Abstract
Microbiome research and improvements in high throughput sequencing technologies revolutionize our current scientific viewpoint. The human associated microbiome is a prominent focus of clinical research. Large cohort studies are often required to investigate the human microbiome composition and its changes in a multitude of human diseases. Reproducible analyses of large cohort samples require standardized protocols in study design, sampling, storage, processing, and data analysis. In particular, the effect of sample storage on actual results is critical for reproducibility. So far, the effect of storage conditions on the results of microbial analysis has been examined for only a few human biological materials (e.g., stool samples). There is a lack of data and information on appropriate storage conditions on other human derived samples, such as skin. Here, we analyzed skin swab samples collected from three different body locations (forearm, V of the chest and back) of eight healthy volunteers. The skin swabs were soaked in sterile buffer and total DNA was isolated after freezing at -80°C for 24 h, 90 or 365 days. Hypervariable regions V1-2 were amplified from total DNA and libraries were sequenced on an Illumina MiSeq desktop sequencer in paired end mode. Data were analyzed using Qiime 1.9.1. Summarizing all body locations per time point, we found no significant differences in alpha diversity and multivariate community analysis among the three time points. Considering body locations separately significant differences in the richness of forearm samples were found between d0 vs. d90 and d90 vs. d365. Significant differences in the relative abundance of major skin genera (Propionibacterium, Streptococcus, Bacteroides, Corynebacterium, and Staphylococcus) were detected in our samples in Bacteroides only among all time points in forearm samples and between d0 vs. d90 and d90 vs. d365 in V of the chest and back samples. Accordingly, significant differences were detected in the ratios of the main phyla Actinobacteria, Firmicutes, and Bacteroidetes: Actinobacteria vs. Bacteroidetes at d0 vs. d90 (p-value = 0.0234), at d0 vs. d365 (p-value = 0.0234) and d90 vs. d365 (p-value = 0.0234) in forearm samples and at d90 vs. d365 in V of the chest (p-value = 0.0234) and back samples (p-value = 0.0234). The ratios of Firmicutes vs. Bacteroidetes showed no significant changes in any of the body locations as well as the ratios of Actinobacteria vs. Firmicutes at any time point. Studies with larger sample sizes are required to verify our results and determine long term storage effects with regard to specific biological questions.
Highlights
Published studies on the human derived microbiome, as the entity of all microbial genomes in and on the human body, have increased tremendously in the last decade from 257 publications in 2005 to 5849 in 2015 [retrieved from PubMed by the search term ‘human microbiome’ and updated from Toh and Allen-Vercoe (2015)]
Tree based hierarchical agglomerative clustering with average linkage dendrogram analysis on Bray– Curtis distances revealed no significant clustering of samples based on the sample storage duration at −80◦C but our results show a clustering by volunteer followed by the clustering per body location (Supplementary Figure 2)
We analyzed the effects of long term storage at −80◦C from skin derived dermal microbiome swab samples on the results of microbial composition
Summary
Published studies on the human derived microbiome, as the entity of all microbial genomes in and on the human body, have increased tremendously in the last decade from 257 publications in 2005 to 5849 in 2015 [retrieved from PubMed by the search term ‘human microbiome’ and updated from Toh and Allen-Vercoe (2015)]. Future needs to implement microbiome analysis in daily clinical procedures are already drafted (Fricke and Rasko, 2014; Klymiuk et al, 2014) This started process of standardization to increase reproducibility, efficiency and quality of data output on microbiome research projects will affect the processing in the wet lab during analyses of sample material. Other studies demonstrated that microbial diversity remains relatively stable among various storage conditions, whereas the relative abundance of main taxa can change dramatically if samples are stored at room temperature for 2 weeks (Cardona et al, 2012) All these studies lack information on long term storage effects on microbiome analyses and on human derived skin samples
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
