 kJ/mol. (For comparison, a carbon-carbon bond is about 360 kJ/mol.) To get the 26 kJ/mol. the two hydrogen bonds have to be formed, and that can only happen of the entities have the right groups in the correct rigid configuration. When guanine bonds with cytosine, three such hydrogen bonds are formed, and the attraction is just under 40 kJ/mol. Guanine can also bind with uracil generating 26 kJ/mol., so information transfer is not necessarily totally accurate.

This binding through hydrogen bonds is critical. The bonding is strong enough to give a significant preference for each mer, but once the polymer gets long enough, the total energy (the sum of the energy of the individual pairs) holding the strands together gets to be those energies above multiplied by the number of pairs. If you have a million pairs, the strength of diamond becomes trivial, yet to reproduce, the strands must be separated. Hydrogen bonds can be separated because as the strands start to separate, water also hydrogen bonds and thus makes up for the linking energy. However, that alone is insufficient because the strand itself would be insoluble in water, and if so, the two strands linked together would remain insoluble (for those who know what this means, entropy strongly favours keeping the strands together). To achieve this, we need something that joins the mers into a chain, adds solubility, forms stable chemical bonds in general but is equally capable of being broken so that if the information creates something that is useless, we can recycle the chemicals. Only phosphate fills these requirements, but phosphate does not bind nucleobases together. Something intermediate is required, and that something is ribose.

In the next posts on this topic, I shall show you where this leads in seeing where life might be.