A popular hypothesis among microbial ecologists is that denitrifying communities in soils are different regarding their ability to achieve a balanced transition from oxic to anoxic conditions, and that their propensity to emit N2O can be predicted by their genetic "fingerprint". The most naive version of this hypothesis is that the propensity to emit N2O can be predicted by gene abundance, i.e. that soils which emit much N2O have high nir/nosZ gene abundance and vice versa. Such hypotheses are commonly tested by searching for correlations between gene abundance data and emissions (or the N2O/N2 product ratio of denitrification. No concistent patterns have emerged, however: the numerous correlations found appear to be utterly spurious.
We believe that regulatory biology is a clue to understand the role of denitrifying communities as sources and sinks for N2O in the environment, and that stringent investigations of model organisms can enhance our understanding of how N2O emissions are regulated, hence strengthen our ability to develop mitigations, i.e. actions that reduce the emissions of N2O.
Our investigatins of Paracoccus denitrificans have revealed bet hedging response to oxygen depletion: all cells express nosZ (N2O reductase), while only a minority express NirS. As a result, the population becomes a strong sink for N2O. The phenomen appears to be widespread among denitrifying organisms (papers are in the pipeline).
Our investigation of nosZ-expression in Pa. denitrificans showed that low pH hampers this expression by interering with the maturation of the enzyme in the periplasm. This could be an anecdote for Paracoccus, but this is clearly not the case: All microbial communities tested so far show exactly the same post transcriptional effect of low pH on the making of N2O reductase! We are currently exploring in more detail what goes wrong at low pH. But the results so far suffice to claim that we know why soil pH has a pervasive effect on the N2O/N2 product ratio of denitrification.
A common notion is that N2O reductase is destabilized/destroyed by oxygen. Our experiments with Pa denitrificans show that this is not the case. During aerobic growth, cells with N2O reductase (expressed during anoxic spell) will preserve the enzyme (but diluting it if growing) during aerobic growth. We have also demonstrated that they preserve nitrite reductase during aerobic growth. This explains the "legacy of anoxia" often observed in soil: once exposed to anoxia, the soils responds faster to subsequent anoxic spells.
A long standing notion is that NOR (the enzyme reducing NO to N2O) is a necessity for survival as a denitrifying organism. This is certainly true when grown in pure culture (no organisms can survive for long in the presence of micromolar concentrations of NO). Nevertheless, denitrifying organisms without NOR have been isolated (Falk et al 2010, see publication list), and more will be published on this in the near future. Thus, it appears perfectly possible to survive as a denitrifier without NOR, probably because NO toxicity is not a problem under natural conditions (NO scavenged by others). The "Black queen hypothesis" seems relevant here...
Several observations of communities and pure cultures have convinced us that post- transcriptional regulation of electron flow to either O2 or NOx is a clue to understand N2O emissions from soils. Transcription of the denitrification genes appears to be universally repressed by oxygen, but aerobic denitrification is a significant phenomenon, provided that the denitrification proteome has been expressed during foregoing anoxia.
The "post anoxic trauma" is a term used for a community exposed to oxygen after an anoxic spell. The figure above shows that Thauera aminoaromatica is unable to stop denitifying if given a sudden dose of oxygen. The same phenomenon was deomstrated for a denitrifying community by Morley et al 2008 (see publication list). In both cases, nitrous oxide reductase activity was apparently inhibited by oxygen. Our understanding of this phenomenon has deepened over the last years: it appears that the "inhibition" is primarily a "deprivation": that the terminal oxidases compete successfully with N2O reductase for electrons. In contrast, NIR and NOR appear to be more competitive vi a vis terminal oxidases.