Validation of heavy-water stable isotope probing for the characterization of rapidly responding soil bacteria

Abstract

Rapid responses of bacteria to sudden changes in their environment can have important implications for the structure and function of microbial communities. In this study, we used heavy-water stable isotope probing (H2(18)O-SIP) to identify bacteria that respond to soil rewetting. First, we conducted experiments to address uncertainties regarding the H2(18)O-SIP method. Using liquid chromatography-mass spectroscopy (LC-MS), we determined that oxygen from H2(18)O was incorporated into all structural components of DNA. Although this incorporation was uneven, we could effectively separate 18O-labeled and unlabeled DNAs derived from laboratory cultures and environmental samples that were incubated with H2(18)O. We found no evidence for ex vivo exchange of oxygen atoms between DNA and extracellular H2O, suggesting that 18O incorporation into DNA is relatively stable. Furthermore, the rate of 18O incorporation into bacterial DNA was high (within 48 to 72 h), coinciding with pulses of CO2 generated from soil rewetting. Second, we examined shifts in the bacterial composition of grassland soils following rewetting, using H2(18)O-SIP and bar-coded pyrosequencing of 16S rRNA genes. For some groups of soil bacteria, we observed coherent responses at a relatively course taxonomic resolution. Following rewetting, the relative recovery of Alphaproteobacteria, Betaproteobacteria, and Gammaproteobacteria increased, while the relative recovery of Chloroflexi and Deltaproteobacteria decreased. Together, our results suggest that H2(18)O-SIP is effective at identifying metabolically active bacteria that influence soil carbon dynamics. Our results contribute to the ecological classification of soil bacteria while providing insight into some of the functional traits that influence the structure and function of microbial communities under dynamic soil moisture regimes.

Fig. 1.

Mass spectra of unlabeled (¹⁶O) and labeled (¹⁸O) dTMP in actively growing E. coli. Bars represent the relative intensities of dTMP ions with various mass-to-charge (m/z) ratios from LC-MS. Relative intensity was based on the number of ions in each m/z ratio relative to the ratio with the largest number of ions detected for unlabeled (6.31e³) or ¹⁸O-labeled (1.23e³) dTMP ions. (B) Diagram of dTMP with an m/z of 335, highlighting the possible locations of ¹⁸O.

Fig. 2.

¹⁸O-DNA SIP of unlabeled (¹⁶O) and labeled (¹⁸O) E. coli rRNA genes before (A) and after (B) a 7-day incubation in H₂¹⁶O. Values are means ± SEM for the ratio of maximum quantities of rRNA gene copies (n = 3) within CsTFA gradients. Ratios are based on the number of gene copies in each buoyant density fraction relative to the buoyant density fraction that contained the maximum number of 16S rRNA gene copies.

Fig. 3.

Rapid incorporation of ¹⁸O from H₂¹⁸O into bacterial DNA following rewetting of dried soils. Panels show the resulting distributions of 16S rRNA genes from replicate (n = 3) grassland soils in CsTFA gradients. Soils were incubated with H₂¹⁸O for 0 h (A), 48 h (B), and 72 h (C). Values are mean ratios of maximum quantities of rRNA genes (n = 3) ± SEM from qPCRs.

Fig. 4.

Taxonomic responses to soil rewetting. Values are mean changes (n = 3) in relative recoveries of taxa after rewetting (72 h) and before rewetting (0 h) with H₂¹⁸O for phyla or classes (A) and families (B). Asterisks denote taxonomic groups that responded significantly to rewetting based on the overlap of 95% confidence intervals with zero change in relative recovery.