Effects of preservation method on canine (Canis lupus familiaris) fecal microbiota
- Published
- Accepted
- Subject Areas
- Biodiversity, Bioinformatics, Genomics, Microbiology, Veterinary Medicine
- Keywords
- Fecal preservation, Microbiome, Canine, 16S rRNA, Gut microbiota, Sample storage, DNA preservation
- Copyright
- © 2018 Horng et al.
- Licence
- This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, reproduction and adaptation in any medium and for any purpose provided that it is properly attributed. For attribution, the original author(s), title, publication source (PeerJ Preprints) and either DOI or URL of the article must be cited.
- Cite this article
- 2018. Effects of preservation method on canine (Canis lupus familiaris) fecal microbiota. PeerJ Preprints 6:e26912v1 https://doi.org/10.7287/peerj.preprints.26912v1
Abstract
Studies involving gut microbiome analysis play an increasing role in the evaluation of health and disease in humans and animals alike. Fecal sampling methods for DNA preservation in laboratory, clinical, and field settings can greatly influence inferences of microbial composition and diversity, but are often inconsistent and under-investigated between studies. Many laboratories have utilized either temperature control or preservation buffers for optimization of DNA preservation, but few studies have evaluated the effects of combining both methods to preserve fecal microbiota. To determine the optimal method for fecal DNA preservation, we collected fecal samples from one canine donor and stored aliquots in RNAlater, 70% ethanol, 50:50 glycerol:PBS, or without buffer at 25°C, 4°C, and -80°C. Fecal DNA was extracted, quantified, and 16S rRNA gene analysis performed on days 0, 7, 14, and 56 to evaluate changes in DNA concentration, purity, and bacterial diversity and composition over time. We detected overall effects on bacterial community of storage buffer (F-value= 6.87, DF= 3, P<0.001), storage temperature (F-value=1.77, DF= 3, P=0.037), and duration of sample storage (F-value=3.68, DF= 3, P<0.001). Changes in bacterial composition were observed in samples stored in -80°C without buffer, a commonly used method for fecal DNA storage, suggesting that simply freezing samples may be suboptimal for bacterial analysis. Fecal preservation with 70% ethanol and RNAlater closely resembled that of fresh samples, though RNAlater yielded significantly lower DNA concentrations (DF=8.57, P<0.001). Although bacterial composition varied with temperature and buffer storage, 70% ethanol was the best method for preserving bacterial DNA in canine feces, yielding the highest DNA concentration and minimal changes in bacterial diversity and composition. The differences observed between samples highlight the need to consider optimized post-collection methods in microbiome research.
Author Comment
This is a submission to PeerJ for review (accepted).
Supplemental Information
Fig S1. Average DNA purity (A260/280) by preservation method from Day 0 to 56 at 25°C, 4°C, and -80°C
Across all buffers and temperatures, DNA purity ± standard error did not change significantly over time.
Fig S2. Shannon diversity index of all samples by buffer, temperature, day, and their interactions
Significant effects were found according to storage buffer (F-value=3.07, DF=3, P=0.03). ***p<0.001, **p<0.01, *p<0.05.
Fig S3. Species richness of all samples by buffer, temperature, day, and their interactions
Significant effects were found according to storage buffer (F-value=12.4, DF=3, P<0.00001), duration of sample storage (F-value=10.8, DF=1, P=0.0016), the interaction between storage buffer and temperature (F-value=3.443, DF=3, P=0.22), and the interaction between storage buffer and duration of sample storage (F-value=9.67, DF=3, P<0.00001). ***p<0.001, **p<0.01, *p<0.05.
Fig S4. Species evenness of all samples by buffer, temperature, day, and their interactions
There were significant effects associated with interactions between storage buffer and storage temperature (F-value=3.98, DF=3, P=0.01), storage buffer and duration of sample storage (F-value=4.9, DF=3, P=0.004), and buffer, storage temperature, and duration of sample storage (F-value=3.1, DF=3, P=0.03). ***p<0.001, **p<0.01, *p<0.05.
Fig S5. Bray Curtis Dissimilarities of samples stored at 25°C, 4°C, and -80°C
Boxplot of Bray Curtis dissimilarities between sample groups representing median, lower quartile, and upper quartile distances bound between 0 and 1.
Fig S6. Comparison of DNA concentration yields obtained from QUBIT and Nanodrop
Across all buffers and temperatures, DNA detection by Nanodrop yielded much higher concentrations than that of QUBIT, particularly in that of RNA later samples.
Table S1. Effects of storage buffer on Shannon Diversity Index based on Tukey’s HSD
The difference between the means (diff), upper and lower levels of the 95% confidence interval around that mean difference, and p-values adjusted using Tukey were determined on R software. Significant interactions (bold) were considered when p<0.05.
Table S2. Effects of buffer, duration of storage, the interaction between buffer and temperature, and the interaction between buffer and duration of storage on Species Richness based on Tukey’s HSD
The difference between the means (diff), upper and lower levels of the 95% confidence interval around that mean difference, and p-values adjusted using Tukey were determined on R software. Significant interactions (bold) were considered when p<0.05.
Table S3. Effects of the interaction between buffer and temperature, buffer and duration of storage, and buffer, temperature, and duration on storage on Species Evenness based on Tukey’s HSD
The difference between the means (diff), upper and lower levels of the 95% confidence interval around that mean difference, and p-values adjusted using Tukey were determined on R software. Significant interactions (bold) were considered when p<0.05.
Table S4. Bray Curtis Dissimilarity distances between groups of samples at 25°C, 4°C, and -80°C
Significance testing of Bray Curtis dissimilarity was performed using a two-sided Student's two-sample t-test. T-statistics and P-values including Bonferonni correction between buffers were evaluated with sample storage at 25°C, 4°C, and -80°C.