An Experimental Study on What Controls the Ratios of 18O/16O and 17O/16O of O2 During Microbial Respiration

Abstract:

¹⁸O/¹⁶O and ¹⁷O/¹⁶O ratios of atmospheric and dissolved oceanic O₂ are key biogeochemical tracers of total photosynthesis and respiration on global to local length scales and glacial/interglacial time scales (Luz et al., 1999). Critical to the use of these ratios as biogeochemical tracers is knowledge of how they are affected by production, consumption, and transport of O₂. We present new measurements of O₂ respiration by E. coli and N. oceanus, an ammonia oxidizing bacterium, to test three assumptions of isotopically enabled models of the O₂ cycle: (i) laboratory-measured respiratory ¹⁸O/¹⁶O isotope effects (¹⁸α) of microorganisms are constant under all experimental and natural conditions (e.g., temperature and growth rate); (ii) the respiratory ‘mass law’ relationship between ¹⁸O/¹⁶O and ¹⁷O/¹⁶O [¹⁷α = (¹⁸α)^β] is universal; and (iii) ¹⁸α and β for aerobic ammonia and organic carbon oxidation are identical.

For E. coli, we find that both ¹⁸α and β are variable. From 37°C to 15°C, ¹⁸α varies linearly with temperature from 17 to 14‰, and β varies linearly from 0.513 to 0.508. ¹⁸α and β do not appear to vary with growth rate (as tested using different carbon sources). Both ¹⁸α and β are lower than previous observations for bacteria: ¹⁸α = 17-20‰ (Kiddon et al., 1993) and β = 0.515 (Luz and Barkan, 2005). We were able to simulate the observed temperature dependence of ¹⁸α and β using a model of respiration with two isotopically discriminating steps: O₂ binding to cytochrome bo oxidase (the respiratory enzyme) and reduction of O₂ to H₂O. Finally, initial results on N. oceanus suggest it has similar values for ¹⁸α and β as previously studied aerobic bacteria that consume organic carbon, providing the first support for assumption (iii).

Based on these results, isotopically constrained biogeochemical models of O₂ cycling may need to consider a temperature dependence for ¹⁸α and β for microbial respiration. For example, these results may explain why the cold, deep ocean has lower obeserved ¹⁸α values (~8‰) than that for warmer surface waters (~19‰; Levine et al., 2009).