Optimization of viral resuspension methods for carbon-rich soils along a permafrost thaw gradient

Abstract

Permafrost stores approximately 50% of global soil carbon (C) in a frozen form; it is thawing rapidly under climate change, and little is known about viral communities in these soils or their roles in C cycling. In permafrost soils, microorganisms contribute significantly to C cycling, and characterizing them has recently been shown to improve prediction of ecosystem function. In other ecosystems, viruses have broad ecosystem and community impacts ranging from host cell mortality and organic matter cycling to horizontal gene transfer and reprogramming of core microbial metabolisms. Here we developed an optimized protocol to extract viruses from three types of high organic-matter peatland soils across a permafrost thaw gradient (palsa, moss-dominated bog, and sedge-dominated fen). Three separate experiments were used to evaluate the impact of chemical buffers, physical dispersion, storage conditions, and concentration and purification methods on viral yields. The most successful protocol, amended potassium citrate buffer with bead-beating or vortexing and BSA, yielded on average as much as 2-fold more virus-like particles (VLPs) g(-1) of soil than other methods tested. All method combinations yielded VLPs g(-1) of soil on the 10(8) order of magnitude across all three soil types. The different storage and concentration methods did not yield significantly more VLPs g(-1) of soil among the soil types. This research provides much-needed guidelines for resuspending viruses from soils, specifically carbon-rich soils, paving the way for incorporating viruses into soil ecology studies.

Keywords: Active layer; Humic-laden; Microbial ecology; Peatland; Permafrost; Phages; Soil viruses; Viral diversity; Viral ecology; Viral methods.

Figure 1. Optimizing viral recovery.

Three experiments tested resuspension, storage, and purification conditions for viral recovery on a range of peatland soils from this thawing permafrost site. Red font color indicates the best-performing option within each set. EFM, epifluorescence microscopy; TEM, transmission electron microscopy; BSA, bovine serum albumin. Experiment 1 used peatland samples from 2013 and Experiments’ 2 & 3 used peatland samples from 2014. Palsa and bog pictures contributed by Anthony Garnello.

Figure 2. The impact of buffers and physical dispersion methods on viral yields.

(A) Viral yields from different buffers (Experiment 1A). PP, sodium pyrophosphate; KC, potassium citrate; AKC, amended potassium citrate. Each treatment was followed by sonication for physical dispersion. (B) Viral yields from different physical dispersion methods (Experiment 1B). VX, vortexing; BB, bead-beating; SC, sonication. Each replicate was in AKC buffer. An ^∗ denotes statistically significant difference (p < 0.05) within the soil type.

Figure 3. The impact of storage conditions on viral yields for different depth soils.

Samples from shallow (1–5 cm; A) or deep (36–40 cm; B) depths, and stored frozen (flash frozen and kept at −80 °C) or chilled (4 °C), were counted using epifluorescence microscopy.

Figure 4. The impact of BSA on virus concentration by Amicon filter.

Deep samples, stored frozen (A) or chilled (B), were used for this comparison and post concentration yields were counted by epifluorescence microscopy. An ^∗ denotes statistically significant (p < 0.05) within the soil type.

Figure 5. Comparison of samples with or without BSA after CsCl gradient purification.

Counts were taken of deep samples that were CsCl gradient purified after Amicon filter concentration (with and without BSA) for samples from each habitat that were (A) frozen or (B) chilled.