AME 83:189-201 (2019)  -  DOI: https://doi.org/10.3354/ame01911

ABSTRACT: The present study compares different methods for calculating hypolimnic denitrification in the stratified water column of Lake Scharmützelsee (Germany) during summer stratification. Denitrification rates were estimated either from nitrate loss under anoxic conditions or with a Michaelis-Menten-based approach. The latter included correction terms to account for temperature variations and oxygen inhibition. For this kinetic approach, 2 critical parameters (cell count and maximum turnover rate) are needed. As gene copy numbers obtained by qPCR do not distinguish between viable and non-viable cells, most probable number (MPN)-based cell counts were also tested for their suitability in the Michaelis-Menten approach. At the beginning of the stratification period, about 1% of the denitrifiers calculated from nirS gene copy numbers were cultivable. Nitrate loss and calculated qPCR-based denitrification rates differed by only 1 order of magnitude (mean rates 40.74 and 1.27 µmol l^-1 d^-1, respectively). During the stratification period, the difference between qPCR-based calculated denitrification rates and measured nitrate loss increased by up to 4 orders of magnitude, resulting in an enormous overestimation of the calculated denitrification rates. Calculations based only on the numbers of cultivable cells from the MPN counts were significantly lower than the nitrate loss in the lake. Neither nirS gene copy numbers nor MPN counts were suitable for determining the real number of metabolically active denitrifiers. Therefore, additional methods (e.g. propidium monoazide-PCR, viability staining or model strains) are needed to identify the real portion of metabolically active cells.

KEY WORDS: Denitrification · Freshwater · Hypolimnion · Viable but non-culturable state · qPCR · Most probable number