Australia: The Land Where Time Began

A biography of the Australian continent 

Palaeoproterozoic Ice Houses - Evolution Oxygen-Mediating Enzymes, Late Origin of Photosystem II

According to the authors1 with regard to the origin of oxygenic photosynthesis there are 2 major geological problems. One problem is the identification of an oxygen source that predates the production of biological oxygen and is capable of driving the evolution of oxygen tolerance. The other problem is the determination of when oxygenic photosynthesis evolved. The accumulation of H2O2 that was produced photochemically at the surface of glaciers and subsequently incorporated into ice is a solution to the first problem. H2O2 would be released by melting at the base of the glacier, O2 would then be produced when the  H2O2 interacts with seawater in an environment that is shielded from the lethal levels of UV radiation that is required in the production of  H2O2. With regard to the second problem, the answers that have been proposed have ranged from 3.8-2.02 Ga. A view that the authors1 describe as sceptical is based on the metals with a redox potential close to that of oxygen, argues for the late end of the range. The weight of geological evidence suggests little, less than 1 ppm, or no atmospheric oxygen in the Late Archaean. Lipid biomarkers are the main evidence suggesting an earlier evolution of oxygenic photosynthesis, though it has been shown by more recent work that 2-methylhopanes, previously believed to be unique biomarkers of cyanobacteria, are also produced in significant quantities by at least 2 strains of anoxygenic phototrophs. The strongest evidence for an age of 2.7 Ga or more is provided by sterane biomarkers, which could also be provided by the common evolutionary pattern of anaerobic enzymes being replaced by oxygen-dependent ones. Enzymes that perform the necessary chemistry are known to exist at the present, though in the biosphere of the present no sterol synthesis pathway has been identified. Oxygenic photosynthesis has been suggested by the analysis carried out by the authors1 could have evolved close in geological time to the Makganyene Snowball Earth Event and suggests there is a causal link between the 2.

According to the authors1 the origin of the atmospheric oxygen is probably the longest-running debate that has yet to be solved in the history of modern science. The debate began in 1845 when Embelman and 1854 Bischof were responsible for some of the earliest publications in biochemistry regarding the balance of oxygen release by carbon and pyrite burial and the consumption of oxygen by the oxidation of iron and manganese, speculating that the levels of oxygen in the atmosphere might have changed with time and changes in biota (Berner & Maasch, 1996). It was argued in 1856 (Koene, 1856/2004) that the initial atmosphere of the Earth had high concentrations of carbon dioxide but had no oxygen and the photosynthetic activity of plants had resulted in a carbon dioxide decline and an increase in oxygen levels. It wasn't until the late 19th century that follow-up publication began to appear (See these reviews Stevenson, 1900; van Hise, 1904; Clarke, 1924).

Evolution of oxygen-mediating enzymes and glacial peroxides

A severe threat to organisms living in an aerobic environment is the ability of dissolved O2 to accept an electron from a suitable donor, such as Fe+2, to form O-2 the superoxide radical, that spontaneously reacts with water to form the hydroxyl radical .OH, which then attacks the sugar/phosphate backbone of DNA. The authors1 suggest  the evolution of O2-mediating enzymes, such as superoxide reductase, was fostered by the need to control this toxic process (Jenny et al., 1999) and superoxide dismutase that converts  O-2 to H2O2 (Wolfe-Simon, 2005), catalase, that converts hydrogen peroxide to water, and various domains that bind oxygen, such as haem, that help stop the formation of superoxide (Niyogi, 1999). Prior to the metabolic use of oxygen, including oxygen production by the oxygen-evolving complex of PSII, it was necessary for oxygen mediating pathways to evolve. It appears the oxygen-evolving complex derives oxygen from the manganese cluster of catalase, which also implies that the enzymes that mediate oxygen came first (MacKay & Hartman, 1991; Blankenship & Hartman, 1998).

The authors1 suggest there must have been a source of superoxide radicals to apply an adaptive pressure for oxygen-mediating pathways to develop, in the atmosphere and hydrosphere of the Earth most geochemical and geological processes are relatively reducing. Redox reactions on the reducing end of the spectrum, that are more reducing than the ferrous-ferric couplet and far from the water/oxygen couplet that is oxidising,  buffer gases emitted from volcanoes. The only known pathway for producing concentrations of molecular oxygen is the photolytic reactions that involve UV radiation and water vapour. The same UV radiation is destructive to complex organic molecules, such as DNA, though the environment under which this process can occur are lethal to all living organisms.

It has been noted that interannual variations in hydrogen peroxide concentrations which reflect the history of the ozone hole over Antarctica (Frey et al., 2005, 2006). Decreases of ozone concentrations in the stratosphere facilitate photochemical reactions that involve water and generate hydrogen peroxide and hydrogen gas, as ozone is the major filter of UV radiation of short wavelength. The hydrogen peroxide condenses and accumulates in the ice, as it has a freezing point near that of water, and the hydrogen diffuses away into the atmosphere.

The Pongola Glaciation of the Late Archaean and the glaciations of the Early Palaeoproterozoic all occurred in atmospheres in which there was little oxygen, and with no ozone screen, the UV radiation was free to penetrate to the surface of the earth that allowed it to produce the mass-independent fractionation (MIF) that is seen in sulphur compounds. The same photochemical processes that act in Antarctica at the present would have acted over entire ice sheets in these ancient glaciations to cause the accumulation of hydrogen peroxide. Any hydrogen peroxide that was produced in the oceans, as it was produced in the ancient atmospheres that were very low in oxygen, it would have simply diffused away and not accumulate. Snow that was laced with hydrogen peroxide would follow normal glacial dynamics in a normal, non-Snowball Earth glaciation, being compressed into glacial ice, being carried by the glaciers to the ocean where it melted , or along the wet-base portions. When the ice melted hydrogen peroxide would disproportionate into oxygen and water (2H2O2 → 2H2O + O2) leading to the protection of the environment from UV radiation, though 'poisoned' with oxygen at trace amounts, the environmental conditions that could be expected to encourage the evolution of enzymes that could mediate the oxygen, which would allow any cells possessing the enzyme to survive in such an atmosphere. It has been reported recently (Anbar, 2007; Kaufman, 2007) from 2 localities on the palaeocontinent of Vaalbarra at about 2.5 Ga that have been suggested to the geochemical fingerprint of the mixing of oxic meltwater and glacial flour, the powdered rock found at the base of a polar icefield. The full span of the Late Archaean and the glacial epochs of the Palaeoproterozoic is not known, as the preservation of glacial deposits depends on a number of factors, though the oldest unit in the Palaeoproterozoic that have been confirmed dates to sometime after 2.45 Ga (Evans, 2000; Young, 2001). These necessary factors include continents being in the correct position, enough accommodation space, and the fate of deposits through geological time. The authors1 suggest that in the time interval between the glacial deposits of the Pongola and  Huronian Glaciations there may have been glacial icehouse conditions similar to those of the Late Palaeozoic and Cainozoic.

Lipid biomarkers from the Archaean - are they contaminants?

According to the authors1 a way of constraining the history of oxygen is to search in sediments for biomarkers that are the products of reactions that are oxygen-dependent. There were 473 biochemical reactions in which molecular oxygen, O2, was a substrate that were known of by 2004 (Raymond & Blankenship, 2004; Raymond & Segre, 2006). It is necessary to verify if the materials under study are of the same age as the rocks, as is the case with any geobiological tracer, The authors1 say that in the 1960s and early 1970s many new techniques were devised for use as geochemical studies of traces in rocks from the Precambrian of porphyrins, fatty acids, alkanes, acyclic isoprenoids, and amino acids (Hayes et al., 1983; Brocks et al., 2003a). Almost all the organic compounds were subsequently found to be hydrocarbons derived from petroleum, from either lab or fields environments. As a result the authors1 claim the workers in this particular field have become the most severe critics of their own work.

There has been much progress on identifying the proper target molecules for study and elucidating the diagenetic processes that operate on them (Knoll et a., 2007). It has been found that by far the best is the lipid fraction from biological membranes, because the topology of their skeletal backbones can be preserved by these lipids, in spite of various changes occurring during diagenesis, such as hydroxyl groups being lost and replaced, degradation of linear side chains and double bond saturation. As potential constraints on oxygen history the degradation products of isoprenoids, especially hopanols and sterols, are of particular importance, preserving enough of their structural information to allow identification of the biochemical pathways by which they were  formed. It is rare, though not unknown, for the same isoprenoid compounds to be produced via different pathways. The main focus of a series of papers (Brocks et a;l., 1999, 2003a b; Summons et al., 1999, 2006); have been the fossil record of hopanols and sterols from the Precambrian. The authors1 recommend 2 excellent reviews of the biosynthesis of sterols and the structure and function of the family of triterpene cyclase enzymes (Lesberg et al., 1998 and Wendt et al., 2000) and for the oxygen dependence for sterol synthesis in yeast (Rosenfeld & Beauvoit, 2003).

According to the authors1 vastly superior techniques and internal controls for reproducibility were used in this body of work than those used by those of the 1960s that have since been discredited. According to the authors1 the age of the molecules containing the biological information has not been resolved completely, resulting in concerns about contamination remain (Brocks et al., 2003b, p.4322), they therefore suggest researchers should be cautious about citing palaeobiological interpretation with reference to the remaining uncertainty of syngeneity. Biomarkers that are characteristically Phanerozoic, such as dinosteranes, that are generally associated with dinoflagellates, are a taxon from the Mesozoic-Cainozoic, as well as no trace of any stages that were geologically intermediate in evolutionary succession in the long sterol biosynthetic pathway, are, as well as others, considered to be possible 'red flags'. Gemmata obscuriglobus, extant planctomycete, has been identified (Pearson et al., 2003), that possesses just such an intermediate sterol synthesis pathway that has been suggested to possibly be on the ancestral eukaryote lineage (Fuerst, 2005), though it is not known if the pathway is an immature relict or has been truncated. Whatever the case, it has been noted (Brocks, 2005) that changes in lipid biomarkers should be similar to the incremental pattern in the fossil record from the Proterozoic, but in the Archaean, sterols have been found that have a fully modern fingerprint almost a billion years prior to the first known fossils that possibly have an affinity with eukaryotes. Examination of hydrocarbons in fluid inclusions, among other new analytical methods (Dutkiewicz et al., 2006; Ueno et al., 2006) are suggested by the authors1 to possibly soon resolve this contamination controversy.

Formation of hopane and sterol - Is molecular oxygen a fundamental requirement?

Verification is important that oxygen is required by the fundamental chemistry involved in providing strong restraints on the levels of palaeo-oxygen, precluding any possible anaerobic mechanisms, and that the basic processes, that are the basis of the references, have not changed over geological time. The authors1 argue that the case for the linking of biosynthetic pathways leading to hopanol and sterol to the presence of oxygenic photosynthesis has been considerably weakened by recent work, even if the pathways evolved during the Hadean or Archaean. The 4 lines of molecular evidence for free molecular oxygen in the surface waters have been summarised (Brocks et al., 2003b, p.4331).

These 4 lines of molecular evidence of free molecular oxygen in the surface waters are:

Molecular fossils of bacteriohopanoids are contained in the bitumens. These lipids have not been isolated from strict anaerobes (Ourisson et al., 1987), though the biosynthesis of hopanoid does not require oxygen.

Cyanobacterial 2-methylhopanes are found at relatively high concentrations in shales from then Archaean. These biomarkers are indirect evidence that oxygen release is occurring within the photic zone.

Oxic conditions during earliest diagenesis of cyanobacterial organic matter is indication of the pattern of degradation of the side-chain of the 2-methylhopane series.

Molecular fossils of sterols are present in bitumens. Dissolved oxygen at concentrations equivalent to 1 % PAL, is required in the biosynthesis of sterol in extant eukaryotes (Jahnke & Klein, 1983).

It has been suggested by recent work that mechanisms not requiring free oxygen in the environment can explain all of the above 4 lines.

Bacteriohopanoids and 2-methylhopanes

Molecular oxygen is not involved in production of hopanpiods (Brocks et al., 2003b) and multiple anaerobic bacteria that produce hopanoids have been identified by subsequent work. It has been found that at least 1 bacterium, Geobacter sulphurreducens, can synthesise diverse hopanoids, though not 2-methylhopanoids, when grown under conditions that are strictly anaerobic (Fischer et al., 2005). It has been reported that large amounts of 2-methylhaponoids are produced by Rhodopseudomonas palustris under anaerobic conditions that mimic those that are presumed to have existed in the Archaean (Rashby el at., 2007). The authors1 include a quote from Rashby et al. (p. 15 102), '... because 2-MeBHPs may be produced by organisms that do not engage in oxygenic photosynthesis and because their biosynthesis does not require molecular oxygen, 2-methylhopanes cannot be used as de facto evidence for oxygenic photosynthesis'.

According to a quote included by the authors1 (Brocks et al., 2003b, p.4330) it is suggested by the pattern of side-chain degradation of C30 to C36 2-methylhopane homologous series, relative to the corresponding C30 to C36 3-methylhopane that in the Late Archaean cyanobacteria lived in an oxygenated micro-environment. In all the samples of C30 to C36 2-methylhopane that have been analysed there is a characteristic carbon number predominance of even-over-odd. Oxidative side-chain cleavage of the bacteriohopanetetrol to a C33 carboxylic acid and subsequently decarboxylation under conditions that were non-reducing is indicated by the elevated abundance of the C32-homologue relative to the C31 and C33.

Anaerobic processes are suggested by the authors1 to being capable of explaining the predominance of the even-over-odd carbon number of 2-methylhopanes relative to 3-methylhopanes. An anaerobic sulphate reducing  bacterial strain that metabolises straight-chain saturated alkanes has been isolated from a sediment that was contaminated by petroleum with a similar 2-carbon removal process (So & Young, 1999). When the bacteria were grown on pure Ceven alkanes the carbon number of the fatty acids they formed were predominantly even, but if grown on Codd alkanes their total fatty acids had a predominantly odd number of carbon atoms. This example is from alkane degradation, and the authors suggest similar chemistry should work for isoprenoid side chains.  It is clear that the anaerobic organisms have a number of different biochemical pathways that involve the addition of small carbon compounds that can produce such patterns of even/odd for which the complexity and mechanism has yet to be resolved (see Aeckersberg et al., 1998; Berthe-Corti & Fetzner, 2002; and references therein, though the side-chains of 2- and 3-methylhopanes would be equally affected by post-depositional degradation.


Conclusions of the authors1

It has been noted that the production and accumulation of hydrogen peroxide on the surface of ice of the polar ice caps (Liang et al., 2006) provides a straightforward mechanism for producing trace, though substantial, concentrations of hydrogen peroxide and free oxygen in the water beneath melting glaciers, in a situation where they are protected from the destructive effects of UV radiation. The evolution of oxygen-mediating enzymes could possibly be driven in these conditions, which would therefore pave the way for the evolution of oxygen-evolving clusters of PSII. According to the authors1 they found no chemical requirement for the presence of molecular oxygen in the biosynthesis of lipid biomarkers that have been presumed to be of Archaean age, that cannot be satisfied  by known anaerobic biochemical mechanisms. They also suggest intensive evolutionary pressure is applied for the anaerobic enzymes, that perform steps that are analogous to steps performed by those requiring oxygen using cofactors of redox-sensitive metals that are poisoned by oxygen (Imlay, 2006), to swap them for others that are not sensitive to oxygen. In the biosynthesis of chlorophyll 3 such substitutions in 17 enzymatic steps are known of, the authors1 argue that 4 substitutions in a sterol synthesis pathway having 25 steps are not unreasonable.

In the Makganyene Snowball Earth and the deposition of the Kalahari Manganese Field is correlated with the evolution of the PSII complex that releases oxygen, as measured by environmental redox indicators. The authors1 suggestthat the destruction of reduced greenhouse gases, such as methane (Pavlov et al., 2000), did not take 400 My, though it was sufficiently rapid to trigger the Makganyene Snowball (Kopp et al., 2005).

See source1 for more information.

Sources & Further reading

  1. Kirschvink, Joseph L, and Robert E Kopp. "Palaeoproterozoic Ice Houses and the Evolution of Oxygen-Mediating Enzymes: The Case for a Late Origin of Photosystem Ii." Philosophical Transactions of the Royal Society B: Biological Sciences 363, no. 1504 (August 27, 2008 2008): 2755-65.
Author: M. H. Monroe
Last updated  03/06/2013
Journey Back Through Time
Experience Australia
Aboriginal Australia
National Parks
Photo Galleries
Site Map
                                                                                           Author: M.H.Monroe  Email:     Sources & Further reading