Short-term preservation does not alter fecal microbiota community independent of the storage temperature. Since stool samples commonly experience a temporary storage after sampling, we wondered whether a short period storage would significantly affect microbiome stabilization. Nine volunteers donated fecal samples and each sample was divided into four aliquots to be stored for 4h at ambient temperature (AT), 4°C, -20°C and − 80°C, respectively. -80°C preservation was used as the control method. Post temporary storage, microbial DNAs were extracted from each aliquot followed by shotgun sequencing. As showed in Fig. 1A, we found a similar α-diversity among all groups estimated by indices of Shannon, Simpson and Evenness, though values were slightly higher in AT and 4°C groups than freezing groups (-20°C and − 80°C), which was possibly caused by the introduction of experimental microorganisms or the altered growth affected by atmospheric oxygen. These results indicated that microbial structures in stool samples were not significantly affected by storage temperature under a short-term preservation within 4 hours. Principle coordinate analysis (PCoA) further revealed that samples across four different temperature groups tended to overlap each other, exhibiting a high similarity of bacterial structure (Fig. 1B). Analogously, the Euclidean distance of microbial communities to -80°C group was not altered among ambient temperature, 4°C and − 20°C groups (Fig. 1C).
As for bacterial composition, we then analyzed the relative abundances of dominant genera and species. At the genera level, six out of nine samples (including samples from No.1 to 5 and No.8) showed an almost identical compositional profile across four temperature conditions (Fig. 1D), suggesting few perturbations caused by different temperatures to microbial community. Other samples displayed a disordered pattern: No.6 sample had a similar composition under AT, 4°C and − 80°C temperatures, the profiles of No.7 sample at AT, 4°C and − 20°C were different from that at -80°C, while for No.9 sample, ambient temperature storage changed the genera abundance, obviously distinct from other temperatures. Regarding the species profiling, we observed a relatively consistent pattern with the genera abundance (Fig. 1E).
Collectively, our results implied that a temporary storage no longer than 4 hours could stabilize both microbial structure and composition in human fecal samples, independent of the storage temperature. These findings also provided a hint that the temperature factor did not matter for a temporary preservation (e.g. <4 hours), thus a short-period storage may be a reliable practice to be adopted at the absence of refrigerators and liquid nitrogen.
A self-prepared preservation buffer (PB) enables to stabilize fecal microbial consortia. It is widely accepted that storage at room temperature can destroy the microbial consortia in sequencing samples. In accordance, we indeed observed the remarkable changes of microbial structure and composition caused by in donated fecal samples under room temperature, as compared with snap-frozen samples with LN (Fig. 2). Importantly, the alterations apparently occurred after even one-day RT treatment, though the storage period last up to 4 weeks. Miguel and colleagues once reported a nucleic acid preservation (NAP) buffer could maintain the quantity and quality of RNA and DNA from mammal samples under field conditions23. We then evaluated whether this lab-prepared preservation buffer (abbreviated as PB in the context) could be applied to stabilize microbial communities under ambient temperature (PB-AT) and high temperature (PB-HT) mimicking the temperature fluctuation during summer transportation. For community diversity, contrary to RT storage, PB application marginally decreased the values of Shannon, Simpson and Evenness indices (Fig. 2A), compared with LN group and the reduce tended to be stable after 3-day preservation. Meanwhile, PB-HT group samples which underwent 2-week PB preservation followed by an extra 50°C treatment displayed a highly similar profile of bacterial diversity with PB-AT group (Fig. 2A). These results clearly presented the inhibitory role of PB buffer in microbiota blooming under room temperature.
Based on distance analysis (Fig. 2B), the relative distance of AT group to LN group was approximately twice further than that of PB-applied group (for Bray-Curtis, 0.6 vs 0.3; for Euclidean, 0.4 vs 0.2), and the spearman coefficients of two PB groups were apparently higher than AT samples when relative to LN group (Fig. 2B). Moreover, samples stored at ambient temperature (included in the blue ellipse) drifted heavily from LN-frozen samples (the red dots) (Fig. 2C), while PB-preserved samples (mainly embraced in the pink ellipse) tended towards LN group, exhibiting that PB-treated groups were in close proximity to LN group regarding the microbial structures.
Furthermore, we performed compositional analysis at the genera and species levels and found that one-day AT preservation substantially altered genera abundance in microbial community, distinct from the pattern displayed in LN group (Fig. 3A), indicating samples should avoid storage at room temperature, even merely for one day. Conversely, PB maintained the genus composition pattern to a large extent, with some mild perturbations in the proportion of genera, e.g. the increase in Prevotella, Bacteroides and Eubacterium; the reduction in Megamonas and Megasphaera (Fig. 3A), yet far less than the destruction of composition resulted from ambient temperature storage. We also observed a great similarity in the relative abundance of dominant genera between PB-AT and PB-HT groups (Fig. 3A). Consistently, the species composition exhibited an analogous pattern to the observed genera profile (Fig. 3B).
According to these findings, the utilization of self-made PB buffer could facilitate to stabilize microbiome in human fecal samples, promoting samples eligible for the following complicated analysis.
PB buffer is suitable for human saliva to maintain microbiota community. Aside from we human beings, investigations in the burgeoning microbiome filed have reached to a wide variety of subjects such as model organisms, companion animals, wildlife, and marine organisms12,14,16,24. Mammalian microbiota not only vastly colonizes in host gastrointestinal tracts but also resides within or on the body, including lung, oral mucosa, skin and vaginal mucosa1, thereby provoking the multifarious samples and the corresponding exploration of preservation methods. In this study, we next tested whether the protective effects of PB would be retained in another sample type. Taking saliva as an example, we collected saliva samples from five volunteers and each sample was divided into 2 aliquots for − 80°C cryopreservation and PB buffer preservation at ambient temperature, respectively. According to the aforementioned results of fecal samples preserved with PB buffer (Fig. 2,3), one-week storage was used as the representative duration in this experiment. Compared with − 80°C condition, treatment with PB buffer did not change the indices of Shannon, Simpson, and Evenness of microbial community, as depicted in Fig. 4A, indicating the α-diversity of saliva microbiota was maintained by PB buffer. PCoA analysis also showed that PB-treated samples clustered together with frozen samples (Fig. 4B). Our results consolidated that application of PB buffer could stabilize the structure of saliva microbiome.
For microbial composition, PB group displayed a similar overall profile of dominant genera with that of control samples (Fig. 4C), although there were some alterations such as the relative abundance of Neisseria and Haemophilus was decreased and Actinomyces was increased post PB storage as compared to each − 80°C control. At the species level, the similar trends could be observed as well (Fig. 4D). Based on the above analysis, we concluded that the self-made PB buffer could prominently maintain the microbial consortia in saliva samples, exhibiting an acceptable suitability.