Page 27 - Read Online
P. 27
Horie et al. Microbiome Res Rep 2024;3:35 https://dx.doi.org/10.20517/mrr.2024.08 Page 5 of 12
16S rRNA gene amplicon sequencing and bacterial flora analysis by QIIME2
Genomic DNA for microbiome analysis from animal feces was prepared using ISOSPIN Fecal DNA
(Nippon Gene Co., Ltd., Tokyo, Japan) in accordance with the manufacturer’s protocol with a beads cell
disrupter (Micro Smash, Tomy Seiko Co., Ltd., Tokyo, Japan). Microbiome analysis based on the V1-V3
region of the 16S rRNA gene was performed by Eurofins Genomics, K.K. (Tokyo, Japan). The reads were
[43]
filtered using fastp 0.23.2 . The primer sequences were removed by truncating 19 bases at the 5’ end of the
forward read and 23 bases at the 5’ end of the reverse read. The truncation of one base at the 3’ end, removal
of reads with an average Q score of less than 30, and truncation of low-quality bases (average Q score of less
than 30) at the 3’ end were performed using a sliding window (window size 4). Sequence analysis was
[44]
performed on the filtered reads using QIIME 2 2022.8 . Amplicon sequence variants (ASVs) were created
using DADA2 (q2-Dada2). Bacterial classification was assigned to representative sequences of each ASV
[45]
using the naive Bayesian classification method with q2-feature-classifier classify-sklearn . The V1-V3
[46]
regions that had 99% homology to the Silva release 138.1 SSU database were used after processing by q2-
feature-classifier fit-classifier-naive-bayes [47,48] . Classification was based on Silva release 138.1, instruments
were curated by RESCRIPt , and extracted V3-V4 regions were based on amplification primer sequences
[49]
and then processed by qiime feature-classifier fit-classifier-naive-bayes. RESCRIPt curation was performed
by removing low-quality sequences (sequences containing more than 5 ambiguous bases or homopolymers
longer than 8 bases), base-length filtering (i.e., removing sequences that did not meet the following criteria:
archaea ≥ 900 bp, bacteria ≥ 1,200 bp, eukaryotes ≥ 1,400 bp), and deleting redundant sequences. ASVs
presumed to be derived from chloroplasts or mitochondria were removed from the summary table.
Principal coordinate analysis and cluster analysis
Differences between the samples were visualized using Principal Coordinate Analysis (PCoA) and cluster
analysis using ASVs. The ASV summary table was diluted to the minimum size using the rarefy function of
[50]
vegan ver. 2.6-4 . Representative ASV sequences were aligned using MAFFT ver. 7.490 with the E-INS-i
option. Based on the aligned sequences, they were optimized using the GTR + CAT model with fasttree ver.
[50]
2.1.11 . A phylogenetic tree was constructed using the maximum likelihood method. Based on the
constructed phylogenetic tree, the unweighted and weighted UniFrac distances between samples were
measured using phyloseq ver. 1.38.0 and R ver. 4.2.2 [51,52] . PCoA was performed using the cmdscale function
in R.
RESULTS
Isolation of lactic acid bacteria from animal feces
Lactic acid bacteria were isolated from the cultured feces of zoo animals, including carnivores (jaguar, tiger,
lion, and puma), piscivores (Asian small-clawed otter, California sea lion, South American sea lion, harbor
seal, Baikal seal, and bottlenose dolphin), herbivores [South American tapir, Malayan tapir, horse (Japanese
native species: Noma-uma), giraffe, and West Indian manatee], and omnivores (Japanese badger, raccoon
dog, and bear). Details of the host animals are shown in Supplementary Table 1. Three species of bear,
carnivorous polar bear, herbivorous sun bear, and intermediate Asian black bear, were included. Details of
the diet provided to each animal are listed in Supplementary Table 2. Lactic acid bacteria were isolated from
the feces of the 20 animals with different eating habits using MRS and LBS agar [Supplementary Table 1].
Former Lactobacillus species were frequently isolated from the feces of carnivorous animals. Almost all of
the colonies that formed on the MRS and LBS agar plates belonged to the former genus Lactobacillus.
Differences in the species of lactic acid bacteria among hosts were small. L. salivarius was the most
frequently isolated, followed by Ligilactobacillus saerimneri (L. saerimneri). Enterococcus faecium was also
commonly isolated. Former Lactobacillus species were also found in the feces of herbivores; however, they
differed from those isolated from the feces of carnivores. Ligilactobacillus equi (L. equi) and
Limosilactobacillus gorillae were isolated from the herbivores, and Ligilactobacillus salivarius was also

