Quantifying bias introduced by sample collection in relative and absolute microbiome measurements.
Author(s): Maghini DG, Dvorak M, Dahlen A, Roos M, Kuersten S, Bhatt AS
Publication: Nat Biotechnol, 2024, Vol. 42, Page 328-338
PubMed ID: 37106038 PubMed Review Paper? No
Purpose of Paper
This paper compared DNA and RNA yields and metagenomic and metatranscriptomic profiles of matched fecal specimens that were frozen at -80°C without preservative or in OMNIgene GUT OMR-200 collection buffer (OMNIgene) or Zymo DNA/RNA Shield Fecal Collection buffer (Zymo) and among specimens preserved in OMNIgene or Zymo and stored at 23°C or 40°C (rather than -80°C) for 7 days.
Conclusion of Paper
The metagenomic and metatranscriptomic profiles of fecal specimens are both affected by preservative and storage temperature of Zymo-preserved specimens, but only the metatrascriptomic profile is significantly affected by storage temperature of OMNIgene preserved specimens. Specimens preserved in Zymo or OMNIgene at -80°C had higher RNA yields and those preserved in Zymo had lower DNA concentrations than those of matched unpreserved frozen specimens. Metagenomic sequencing found relative enrichment of all Bacteroidetes and depletion of many Firmicutes and Actinobacteria in specimens preserved at -80°C in Zymo or OMNIgene relative to unpreserved specimens. Real-time PCR identified an increase in microbial DNA, higher counts of Bacteroidetes and Firmicutes, and a higher ratio of Bacteroidetes to Firmicutes when specimens were stored in OMNIgene at -80°C compared to unpreserved specimens; when specimens were stored in Zymo at -80°C, only the Bacteroidetes to Firmicutes ratio increased relative to unpreserved frozen specimens. Metatranscriptomic analysis found that Zymo and OMNIgene preserved specimens stored at -80°C had significant enrichment of Bacteroidetes and depletion of Firmicutes and viruses relative to the matched unpreserved frozen specimen. Specimens preserved at -80°C in either preservative had significantly lower Shannon entropy and Inverse Simpson Indices based on metagenomic sequencing, but Shannon entropy based on metatranscriptomic sequencing was not affected by preservative. Metagenomic profiles of those stored in Zymo were more similar to unpreserved frozen specimens than specimens frozen in OMNIgene.
Among the specimens preserved with Zymo, storage for 7 days at 23°C reduced DNA yields and storage for 7 days at 40°C reduced DNA and RNA yields relative to Zymo preserved specimens stored at -80°C. Metagenomic analysis found enriched levels of Bacteroidetes and decreased relative counts of many Firmicutes and Actinobacteria in Zymo preserved specimens when stored at 23 or 40°C rather than -80°C. Real-time PCR identified decreases in microbial DNA, Bacteroidetes, and Firmicutes counts and an increase in Bacteroidetes to Firmicutes ratio when specimens in Zymo were stored at 23 or 40°C rather than -80°C. Using metagenomic data, the Shannon entropy was higher and the Inverse Simpson Index was lower when Zymo preserved specimens were stored at 23 or 40°C than -80°C, but the metatranscriptomic Shannon entropy was lower when Zymo preserved specimens were stored at 40 than -80°C. Bray–Curtis dissimilarities from frozen unpreserved specimens were larger in both metagenomic and metatranscriptomic analyses when Zymo preserved specimens were stored at 23 or 40°C than -80°C.
While storage temperature did not affect the DNA concentration of OMNIgene preserved specimens, RNA concentrations of OMNIgene preserved specimens were lower after storage at 23 than -80°C and specimens stored at 40°C did not yield quantifiable RNA. Metagenomic analysis found little to no consistent effect of storage temperature on the abundance of the phyla in OMNIgene preserved specimens. Real-time PCR confirmed no real differences in Bacteroidetes and Firmicutes counts or their ratio in OMNIgene specimens among storage temperatures. However, metatranscriptomic analysis identified enrichment of Actinobacteria, viruses, and fungi in OMNIgene specimens stored at 23°C compared to those stored at -80°C. Shannon entropy based on metagenomic or metatranscriptomic data was unaffected by the temperature of storage of OMNIgene preserved specimens, but Bray–Curtis dissimilarity based on the metatrascriptomic or metagenomic data increased when specimens in OMNIgene were stored at 23°C relative to the similarly preserved specimen stored at -80°C.
Studies
-
Study Purpose
This study compared the yield of DNA and metagenomic profile of matched fecal specimens frozen at -80°C without preservative or preserved in OMNIgene GUT OMR-200 collection buffer (OMNIgene) or Zymo DNA/RNA Shield Fecal Collection buffer (Zymo) and among specimens preserved in OMNIgene or Zymo and stored at 23°C or 40°C (rather than -80°C) for 7 days. Feces were collected from 10 donors (diagnosis not specified). Each specimen was aliquoted into plain tubes and those containing OMNIgene GUT OMR-200 collection solution or Zymo DNA/RNA Shield Fecal Collection buffer. Triplicate aliquots from each tube type were stored at -80°C. Fecal aliquots that were stored in OMNIgene or Zymo were also stored at 23°C or 40°C for 7 days before freezing. DNA was extracted using the QIAamp PowerFecal Pro DNA Kit with the modification that a vacuum manifold replaced centrifugation. DNA was quantified using Qubit. Sequencing libraries were prepared with the Illumina DNA Prep Kit and sequenced on a NovaSeq 6000 machine. 16S rRNA was quantified by real-time PCR.
Summary of Findings:
DNA concentration was significantly lower in specimens stored in Zymo at -80°C than in unpreserved specimens stored -80°C (P<0.001) and in specimens stored in Zymo at 23°C or 40°C than at -80°C (P<0.001, both), but were unaffected by preservation in OMNIgene. The metagenomic profile showed the expected inter-donor variability but preservative and storage temperature-specific changes in abundance at the genus and phyla level were also observed. Notably, there was enrichment of all Bacteroidetes and depletion of many Firmicutes and Actinobacteria in specimens stored in either OMNIgene or Zymo at -80°C relative to unpreserved specimens stored at -80°C. While storage at 23°C or 40°C enriched levels of Bacteroidetes and decreased relative counts of many Firmicutes and Actinobacteria in specimens preserved in Zymo relative to the Zymo-preserved specimens stored at -80°C, storage temperature had little to no effect on the abundance of Bacteroidetes in OMNIgene-preserved specimens. In OMNIgene-preserved specimens, storage temperature-related effects were not consistent on the phyla level, with both enrichment and depletion of different genera observed. On the genus level, the Shannon entropy and Inverse Simpson Indices were significantly lower in specimens stored in either OMNIgene or Zymo at -80°C relative to unpreserved specimens stored at -80°C (P≤0.001, both for OMNIgene; and P=0.002 and P≤0.001, respectively for Zymo). Shannon entropy was higher and the Inverse Simpson Index was lower when Zymo-preserved specimens were stored at 23°C (P=0.009 and P≤0.01, respectively) or 40°C (P≤0.001, both) than -80°C. When immediately frozen, feces stored in Zymo were more similar to unpreserved frozen specimens than specimens frozen in OMNIgene (Bray–Curtis dissimilarity of 0.14 versus 0.16, P=0.022). However, specimens stored at 23°C or 40°C in Zymo preservative showed larger Bray–Curtis dissimilarities from frozen unpreserved specimens than when stored at -80°C.
Real-time PCR identified an increase in microbial DNA when specimens were stored at -80°C in OMNIgene compared to unpreserved specimens (P≤0.05) and decreased microbial DNA when specimens in Zymo were stored at 23°C or 40°C rather than -80°C (P≤0.05 and P≤0.01, respectively). ANOVA found that more of the variation in microbial abundance was explained by preservative and storage temperature than donor-specific differences (37.6% versus 25.9%). Feces stored at -80°C in OMNIgene had significantly higher counts of Bacteroidetes and Firmicutes and a higher ratio of Bacteroidetes to Firmicutes than in unpreserved specimens stored at -80°C (P<0.001), but no differences in Bacteroidetes and Firmicutes counts or their ratio were observed in specimens stored in OMNIgene among the storage temperatures examined. In contrast, feces stored at -80°C in Zymo preservative had comparable counts of Bacteroidetes and Firmicutes relative to unpreserved specimens stored at -80°C, but displayed temperature-dependent decreases in Bacteroidetes and Firmicutes with increased storage temperature (P<0.05) and an increase in Bacteroidetes to Firmicutes ratio when specimens were preserved in Zymo at -80°C (P≤0.001) or when preserved in Zymo with increased storage temperature (P≤0.001, all). The authors conclude that both preservatives cause a shift in the metagenomic profile and, in Zymo preserved specimens, there was an additional effect of storage temperature on the metagenomic profile.
Biospecimens
Preservative Types
- Other Preservative
- Frozen
Diagnoses:
- Not specified
Platform:
Analyte Technology Platform DNA Fluorometry DNA Next generation sequencing DNA Real-time qPCR Pre-analytical Factors:
Classification Pre-analytical Factor Value(s) Storage Storage temperature 23°C
40°C
-80°C
Biospecimen Preservation Type of fixation/preservation Zymo DNA/RNA Shield
OMNIgene GUT
Frozen
-
Study Purpose
This study compared the RNA yield and metatranscriptomic profile of matched fecal specimens that were frozen at -80°C without preservative or in OMNIgene GUT OMR-200 collection buffer (OMNIgene) or Zymo DNA/RNA Shield Fecal Collection buffer (Zymo) and among specimens preserved in OMNIgene or Zymo and stored at 23°C or 40°C (rather than -80°C) for 7 days. Feces were collected from 10 donors (diagnosis not specified). Each specimen was aliquoted into plain tubes, OMNIgene GUT OMR-200 collection tubes, and tubes containing Zymo DNA/RNA Shield Fecal Collection buffer. Triplicate aliquots from each tube type were stored at -80°C. Fecal aliquots stored in OMNIgene or Zymo were also stored at 23°C or 40°C for 7 days before freezing. RNA was extracted using the RNeasy PowerMicrobiome Kit with a vacuum manifold in lieu of centrifugation. RNA was quantified using Qubit. Sequencing libraries were prepared using the Illumina Stranded Total RNA Prep with Ribo-Zero Plus Microbiome kit and sequenced on a NovaSeq 6000 machine.
Summary of Findings:
RNA was undetectable in fecal specimens stored in OMNIgene for 7 days at 40°C and was lower from specimens stored in Zymo or OMNIgene at -80°C than in unpreserved specimens stored -80°C (P<0.001, both). RNA concentrations were also lower when specimens were stored in OMNIgene at 23°C than at -80°C (P<0.05) or in Zymo at 40°C than at -80°C (P<0.001). The metatranscriptomic profile showed the expected inter-donor variability but preservative and storage temperature-specific changes in abundance at the genus and phyla level were also observed. Relative to unpreserved specimens stored at -80°C, Zymo- and OMNIgene-preserved specimens stored at -80°C showed significant enrichment of Bacteroidetes (P≤0.001, both) and depletion of Firmicutes (P≤0.01, both) and viruses (P≤0.05 and P≤0.01, respectively); in addition, frozen specimens in Zymo had a depletion of Actinobacteria (P≤0.01) and Fungi (P≤0.05). Actinobacteria, viruses, and fungi were all enriched in OMNIgene specimens stored at 23°C compared to those stored at -80°C (P≤0.001, P≤0.05, and P≤0.001, respectively). Compared to Zymo specimens stored at -80°C, Zymo specimens stored at 23°C had depleted Firmicutes (P≤0.01) and enrichment of fungi (P≤0.01) and specimens stored at 40°C had an enrichment of Bacteriodetes (P≤0.001), viruses (P≤0.001), and fungi (P≤0.01) and depleted Firmicutes (P≤0.001) and Actinobacteria (P≤0.05). Shannon entropy was similar among specimens stored at -80°C that were unpreserved or preserved in Zymo or OMNIgene, but was lower in specimens stored in Zymo at 40°C than those stored at -80°C (P≤0.01) and higher in specimens stored in OMNIgene at 23°C than at -80°C (P≤0.001). Beta-diversity was also affected by storage temperature as Bray–Curtis dissimilarity was significantly increased relative to similarly preserved specimen stored at -80°C when specimens in OMNIgene were stored at 23°C (P≤0.01) or in Zymo at 23°C or 40°C (P≤0.001, both). The authors conclude that both preservatives cause a shift in the metatranscriptomic profile, and further shifts in the metatranscriptomic profile are observed with increased storage temperature, but they conclude that use of Zymo is preferrable to OMNIgene for metatranscriptomics studies.
Biospecimens
Preservative Types
- Other Preservative
- Frozen
Diagnoses:
- Not specified
Platform:
Analyte Technology Platform RNA Next generation sequencing RNA Fluorometry Pre-analytical Factors:
Classification Pre-analytical Factor Value(s) Storage Storage temperature 23°C
40°C
-80°C
Biospecimen Preservation Type of fixation/preservation Zymo DNA/RNA Shield
OMNIgene GUT
Frozen