The impact of freeze-drying infant fecal samples on measures of their bacterial community profiles and milk-derived oligosaccharide content

Abstract

Infant fecal samples are commonly studied to investigate the impacts of breastfeeding on the development of the microbiota and subsequent health effects. Comparisons of infants living in different geographic regions and environmental contexts are needed to aid our understanding of evolutionarily-selected milk adaptations. However, the preservation of fecal samples from individuals in remote locales until they can be processed can be a challenge. Freeze-drying (lyophilization) offers a cost-effective way to preserve some biological samples for transport and analysis at a later date. Currently, it is unknown what, if any, biases are introduced into various analyses by the freeze-drying process. Here, we investigated how freeze-drying affected analysis of two relevant and intertwined aspects of infant fecal samples, marker gene amplicon sequencing of the bacterial community and the fecal oligosaccharide profile (undigested human milk oligosaccharides). No differences were discovered between the fecal oligosaccharide profiles of wet and freeze-dried samples. The marker gene sequencing data showed an increase in proportional representation of Bacteriodes and a decrease in detection of bifidobacteria and members of class Bacilli after freeze-drying. This sample treatment bias may possibly be related to the cell morphology of these different taxa (Gram status). However, these effects did not overwhelm the natural variation among individuals, as the community data still strongly grouped by subject and not by freeze-drying status. We also found that compensating for sample concentration during freeze-drying, while not necessary, was also not detrimental. Freeze-drying may therefore be an acceptable method of sample preservation and mass reduction for some studies of microbial ecology and milk glycan analysis.

Keywords: Fecal microbiome; Freeze-drying; Human microbiome; Human milk oligosaccharides; Infants; Lyophilization.

Conflict of interest statement

DAM, JBG, and CBL are co-founders of Evolve Biosystems, a company focused on diet-based manipulation of the gut microbiota. Evolve Biosystems had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Figures

Figure 1. Bacterial communities of freeze-dried and wet feces.

The bacterial community structures of the feces of 24 test infants. Relative abundances of each bacterial taxon are shown. Each sample is grouped together, with the two wet replicates first, followed by the two low-mass duplicates, and lastly the high-mass duplicates.

Figure 2. Comparison of fecal oligosaccharide measures between freeze-dried and wet fecal samples.

Absolute and relative abundances of fecal oligosaccharides.

Figure 3. Cladograms representing the taxa enriched under various treatments.

(A) shows the differential features of wet and dry (both high- and low-mass) replicates in an all-against-all comparison (classes = wet and dry; subclasses = wet, high, and low). (B) shows the results separated by high-mass and low-mass as well (classes = wet, high, and low) in a one-against-all comparison.

Figure 4. Principle Coordinates Analysis (PCoA) plots showing the clustering of samples by various metadata.

(A) by Wet vs. Dry, (B) Wet vs. High Mass Lyophilized vs. Low Mass Lyophilized, (C) by Infant (subject).