A biomolecular isolation framework for eco-systems biology

Abstract

Mixed microbial communities are complex, dynamic and heterogeneous. It is therefore essential that biomolecular fractions obtained for high-throughput omic analyses are representative of single samples to facilitate meaningful data integration, analysis and modeling. We have developed a new methodological framework for the reproducible isolation of high-quality genomic DNA, large and small RNA, proteins, and polar and non-polar metabolites from single unique mixed microbial community samples. The methodology is based around reproducible cryogenic sample preservation and cell lysis. Metabolites are extracted first using organic solvents, followed by the sequential isolation of nucleic acids and proteins using chromatographic spin columns. The methodology was validated by comparison to traditional dedicated and simultaneous biomolecular isolation methods. To prove the broad applicability of the methodology, we applied it to microbial consortia of biotechnological, environmental and biomedical research interest. The developed methodological framework lays the foundation for standardized molecular eco-systematic studies on a range of different microbial communities in the future.

Figure 1

The developed biomolecular isolation framework highlighting the NA-based methodological workflow. Legend: () Sample processing and preservation: immediate snap-freezing by immersion in liquid nitrogen for LAO-enriched mixed microbial communities; concentration by tangential flow filtration followed by high-speed centrifugation, then snap-freezing of the resulting cell pellet for freshwater mixed microbial communities; homogenization with RNAlater followed by centrifugation steps before snap-freezing for fresh human fecal samples (Materials and methods). () Cryomilling and metabolite extraction: cryomilling of cell pellets and solvent extraction of the intracellular polar and non-polar metabolite fractions (extracellular fractions were only prepared for the LAO-enriched microbial communities; Materials and methods). () Physicochemical biomacromolecular isolation: use of sequential physicochemical separation based around chromatographic spin columns following bead-beating in the modified NA-lysis buffer, resulting in the isolation of high-purity biomacromolecular fractions (Materials and methods).

Figure 2

Metabolome heterogeneity within LAO-enriched microbial community samples. (a) Scatter plot of the two first principal components obtained using principal component analysis (PCA) of the normalized metabolomics data derived from the four biological replicates (islets; I1-I4), for each of the four distinct sampling dates (D1-D4). Each microbial community metabolome is indicated by a dot and color-coded according to sampling date. (b) Between-class PCA of the individual technical replicates for each biological replicate (islets; I1-I4). (a and b) The center of gravity for each date/islet cluster is marked by a rectangle and the colored ellipse covers 67% of the samples belonging to the cluster. (c) Hierarchical clustering of the normalized metabolomics data using the Pearson product moment correlation coefficient.

Figure 3

Efficiencies of different cell lysis methods. (a–c) Representative epifluorescence micrographs of microbial cells from a representative LAO-enriched microbial community sample stained with the Live/Dead stain (intact cells highlighted in green, disrupted cells with a compromised cell membrane in red). Scale bar is equivalent to 10 μm. (a) Sample having undergone a single freeze-thaw cycle. (b) Sample having undergone polar and non-polar metabolite extractions. (c) Sample having undergone the additional mechanical and chemical lysis step using the NA kit's modified lysis buffer. (d) Bar chart highlighting the percentages of cells disrupted by the different treatments (n=10; error bars represent s.d.). x-axis legend: FT, sample having undergone a single freeze-thaw cycle, reflecting panel a; NA, sample having been subjected to cell lysis in the modified NA lysis buffer, reflecting panel c; QA, QA lysis buffer; TR, TR-based lysis; M, sample after polar and non-polar metabolite extractions reflecting panel b; RM-A, sample having been subjected to the dedicated DNA extraction reference method; RM-B, dedicated RNA extraction method; RM-C, dedicated protein extraction method.

Figure 4

Quality and quantity of biomacromolecular fractions isolated from the representative LAO-enriched microbial community sample using either the NA-, QA- and TR-based method (following prior metabolite extractions) or using the reference methods (no metabolite extractions were carried out before the respective extractions). (a and b) Representative Agilent Bioanalyzer 2100 electropherograms of the total RNA and small RNA fractions, respectively. (c) Agarose gel image highlighting representative genomic DNA fractions obtained (Mean amount (n=3) loaded in μg±s.d., from right to left; NA: 0.35±0.17; QA: 0.35±0.08; TR: 0.83±0.85; RM: 0.08±0.04) and (d) SDS-PAGE image of representative protein fractions (Mean amount (n=3) loaded in μg±s.d., from right to left; NA, first elution: 3.20±0.19; QA: 5.44±1.06; TR: 3.88±0.30; RM: 4.62±0.09). The arrow and the box represent the dominant gel band which was submitted to tryptic digestion and MALDI-ToF/ToF analysis. (e) Biomacromolecular yield obtained for the small RNA, □ total and large RNA, DNA and protein (first elution) fractions (n=5, error bars represent s.d.). FU, fluorescent unit; L, ladder; M, marker; NA, NA-based method; Norm., normalized; nt, nucleotides; RM, reference methods; TR, TR-based method; QA, QA-based method.

Figure 5

Application of the developed biomolecular isolation methodology to a LAO-enriched microbial community (n=3), river water filtrate (n=3) and human feces (n=3). (a–c) Left-hand panels represent LAO-enriched microbial communities, middle panels represent river water filtrate and right-hand panels represent human feces. (a) Representative GC–MS total ion chromatograms of polar and non-polar metabolite fractions. Representative Agilent Bioanalyzer 2100 electropherograms of the (b) total RNA fractions and (c) small RNA fractions. (d) Agarose gel electrophoresis image of the genomic DNA fractions (Mean amount loaded in μg±s.d., right to left; LAO: 0.63±0.28; river water: 0.35±0.03; feces: 0.61±0.26) for each of the three technical replicates considered. (e) SDS-PAGE image of protein fractions, first elution (Mean amount loaded in μg±s.d., right to left; LAO: 1.19±0.40; river water: 1.70±0.63; feces: 1.19±1.21) for each of the three technical replicates considered. L, ladder; RT, retention time.

Figure 6

Three-way comparison of microbial community metabolomes obtained from a LAO-enriched microbial community, river water filtrate and human feces. Each dot represents the relative abundance of a metabolite in the three different environmental samples. Exemplary metabolite signatures for each microbiome are highlighted with their respective abundance levels on the right. Error bars represent s.d.