1. Weird Life in a Shadow Biosphere

Wolfe-Simon et al. (2010) in their recently highly publicized article, have discussed intriguing results which they believe are related to the origins of life, as based on measurements on the presence of arsenic in various small molecules in the bacterium GFAJ-1 of the family Halomonodacaea of the Gammaprotobacteria, which are related to the well-known Escherichia coli. This bacterium had been isolated from a hypersaline and alkaline lake in California, Mono Lake, which contains high concentrations of dissolved arsenic. Under aerobic laboratory conditions with high arsenate (AsO₄^-3) and low phosphate (PO₄^-3) concentrations this bacterium would have substituted arsenic for its phosphorus in various molecules, including at several locations in its DNA. Although this article did not go deeply into evolutionary implications of this finding, subsequent media discussions concentrated on consequences with regard to the origin of life both on Earth and on other planets within or outside the solar system. The substitution of phosphorus by arsenic would imply that weird metabolisms other than one based on the current six elements, carbon, hydrogen, nitrogen, oxygen, sulfur and phosphorus, might have been at the basis of early life. And this would give a new possibility of life of another form originating under different conditions on other planets as well. Also, different, chemically weird life forms may still exist under different conditions here on Earth, constituting a shadow biosphere. On the other hand, this elemental substitution would be in line with other, known, substitutions such as the metal tungsten by molybdenum.

This article with its experimental underpinning was a logical follow-up of two earlier papers (Davies et al., 2009; Wolfe-Simon et al., 2009) of which, to my mind, the paper by Wolfe-Simon et al. (2009) is the most interesting. It is written as a logical extension of a much earlier, classical paper by Westheimer (1987) in which this author asked why nature chose phosphorus in its metabolically principal molecules. Wolfe-Simon et al. (2009) now asked why nature also chose arsenic, and possibly even did so before phosphorus. They then elaborated the idea that arsenic could have been the evolutionarily primary element in ATP, in nucleotides and RNA, and finally in DNA. At a later stage it would have been substituted by phosphorus, found just above it in the same period within the Periodic 4 System of Elements (PSE). As such, phosphorus has chemical properties similar to arsenic. However, because it is smaller, the bond lengths of phosphorus can be shorter and therefore more stable, which can be of advantage in living systems. Because of its chemical similarity, it could easily substitute arsenic without major changes having to be made in the pertaining molecules and their metabolic operation.

This earlier article, as well as the next two, therefore nicely filled a gap in the reasoning Fedonkin and I had developed some years previously (Hengeveld and Fedonkin, 2007) in an article containing the same idea but worked out in an analysis of the PSE as a whole. Moreover, our article put the evolutionary trend presumably followed into the context of a changing environment to which the early systems had to adapt. The chemical environment possibly started from reducing conditions, after which it became increasingly more oxidizing. This had repercussions for the selection of chemical elements that constituted the systems successively. The early selection would have concerned elements with the weakest bonds because the compounds had to be formed and broken up without the use of enzymes, whereas the systems were also still simple enough not to require elaborate systems of standardization to keep order in their dynamic. We therefore deduced that selenium might have been selected initially, after which it was first substituted by sulfur, and this, in turn, by oxygen. Similar trends were recognised among other non-metals as well as among metals. The deduced biological evolution thus followed constraints imposed by the physical properties of the elements, therefore following the structure of the PSE. One significant consequence of this deduction was that the currently dominating elements at the top right of the PSE, carbon, nitrogen, and oxygen, were latecomers rather than upstarters of life. Although we did discuss the place of arsenic in this scheme, were felt too uncertain to mention it in our paper, although we realized the potential significance of its toxicity in this connection. This aspect was beautifully given shape in Wolfe-Simon et al. (2009), and worked out in its two follow-up papers.

The fact that this possible early evolutionary trend could be deduced from physical properties of the elements on which the PSE is based is significant for three reasons. Firstly, predictions can be made, which can then be tested experimentally, as was indeed done in Wolfe-Simon et al. (2010). Secondly, these physical properties imposed constraints on the conditions under which living systems can start 5 up and develop. This means that they put constraints on the evolution of the metabolism of living systems, rather than open up new, weird possibilities. Given enough time, the same elements will select out. And finally, it is unlikely that early forms of life still exist in some shadow biosphere, even here on Earth, because the environmental conditions changed radically over the few billion years since life originated, preventing primitive systems from functioning as conditions changed. Not only were the experiments of Wolfe-Simon et al. (2010) done with an evolutionarily derived bacterium, but they were done under aerobic and therefore under very oxidizing conditions rather than under the reducing ones prevalent at life’s upstart, and conditional to this upstart. One may therefore wonder whether arsenic can still have its place in present-day organisms and under present-day conditions. It may be that meanwhile life in its initial form has become obsolete, along with arsenic as a chemical constituent. Because of its similarity to phosphorus, it can now only confuse phosphate metabolism and thus become toxic (Wolfe-Simon et al., 2009).

Similarly, the fact that such a possible trend can be deduced rather restricts our possibility of finding life on other planets than widen it, certainly in its more derived form as found here on Earth. As a system, any living organism depends on interactions between a great number of components of an ever-increasing complexity with respect to their form and functioning. The number of interactions is therefore vast, some of which may be prohibited biologically whereas others are promoted. Elsasser (1987) therefore speaks of the incalculability of this number, a number that soon becomes greater than the number of atoms in the universe. The chance that similar life forms with the same set of molecules with their specific interactions could also be formed elsewhere is therefore minute; in my opinion too small to be considered seriously. Therefore, the possibility of life using the same elements, and this in the same evolutionary sequence, elsewhere in the universe becomes greater, the stronger the physical restrictions. However, the possibility of living systems developing elsewhere, in the same form and complexity as here on Earth, gets smaller.

The next step that should be taken by the group of Wolfe-Simon is therefore not in the direction of trying to find weird, early life forms in some forgotten, shadowy place here on Earth, nor on some other planet. Rather, it should be in the direction of understanding the system as it is at present, and as it originated and developed on Earth in the distant past. In that sense, their studies should remain confined to an analysis of a realized system instead of that of an imaginary one in an unknown place. This is a scientifically reliable and well-trodden path. Only then, can one obtain testable results and, hence, insight into the world around us.

For example, instead of looking at individual molecules themselves, one can also investigate their biochemical role and functioning. In the case of phosphates and other molecules derived from them, one can think of their role in the energy transfer within cells. It may be inferred that arsenates and their derivatives have had similar roles, which were then gradually taken over by the phosphates. Then, the question is, what happened to this energy downstream in the early metabolism? Was there any relationship or interaction with molecules that contained selenium? Selenium as a direct neighbor of arsenic would have been present in the earliest life forms as well (Hengeveld and Fedonkin, 2007). Finding out whether or not this is feasible could be a first step in reconstructing an interactive, metabolic system. The questions arising then are how living systems could have built up and how they could have functioned. How they could have originated, functioned, and developed evolutionarily as energy dissipating systems. Basic to their chemistry, it is principally the physics of the systems that is of interest. Life as a phenomenon poses in the first place a physical, thermodynamic problem; its thermodynamics define the shape and functioning of the individual molecules, their interactions, and the system as a whole. It is therefore its thermodynamic functioning that first needs to be understood before any further steps can be taken, whichever they may be.

Another way forward is to investigate whether arsenic occurs in certain molecules active in metabolism, such as ATP, or certain nucleotides. Several of these could have had an ancient history in life and might thus indicate a role for arsenic in the beginning of life. Similarly, in that case, it may be instructive to investigate with which molecules it could interact and to find out whether these other molecules have had an ancient history as well. In this case, the kind of interaction may be interesting because this can indicate its possible function in the early systems.

Thus, there is much to be investigated straightaway instead of trying to find out about weird life forms and where they occur; in the end the answer will probably remain elusive.

2. Conclusion

It is to be expected that under certain – laboratory – conditions arsenic can replace phosphorus. But as yet it is not known whether it played some evolutionary role in the origin of life on Earth. It is unlikely that remnants of former life forms, if they ever existed, will be found as living fossils. The article by Wolfe-Simon et al. (2010) has the merit of drawing to arsenic the attention it deserves. Given its place in the PSE and given its toxicological effects it certainly is an intriguing element. The next step to be taken is to investigate in what kind of organisms arsenic is normally found, in what parts of the metabolism, and how it functions. Moreover, the question arises: can these organisms and the relevant parts of their metabolism be considered ancient? DNA is probably already too derived to be of use. At this moment, thinking about further implications with respect to the origin of life here on Earth and elsewhere in the universe, however tempting, deflects from the work that still needs to be done on biogenesis. Apart from this, we should first be clear about a likely origin and evolution of life here on Earth before we can justifiably search for life in some stage of development elsewhere.