aerobic metabolism is impossible, and there is limited atmospheric protection against ultraviolet (UV) radiation that destroys DNA (Kasting, 1987). The rate of nitrate production in the atmosphere by lightning may well have been an order of magnitude lower than today (Yung and McElroy, 1979; Levine et al., 1982); however, the difference between the present-day rate of nitrate production (Borucki and Chameides, 1984) and the rate 2500 to 2000 Ma depends on the CO2 pressure in the Paleoproterozoic atmosphere. Kasting (1990) proposed that at a PCO2 of 0.2 atmosphere (atm), the NO production rate in the absence of O2 is only about a factor of two lower than at present. At a CO 2 pressure of 0.02 atm (i.e., 60 PAL), the NO fixation rate would probably be only one-tenth of the present rate. Denitrification was probably more widespread and intense in a low-O2 ocean than in the present ocean. Therefore, NO3- was almost certainly in very short supply prior to 2100 Ma, and biological N2 fixation must have been the principal source of usable nitrogen for primary producers. H2O 2 may have been an important oxidant on the early anoxic Earth (Kasting et al., 1987); it is possible that biochemical defenses against molecules generally regarded as reactive intermediates in oxygen biochemistry evolved before O2 itself became a significant constituent of the atmosphere (McKay and Hartman, 1991).
The conditions described in the previous paragraph certainly apply to the biota that existed before the evolution of cyanobacterial photosynthesis. How long such conditions persisted after the advent of oxygenic photosynthesis is unclear. As noted above, the antiquity of oxygenic cyanobacteria is poorly constrained, although it could easily be as great as 3500 Ma, the age of the oldest negligibly metamorphosed sedimentary rocks (Knoll, 1979; Schopf and Packer, 1987). As noted previously, a PO2 of approximately 1 to 2% PAL appears likely for an extended period prior to 2100 Ma (see also Towe, 1990). This is an oxygen level of both biological and environmental significance. At about 1% PAL, aerobic metabolism by single-celled organisms becomes possible, while an effective ozone screen expands the ecological possibilities of life. When PO2 rose to 1 to 2% PAL, aerobic metabolism probably followed quickly in organisms already protected against oxygen toxicity. In particular, bacteria capable of aerobic respiration, with its tremendous energetic advantage over fermentation, probably radiated rapidly (and polyphyletically) from photosynthetic ancestors. Flavin-based oxygen-utilizing pathways evolved in archaebacteria and in amitochondrial eukaryotes. At this O2 level, nitrate production levels in the atmosphere may well have been significantly lower than today's (see above). Nitrogen fixers, therefore, could have retained a considerable advantage in primary production.
Ancestral eukaryotes formed endosymbiotic associations with purple bacterial aerobes, gaining the benefits of aerobic respiration. Symbioses with photosynthetic prokaryotes may not have formed concurrently, however. Nitrogen fixation is unknown in plastids and appears to be prohibited (Postgate and Eady, 1988); the reasons for this are not clear, but may involve oxygen toxicity. Although some early algae might have obtained nitrogen heterotrophically, obligately photosynthetic eukaryotes (including all extant megascopic algae) are unlikely to have occurred in the absence of significant quantities of nitrate in the environment.
These considerations suggest a different biological focus for the 2100 Ma oxygen event. It is not that fundamentally new metabolisms were made possible, but rather that as oxygen increased to levels above 10% PAL, nitrate availability may well have increased dramatically (see above). Obligately photosynthetic eukaryotes would then have become feasible. With their ability to avoid formation of nutrient-depleted boundary layers adjacent to cells, eukaryotic primary producers would soon have become ecologically important as primary producers. Thus, it is not surprising that 2100 Ma shales contain megascopic algae or that slightly younger rocks contain abundant acritarchs whose morphology and distribution are similar to those of younger eukaryotic phytoplankton.
How does this environmental scenario compare with the known phylogeny of eukaryotes? Figure 1.5 shows evolutionary relationships among living eukaryotes as determined by Sogin et al. (1989). At the base of the tree is Giardia, a common pathogen in the digestive system of vertebrates. Biochemically, Giardia shares more features with prokaryotes than any other known eukaryote. Ultrastructurally, however, it is clearly a true eukaryote; it contains a membrane-bounded nucleus, undulipodia (9+2 flagella), and a cytoskeleton (albeit a biochemically very simple one). On the other hand, Giardia has no mitochondria and no well-developed ER or Golgi apparatus. These organisms are heterotrophic, engulfing particulate food (phagocytosis) and absorbing dissolved organic molecules. Food is metabolized by the classic Embden-Meyerhoff pathway of glycolysis. Giardia cells are not capable of classical aerobiosis, but can use oxygen as a terminal acceptor of reducing equivalents. This system uses flavins and iron-sulfur proteins, and does not include cytochromes; it appears that the cells derive little energetic benefit from this reaction (Müller, 1988).
The next branches in Figure 1.5 are occupied by the microsporidia, trichomonads, and related protists. Both groups have clearly become specialized as obligate parasites (microsporidians are apparently dependent on an external source of ATP), but they retain features that complement the picture of early eukaryotes developed from Giardia.