Microbiota taxonomic composition
There were 2,526 unique ASVs present in the samples. Twenty-four phyla
were identified and samples were dominated by Firmicutes (52.6%),
Proteobactiera (35.1%), and Actinobacteriota (8.5%; Fig. 2). All other
phyla had mean relative abundances of <1%. Across the 58
samples, there were 551 genera identified and twelve genera had relative
abundances >1%. The most commonly observed genera wereCandidatus Arthromitus (22.3%), Escherichia-Shigella(13.8%), Enterococcus (13.5%), Ligilactobacillus(9.1%), Klebsiella (7.8%), Cronobacter (4.8%),Erysipelatoclostridium (2.3%), Rothia (2.2%),Clostridium sensu stricto 1 (1.9%), Campylobacter(1.6%), Corynebacterium (1.4%), and Kocuria (1.1%).
At the phylum level, location significantly affected the relative
abundance of Firmicutes and Proteobacteria (Table 1). Specifically,
nestlings from urban sites had higher relative abundances of Firmicutes
(χ2 = 13.74, Padj = 0.008) and
lower abundances of Proteobacteria (χ2 = 8.35,Padj = 0.01) compared to nestlings from non-urban
field sites (Figure 2). At the genus level, location significantly
affected the abundance of Candidatus Arthromitus,Klebsiella , Erysipelatoclostridium , and Rothia(Table 1, Figure S3). Candidatus Arthromitus was highly abundant
in urban nestlings, but rarely observed in fecal samples collected from
non-urban nestlings (Figure 3, χ2 = 10.94,Padj = 0.01). Conversely, Klebsiella (
χ2=7.13, Padj = 0.03) andErysipelatoclostridium (χ2 = 6.62,Pad j = 0.03) were observed at higher abundances
in fecal samples from non-urban nestlings (Figure 3). The genusRothia was more common in nestlings from urban nests and was
rarely observed in non-urban nestlings (χ2 = 7.29,Pad j = 0.03). This trend was difficult to
visualize due to two individuals that had very high relative abundance
values (Figure S4: one non-urban nestling: 41.89% Rothia , and
one urban nestling: 63.43% Rothia ). Additional outliers were
detected and removed using the boxplot.stats function in R (non-urban:
1.25%, urban nestlings: 8.36%, 2.77%, 2.15%, 0.73%). Location
(urban, non-urban) still significantly affected the relative abundance
of Rothia with the two greatest outliers removed (Figure S4:
χ2 = 10.11, Pad j = 0.01) and
with the five additional outliers identified by the boxplot.stats
function removed (shown in Fig 3, χ2 = 12.57,Pad j = 0.005).
Parasite treatment did not significantly influence the relative
abundance of the three most abundant phyla or the abundance of any of
the twelve most abundant genera (Table 1). When analyzing only nestling
samples from sham-fumigated nests, there was still a significant effect
of location on the abundance of Firmicutes and Proteobacteria (Figure
S5). However, there was no effect of urbanization on the abundance of
the top twelve genera within the sham-fumigated nests (Table 1).