1. The Early Earth

It is generally accepted that the Earth formed by accretion of planetesimals consisting of mainly rocky material with some ice over 4.5 billion years ago (Dalrymple 2001). Accretion culminated in the impact of the planet "Theia" which struck the Earth and sheared off part of the crust and mantle to produce the moon (Belbruno and Gott III 2005; Wood and Halliday, 2005). Gravitational energy was released as the Earth formed and this energy, aided by that from radioactive decay resulted in the heating of the protoplanet to around 1000^oC. This would be enough to destroy most of the organics. Our hot Earth was cooled through radiation to cold space. As it cooled, water condensed and gases such as nitrogen and carbon dioxide were retained in the atmosphere. The Moon is a comparatively large satellite compared to planet Earth and was also considerably closer than it is today. It is also been suggested that early Earth was covered in an ocean with little or no dry land. Calculations of present Earth water would suggest that the Hadean Ocean was several kilometres deep (Debenedetti and Stanley, 2003).

The early atmosphere was almost certainly denser than at present (around 10x) and rich in carbon dioxide, resulting in an acidic Ocean (Morse and McKenzie 1998). The Sun, then a young star, was in the "T Tauri" stage when the energy output is variable but certainly less than that of the present mature star (Caffe et al. 1987). The temperature may have been lower than today.

A large body of evidence suggests that life emerged early (probably more 3.5 billion years ago; Mojzsis, et al., 1996; Nemchin et al. 2008; O'Neil et al. 2008). The surface environment was probably hostile, with a moon whipping up the Ocean, meteorite and comet strikes common and an Ocean pH of around 5.

2. A Short History of Scientific Ideas on the Origin of Life.

The modern theory of the "organic soup" probably begins with the "warm little pond" mentioned by Charles Darwin in his letter to his friend Joseph Hooker in 1871 (Darwin 1888). It is unlikely that there were many, or any, ponds around on early Earth. If there were any, they would have been liable to have been destroyed by the volatile conditions that prevailed, such as the shocks following meteorite and comet strikes.

In the late nineteenth century the widely held belief that life such as bacteria could arise by spontaneous generation was disproved by Pasteur among others (Farley 1977). The similar concept that life could arise from inanimate chemicals also became unfashionable. Oparin (1924) and Haldane (Haldane 1929) revived this idea, both independently suggesting that a reducing atmosphere through which lightning passed would produce a wide variety of chemicals (Dronamraju 1968). These chemicals would dissolve in water to produce the "primordial soup".

The concept of abiogenesis gained further momentum in the early 1950s when Miller (1953) passed sparks through a methane rich reducing atmosphere and made a number of amino acids. About the same time, it was shown that nucleic acids were the repository of genetic information (Avery et al. 1944). These two results produced two theories. One suggested a protein world on the basis that life was dependent on enzymes and these would have to emerge first, the other posited an "RNA World."

According to the "protein first" theory of life's origins, amino acids joined up spontaneously. A typical experiment was reported by Fox and Dose (1977). They made "Proteinoids" by heating amino acid mixtures above 120^oC. They avoided the pitfall of producing diketopiperazines by enriching the mixture with lysine. But such high temperatures racemised the amino acids and the products were branched polypeptides. They postulated that rain would wash these polymers into the ocean and then achieve life. According to theory, the proteinoids would have absorbed water, swelled and split into smaller units imitating growing bacteria. But the system had no memory and thus no genetic mechanism of adapting to the environment.

The "RNA World" proponents had a system with potential memory. When it was demonstrated that RNA could be catalytic (Kruger et al. 1982), it looked as though the problem of life's origins was solved. However, this system also had its flaws. Nobody could make the monomers of RNA, RNA cannot copy itself from monomers and the spectrum of its catalytic ability is limited (Mills and Kenyon 1996). Despite these problems, the "RNA world" idea became dominant in the 1990s and remains so today.

Both the RNA world and protein first hypotheses have common problems. They are both centred on polymers that are formed with the elimination of water. The monomers would be in dilute solution, and simple mass action calculations demonstrate that this is a major problem. At equilibrium K1/K2= [A]ⁿ x [B]^m/ [C]p x [D]^q for the reaction nA+mB = pC +qD, where K1/K2 rate constants for the forward and reverse reactions. For amino acids undergoing polymerisation with the removal of water in water, the concentration of water will be around 54 molar, while the amino acids will almost certainly be present at millimolar concentrations or lower. Such a system also assumes that the mere presence of the chemicals and their polymers is enough to achieve life. This is manifestly untrue as a dead organism can contain all the polymers necessary for life and in exactly the same concentrations as in life, and yet remain dead. Energy is needed to oppose the drive to entropy. Mary Shelley (1818) recognised this in her novel "Frankenstein" where outside energy had to be applied to bring the monster to life. She notes a large force such as lightning destroys, but the small voltages of Galvanism might spark life.

One proposed solution to these problems is to shift the origin of life to another part of the universe. "Panspermia" (Crick and Orgel 1973, Joseph 2009) in its various forms still has the problems of explaining the origin of life, but also a "delivery system" is now required. There is then the need to explain how this life could exist and thrive outside its initial privileged environment on a planet such as Earth. Joseph (2009; Joseph and Schild 2010) has attempted to provide detailed answers to these questions, though the ultimate question still remains: how did life begin?

Science builds on previous theories and results. Building on the protein first and RNA worlds, both of which incorporate sound ideas, we would like to explore ideas based on a combination of the two. Both theories require a source of energy and a dynamic flow that introduces substrates and removes waste. We believe an Alkaline World provides the answers to these riddles.

3. The Alkaline World

As the moon orbits Earth, not only does it perturb the oceans, but also Earth’s crust (Stacey 1992). This happens at the present time but the effect would have been far greater when the moon was closer to Earth. Therefore, it can be assumed that cracks in the crust appeared and water seeped into the rock, which would have been mainly olivine. This cold acidic water became involved in the process of serpentinization of the crust (Russell et al. 1993, 1994, 2010; Russell and Hall 1997, Holm 2006), which makes it warmer, reducing and alkaline.

This process also enriches the solution with ions such a magnesium, silica and sulfide (Mielke et al. 2010). As the warm alkaline sulfide-bearing seepage emerged into the Hadean Ocean with its load of ferrous iron derived from very hot acidic springs, metal sulfide precipitates formed semi-permeable tubes and micro-geodic compartments comprising hydroxysilicates with dispersed nanocrysts of sulfide minerals such as mackinawite ([Fe₂S₂]n) (Martin and Russell 2007; Mielke et al., 2010). Across the tube wall there would have formed a proton gradient of about 4 pH units, enough to provide the energy for life (Fig. 1). Sulfide precipitates with the ferrous ions dissolved in the ocean to form a tube of the semi-permeable minerals such as Mackinawite (Martin and Russell 2007). Therefore, across the tube wall there would have developed a proton gradient of about 4 pH units, enough to provide the energy for life (Fig. 1) if it can be tapped.

As in existing life, high energy bonds can be synthesised by a proton gradient (Mitchell 1961, Williams 1975). Typically these are polyphosphates but thioesters may play a role especially in early life, and may mediate high energy phosphate synthesis. Phosphorylation of a substrate such as glyceraldehyde leads to the carbohydrate metabolism as described below and from there to nucleotides and a genetic material.

This proposed system also provides a dynamic environment for the elimination of waste. Many "origin of life" experiments are conducted in a vessel where the reaction takes place and a "work up" follows to identify the products. In life, there is an inherent system of removal of unwanted products. The integrity of the system is maintained and potentially toxic or interfering chemicals are removed, and absence of such a process leads to system impairment. In living entities, these systems range from porins in bacteria, vacuoles in amoebae to kidneys in mammals. All living entities have some system for dealing with waste molecules.

The flow through the tube would cause the trapping of different molecules in different environments. Sugar phosphates are trapped in a matrix of alkaline earth ions (magnesium and probably also calcium) in an alkaline environment, RNA precipitated in a neutral/acid environment will trap amino acids (Mellersh 1993, Mellersh and Wilkinson 2000). Molecules and ions that are not trapped will flow back into the Ocean as waste (Fig. 1).

4. The chemistry of the Alkaline World

Amino acids can be synthesised from ammonia and simple organic chains such as α- ketoglutarate. Hydrogen cyanide is readily polymerised to the nucleobase adenine. This and similar chemistries in alkaline conditions are very well covered by Holm and Neubeck (2009). In this paper we want to focus on the synthesis of the carbohydrates and subsequently a nucleotide in an alkaline environment.

Carbohydrates have many functions in living organisms yet rarely get mentioned as important in the origin of life. They are central to the production of energy both by photosynthesis and glycolysis. As polymers they are an energy store and have structural roles as cellulose, chitin, ground substance and others. The obvious route of synthesis is through the formose reaction first performed by Butlerow in 1861. It has however fallen into disrepute because not only does it produce the carbohydrates of life, it makes many others. These problems have been reviewed by Orgel (2004) and Holm et al. (2006). Orgel (2004) ends his review saying "if it could be directed towards the synthesis of ribose, [it] would produce an ideal route". Holm et al. (2006) concludes by saying "the formose reaction must still be considered to be of great potential if we could identify some selective mechanism that would interact with it."

Life uses a reaction very similar to that of Butlerow but controls it using phosphorylated substrates. The charge and bulk of the phosphate controls the orientation and limits the reaction possibilities within the enzyme. To recap on the formose reaction, it is the polymerization of formaldehyde CH₂O which occurs in alkaline solution. The sugars are (CH₂O)n. The first two sugars, formaldehyde and glycoaldehyde are not very reactive together. When we get to n=3 the situation alters dramatically (Orgel 2004). It is therefore probable that the simple polymerisation of CH₂O is not the only route to this metabolism but that others such as thioacetates contribute more until n=3. It is interesting that acetyl CoA is important in this respect in contemporary biochemistry (Russell and Martin 2004). Glyceraldehyde being a cisdiol is easily phosphorylated (Kolb and Orgel 1996) to glyceraldehyde-3-phosphate (GA-3-P) (Fig. 2).

GA-3-P has two important features which are essential to the understanding of how life emerges in the Alkaline World. GA-3-P reversibly isomerises to the other phosphorylated triose, dihydroxacetone phosphate (DHAP) (Fig. 2). They can present different functional groups to each other. GA-3-P has an aldehyde group with the carbon having a partial positive charge and the DHAP has an enolate in which the end carbon is negatively charged. Secondly, it is chiral, and that initial chirality will persist and direct the formation of new chiral centres. In Figure 3, the GA-3-P is S. Charge and bulk repulsions leads to the oxygen labelled 2 being anti to the alcohol (3). As the enolate and the aldehyde approach, the enolate oxygen (1) will be forced to be anti to 2.

In this way the chiral centre has dictated the position of the oxygens 1 and 2 so that when the bond between the two carbons is formed the stereochemistry at the new chiral centres is predetermined and fructose is the product. In this case L-fructose but if the chiral centre of GA-3-P had been R, it would have been D-fructose.

In the classical formose reaction the reactants are in solution and have freedom. In the seepage tube the reactants are precipitated by their phosphate groups in a magnesium phosphate matrix. The hydrophilic sugar moiety is projected into the surrounding aqueous layer. The distance between the phosphate centres in such lattices is approximately 7Ĺ. The functional groups of the trioses fixed to the surface can only react if they can "bridge the gap". This effectively means that the distance between the phosphorus atom and the functional group on the same triose must be greater than 4.5Ĺ.

Glyceraldehyde can form either the 2- or 3- phosphate. The 2-phosphate has a maximum phosphorus to C1 distance of 4Ĺ and cannot participate in a reaction with DHAP. Its fate is either to lose its phosphate and be swept along with the flow to go to waste in the Ocean or to isomerise to GA-3-P. In alkali DHAP becomes an enolate at the non-phosphorylated end conferring a negative charge on the terminal carbon which can react with the positive of the 1-carbon of GA-3-P to form a bond. The flow keeps the system clean by removing non-phosphorylated molecules from the system and retaining those with phosphate groups. Retention by charge is one of the major functions of phosphate in biochemistry (Westheimer 1997), the other being to provide leaving groups which could operate in a biological system.

GA-3-P reacts with DHAP to produce fructose-1,6-diphosphate, which at high pH loses its 1-phosphate to become fructose-6-phosphate (Fig. 4, reaction 1). Fructose-6- phosphate can then react with GA-3-P to form a 1,9-diphosphorylated nonose which is unstable and breaks down to erythrose-4-phosphate and ribulose-5-phosphate (Fig. 4, reaction 2). Erythrose-4-phosphate can react with DHAP to form seduloheptose- 1,7-diphosphate which loses its 1-phosphate under alkaline conditions to produce seduloheptose-7-phosphate (Fig. 4, reaction 3). Seduloheptose-7-phosphate will react with GA-3-P to form a 1,10-diphosphorylated decose, which will break down to form ribose-5-phosphate and ribulose(ketopentose)-5-phosphate (Fig. 4 reaction 4). In the alkaline conditions the Lobry de Bruyn-van Ekenstein transformation establishes an equilibrium between aldoses and ketoses. As ribose-5-phosphate is used for further reactions it is replaced (Fig. 4, reaction 5). In this way the formose reaction has been "directed" to ribose-5-phosphate.

The importance of these pathways is that they mimic the chemistry of modern enzymes. For instance aldolases coordinate the phosphate group of the sugars with magnesium ions, and magnesium ions are used to stabilise the enolate in Class II aldolases (bacterial). Class I aldolases, those of "higher organisms", use essentially the same route, but with an enamine intermediate (Bugg 1997). The Alkaline World uses less efficient and less controllable routes to achieve the same effect until protein synthesis can emerge to produce enzymes to takeover theses functions.

Nucleotides such as AMP have never been synthesised in a plausible prebiotic environment and this synthesis has proved a major stumbling block in the chemical origin of life (Orgel 2004). Almost all the experiments have been carried out at neutral pH, but the alkaline world might offer a solution. A lone pair on N9 of a purine or N1 of a pyrimidine would allow the base to react with the carbonyl of ribose -5-phosphate. This is an obvious route to nucleotide formation (Powner et al. 2009, Bean et al. 2006) and is the mechanism by which the enzyme adenine phosphoribosyl transferase activates adenine (Phillips et al. 1999). The pKa of the adenine N9 is around 10.1 (Gonnella et al. 1983), which could be achieved in the Alkaline World.

Ribose is usually depicted in the ring form, but is also present in the open form with the aldehyde group open to attack by a nucleophile. This is evidenced by the ability of ribose to act as a "reducing sugar". The carbon of the aldehyde group has a positive charge and the adenine N9 a lone pair (Fig. 5). When the bond N9 to C1 of ribose is formed a new chiral centre is formed. The two stereoisomers have different energies and the anomerisation to form AMP is favoured in the R isomer as the oxygen at O1’ hinders O4’ in approach to C1 in the S isomer but not the R isomer.

Magnesium is required to catalyse the formation of AMP from adenine and phosphoribosyl pyrophosphate (Phillips et al. 1999). It may also act as the hydroxyl acceptor from C1’ (Bean et al. 2007) in the formation of nucleotides. Herrero and Terrón (1999) have shown that magnesium ions facilitate the stacking of AMP which would presumably promote oligomer formation from activated nucleotides. Magnesium is an essential ion in most nucleic acid chemistry and it is yet another advantage of the alkaline world that it is present, probably at concentrations close to saturation, in the seepage.

5. Conclusion

The Alkaline World is a dynamic model that includes energy, flow and waste removal. Precipitation of magnesium phosphates in alkaline conditions results in the direction of a Buterlow-type reaction leading to ribose-5-phosphate, through a cascade of phosphorylated sugars. The high pH also allows the abstraction of the proton at the nitrogen of the base which attaches to C1 of ribose (Gonnella et al. 1983), potentially a crucial stage in nucleotide synthesis (Powner et al. 2009, Orgel 2004).

Formation of sugars by such reactions at the early stage of the origin of life has the potential to leave a metabolic dowry allowing future metabolisms to emerge. The essential phosphorylated sugars for the "dark reaction" of photosynthesis are present. GA-3-P is ready for amination to amino acids. Other molecules required for modern biochemical pathways such as phospholipids are also in place.