Literature DB >> 31825977

California condor microbiomes: Bacterial variety and functional properties in captive-bred individuals.

Lindsey Jacobs1, Benjamin H McMahon1, Joel Berendzen2, Jonathan Longmire1, Cheryl Gleasner1, Nicolas W Hengartner1, Momchilo Vuyisich3, Judith R Cohn1, Marti Jenkins4, Andrew W Bartlow1, Jeanne M Fair1.   

Abstract

Around the world, scavenging birds such as vultures and condors have been experiencing drastic population declines. Scavenging birds have a distinct digestive process to deal with higher amounts of bacteria in their primary diet of carcasses in varying levels of decay. These observations motivate us to present an analysis of captive and healthy California condor (Gymnogyps californianus) microbiomes to characterize a population raised together under similar conditions. Shotgun metagenomic DNA sequences were analyzed from fecal and cloacal samples of captive birds. Classification of shotgun DNA sequence data with peptide signatures using the Sequedex package provided both phylogenetic and functional profiles, as well as individually annotated reads for targeted confirmatory analysis. We observed bacterial species previously associated with birds and gut microbiomes, including both virulent and opportunistic pathogens such as Clostridium perfringens, Propionibacterium acnes, Shigella flexneri, and Fusobacterium mortiferum, common flora such as Lactobacillus johnsonii, Lactobacillus ruminus, and Bacteroides vulgatus, and mucosal microbes such as Delftia acidovorans, Stenotrophomonas maltophilia, and Corynebacterium falsnii. Classification using shotgun metagenomic reads from phylogenetic marker genes was consistent with, and more specific than, analysis based on 16S rDNA data. Classification of samples based on either phylogenetic or functional profiles of genomic fragments differentiated three types of samples: fecal, mature cloacal and immature cloacal, with immature birds having approximately 40% higher diversity of microbes.

Entities:  

Year:  2019        PMID： 31825977      PMCID： PMC6905524          DOI： 10.1371/journal.pone.0225858

Source DB:  PubMed          Journal:  PLoS One        ISSN： 1932-6203            Impact factor:   3.240


Introduction

The California condor (Gymnogyps californianus) is a species that dramatically declined during the 20th century from poaching, lead poisoning, and changes in habitat [1]. In 1987, the last remaining individuals were captured from the wild to take part in a breeding program. This breeding program was successful, and in 1992 a small number of condors were reintroduced into the wild. Today there are over 500 California condors living in the wild or in captivity. However, there is less genetic diversity in today’s California condors because the 20th century dramatic decline in condor numbers also resulted in an 80% reduction of genetic diversity [2]. This lack of genetic diversity could have health effects, effects which could be enhanced by the potentially pathogenic infectious diseases California condors are exposed to in the wild [3]. Due to these potential impactors on condor health, indicators of condor health both in captivity and in the wild are of great interest to assess both population and individual health and condition. The microbiome (the collective genome of microbial species in a sample) may be a good indicator of health and condition in avian species [4]. Gut microbiota influence the health and physiology of vertebrates, indicating stressors faced by condors could be assessed using gut microbiota [5]. For example, lower nestling body condition in Great Tits (Parus major) is correlated with large shifts in microbial diversity and composition [6]. Metagenomic profiling of microbial communities has already been applied to provide both phylogenetic and functional information for environmental samples [7], wildlife [8], domestic animals [9], and endangered species [10, 11]. Thorough reviews of the avian microbiome exist and could be used in conservation efforts for avian species such as the California condor [4, 5, 12]. In conservation efforts for any species, the microbiome of healthy animals in captivity will provide a valuable resource for comparison with microbial data for wild individuals of the same species [13]. The avian lifestyle is unique from that of many animals, suggesting the role of bacteria in the avian microbiome may be different from what is typically seen. This is particularly true for scavenging birds such as vultures, in which near-neighbor human pathogens have been found in abundance [14]. It is thought that vultures tolerate these potentially pathogenic bacteria in order to benefit from the bacterial breakdown of carrion [15, 16]. To safely digest non-beneficial bacteria ingested from their primary diet of carcasses in varying levels of decay, scavenging birds such as vultures and condors have an extremely acidic digestive system (1.9 pH in vultures compared to 2.9 pH in humans) [17, 18]. Signatures of health in scavenger birds may be significantly different from other bird species due to this digestive adaptation, and analysis of the microbiome in representative healthy scavenger birds will be necessary for useful comparison with wild individuals. However, recent studies increasingly note the effect of captivity on the microbiome of animals [19, 20]. This is particularly true if the gut microbiome may be impacted by changes in diet from wild counterparts [21]. In addition, the importance of studying the microbiome in wildlife from an evolutionary standpoint has been highlighted [22]. Despite these issues, and prior to samples from wild condors being available, a baseline of variability and diversity of microbiome communities in captive condors should be assessed. In birds, the microbiome of growing chicks and nestlings have been shown to both have increased and decreased microbial diversity [6, 23–26]. These findings emphasize the importance of finding a baseline for California condors. Advances in signature-based analysis of genomic data demonstrate an ability to rapidly extract both phylogenetic and functional information from shotgun metagenomics data [12]. Compared to the 16S rDNA sequencing approach, shotgun metagenomics allows for the analysis of bacterial diversity in samples, yet also provides information on the presence of fungal, parasitic, and DNA virus pathogens [27]. In addition, sequencing the 16S rDNA gene, while providing an economical and well-understood technique to survey the phylogenetic composition of microbial communities, does not provide information regarding metabolic pathways or antimicrobial resistance patterns. Shotgun metagenomic sequencing promises to identify resistance genes, their point mutations, and other gene functional changes that can identify events such as exposure to lead contaminants. In this work, we apply the Sequedex metagenomics analysis package [28] to demonstrate the analyses of these features on microbiome data from California condors. Furthermore, we compare 16S rDNA and shotgun metagenomics sequencing in metrics of microbial composition and microbial diversity. The aims of our study are twofold. The first objective is to describe the microbiome in healthy California condors of different ages in captivity to characterize individual variation in the microbial community structure and bacterial genes under a similar environment. Captive individuals of condors of different ages offer an opportunity to test whether the diversity of microbial communities increases or decreases with age. Our second objective is to compare the sequencing approach of 16S rDNA with shotgun metagenomic sequencing for the condor microbiome. Data from captive California condors can be used as a baseline for future comparison of microbiome variation and diversity with that of wild California condors.

Results

Twenty species of bacteria comprised 89% of the bacterial identifications in the captive healthy California condor microbiomes (Table 1). All of these bacteria had previously been associated with birds and/or gut microbiomes (For example, see Ballou 2016, Larsen 2015, Prabhakar 2012, Vela 2015, and Vital 2014). For most samples, ~30,000 rDNA reads were obtained, while the shotgun sequencing yield was 10–15 million reads per sample. Between 4–58% of the metagenomic shotgun reads were identified as bacterial when analyzed with Sequedex, with 40% typical for the fecal samples. There was more dispersion in the percent of recognized reads in cloacal samples due to the variability in the amount of condor DNA present. Condor DNA contains a much smaller fraction of coding DNA than bacteria, and thus has a smaller fraction of reads that were recognized by the peptide signatures in Sequedex.
Table 1

Attributes of collected fecal and cloacal samples.

Both 16S and DNA shotgun metagenomic sequencing were performed on most samples. The two most-prevalent species are indicated for each individual in Table 1. Dominant species ascribed by the DNA shotgun metagenomics: Cje = Corynebacterium jeikeium, Pac = Propionibacterium acnes, Sfl = Shigella flexneri, Cpe = Clostridium perfringens, Fva = Fusobacterium varium, Bov = Bacteroides ovatus, Ckr = Corynebacterium kroppenstedtii, Eco = Escherichia coli, Sps = Staphylococcus pseudintermedius, Pav = Propionibacterium avidum, Dac = Delftia acidovorans, Cul = Corynebacterium ulcerans, Bsu = Brevundimonas subvibrioides, Esp = Enterococcus sp., Sma = Stenotrophomonas maltophilia, Ljo = Lactobacillus johnsonii, Fmo = Fusobacterium mortiferum, Pan = Peptostreptococcus anaerobius, Csp = Clostridium sp., Bvu = Bacteroides vulgatus, Cce = Clostridium celatum.

SexHatch yearBird #Sample Date16S readsMetagenomic reads% reads recognizedDominant species
CloacalM201369717-Dec-1339,23411,126,43815.8Cje, Pac
F201370718-Dec-1343,53712,056,81015.5Cje,Sfl
M20137068-Jan-14013,453,9023.8Sfl, Cje
M20137166-Jan-1423,65114,265,2788.4Sfl, Cpe
F201372122-Jan-1428,77115,460,7444.3Sfl, Fva
M201368514-Dec-1355,02413,786,20017.2Bov, Cpe
M201369617-Dec-1315,65913,051,19028.2Ckr, Eco
F201369816-Dec-1334,0568,506,85212.7Ckr, Sps
M201370016-Dec-1319,5348,994,38224Ckr, Pav
M201370118-Dec-1325,68215,598,71816.7Dac, Ckr
M201370420-Dec-1329,78213,051,19018.3Dac, Eco
M201369927-Dec-1319,36410,933,3367.3Cul, Dac
M19916014-Dec-1325,71311,785,83613Dac, Bsu
M201371121-Dec-138,20711,974,71837.5Dac, Esp
M201370327-Dec-1321,49515,304,92418.5Cac, Sma
F199716614-Dec-137,47413,353,70612.6Dac, Cje
M201058015-Dec-132,77212,457,47657.7Dac, Sma
F19916915-Dec-132,14914,452,10851.6Dac, Sma
M20137189-Jan-143,68512,140,74414.5Dac, Sma
F19915912-Dec-132,10411,789,40845.4Dac, Sma
FecalM201368514-Dec-1347,01416,659,09848.6Cpe, Ljo
F201369816-Dec-1329,58815,149,97232.8Cpe, Fmo
M20137166-Jan-1441,89816,279,53639.1Cpe, Fva
M20033091-Oct-1342,61015,212,90445.2Cpe, Pan
M201369617-Dec-1349,10516,654,56443.5Csp, Fva
M201370420-Dec-1332,17115,526,65838.8Cpe, Fva
M201058015-Dec-1337,65215,745,05238.2Fva, Cpe
M201369927-Dec-1347,37415,381,10645Cpe, Fva
M20133097-Oct-13014,752,24220.8Fva, Cce
M201370118-Dec-1332,32315,346,16031.2Bvu, Sfl

Attributes of collected fecal and cloacal samples.

Both 16S and DNA shotgun metagenomic sequencing were performed on most samples. The two most-prevalent species are indicated for each individual in Table 1. Dominant species ascribed by the DNA shotgun metagenomics: Cje = Corynebacterium jeikeium, Pac = Propionibacterium acnes, Sfl = Shigella flexneri, Cpe = Clostridium perfringens, Fva = Fusobacterium varium, Bov = Bacteroides ovatus, Ckr = Corynebacterium kroppenstedtii, Eco = Escherichia coli, Sps = Staphylococcus pseudintermedius, Pav = Propionibacterium avidum, Dac = Delftia acidovorans, Cul = Corynebacterium ulcerans, Bsu = Brevundimonas subvibrioides, Esp = Enterococcus sp., Sma = Stenotrophomonas maltophilia, Ljo = Lactobacillus johnsonii, Fmo = Fusobacterium mortiferum, Pan = Peptostreptococcus anaerobius, Csp = Clostridium sp., Bvu = Bacteroides vulgatus, Cce = Clostridium celatum.

Microbial composition

As described in the Sequedex documentation (https://sequedex.readthedocs.io/en/latest/) and shown in S1 Fig, each read recognized by Sequedex is placed at a node of the phylogenetic tree, according to the signature peptides contained in that read. The patterns of assigned reads shown are typical of those of single species, with the reads concentrated in a line from the root of the tree towards the leaves. This pattern arises because 151 base pair reads contain only a limited amount of phylogenetic information, with some reads specifically identified near the leaves of the tree and others that more ambiguously identify the sample’s organisms. All of these reads should, however, be counted to estimate the relative abundance of the different phylogenetic categories found in the sample. S1 Fig shows Sequedex’s phylogenetic assignment of reads for abundant taxa in four regions of the tree, with over 100,000 reads assigned in each of the examples provided. Each of the examples shown is consistent with the presence of a single species in that area of the phylogenetic tree, with the Lactobacillus example typical of a close match to two closely related species in the reference database, and the Clostridium sp. example consistent with a close match to a reference organism (C. perfringens in this case) that is distinct from other organisms in the reference database. Both the Fusobacterium and Propionibacterium examples are more ambiguous, reflecting either a mixture of related organisms or a novel organism with a genome inventory acquired from several of the reference organisms. The within-sample Simpson’s diversity based on reads rolled up into phyla significantly differed among the three groups (ANOVA: F2,27 = 19.41, P < 0.001; Fig 1) and was highest in the cloacal samples of the younger birds (‘Cloacal, immature’, Fig 1). Cloacal mature had the lowest diversity. Post-hoc tests show that all three groups significantly differ from each other (Tukey post-hoc test: all p-values < 0.05). In the immature cloacal samples, a variety of bacteria are present representing the genera Corynebacterium, Shigella, Fusobacterium, and Staphylococcus. The mature cloacal samples were dominated by a single organism, Delftia acidovorans. Clostridium and Fusobacterium were common genera in the fecal samples, although one sample was composed predominantly of Bacteroidetes and Shigella.
Fig 1

A stacked bar chart of the relative abundance of the bacterial composition of each of the samples is shown for selected phyla, indicated to the right of the Fig.

Predominant species comprising each phylum are indicated on the bar chart. Bacterial prevalence is estimated for each sample by Sequedex classification of the shotgun metagenomics data, and species were identified with both 16S sequence data (296 nucleotide per read) and selected RNA polymerase reads identified by Sequedex as being phylogenetically informative. Forward and reverse reads were analyzed separately and were closely related. The separation into four groups was made on the basis of sample type, (cloacal vs. fecal) and phylogenetic and functional profiles (cloacal: immature vs. mature). Complete results are provided in S1a File.

A stacked bar chart of the relative abundance of the bacterial composition of each of the samples is shown for selected phyla, indicated to the right of the Fig.

Predominant species comprising each phylum are indicated on the bar chart. Bacterial prevalence is estimated for each sample by Sequedex classification of the shotgun metagenomics data, and species were identified with both 16S sequence data (296 nucleotide per read) and selected RNA polymerase reads identified by Sequedex as being phylogenetically informative. Forward and reverse reads were analyzed separately and were closely related. The separation into four groups was made on the basis of sample type, (cloacal vs. fecal) and phylogenetic and functional profiles (cloacal: immature vs. mature). Complete results are provided in S1a File. When analyzing data at the node level (data from Fig 2), species richness significantly differed among the three groups (ANOVA: F2,27 = 33.95, P < 0.001). Again, all three groups significantly differed (Tukey post-hoc test: all p-values < 0.05) and the cloacal immature samples had the highest number of species. Cloacal mature had the fewest number of species. The three groups significantly differed in terms of Simpson’s diversity (ANOVA: F2,27 = 55.09, P < 0.001). Cloacal mature had lower microbial diversity than cloacal immature (Tukey post-hoc test: P < 0.001) and fecal mature (Tukey post-hoc test: P < 0.001) samples. Cloacal immature and fecal mature did not significantly differ from each other (Tukey post-hoc test: P = 0.81).
Fig 2

Sample-to-sample variability of phylogenetic composition across microbiome samples is computed by comparing the normalized number of reads recruited to specific nodes on the phylogenetic tree with Sequedex.

The number of reads in each sample that were assigned to the most-populated nodes of the reference phylogenetic tree (S1 Fig) are shown as rows in this heat map. These rows represent nineteen of the most common species observed, and the color indicates relative abundance, with dark blue indicating zero, and red more than half of the sample. These most-populated nodes account for more than 90% of the bacteria present in the sample without needing to differentiate the case of dominant species from diverse communities in particular areas of the phylogeny. Species names were determined using BLASTN with phylogenetically discriminating reads from the rpoB and rpoC polymerase reads. Complete results are provided in S1b File.

Sample-to-sample variability of phylogenetic composition across microbiome samples is computed by comparing the normalized number of reads recruited to specific nodes on the phylogenetic tree with Sequedex.

The number of reads in each sample that were assigned to the most-populated nodes of the reference phylogenetic tree (S1 Fig) are shown as rows in this heat map. These rows represent nineteen of the most common species observed, and the color indicates relative abundance, with dark blue indicating zero, and red more than half of the sample. These most-populated nodes account for more than 90% of the bacteria present in the sample without needing to differentiate the case of dominant species from diverse communities in particular areas of the phylogeny. Species names were determined using BLASTN with phylogenetically discriminating reads from the rpoB and rpoC polymerase reads. Complete results are provided in S1b File. Although not based on an alignment, Sequedex analysis of shotgun metagenomic reads allows the comparison of between-sample, or beta, diversity (Fig 2). In practice, highly populated reads typically occur in related groups, discussed above and shown in S1 Fig, so it is possible to assign these groups of nodes to particular species present in the family by sampling a few dozen phylogenetic marker reads (eg. RNA polymerase) from each node and examining the returned hits with BLASTN from NCBI’s nonredundant database. These species names are provided to the left of the heat map in Fig 2. Complete results across all phylogenetic categories and samples is provided as a spreadsheet in S1 File. While the analysis shown in Fig 2 does not differentiate subtle phylogenetic differences between samples, the general similarity of intensities across related nodes provides an indication of the similarity of community composition across samples. This analysis is related to OTU analysis based on alignments of 16S reads, as can be seen in a heat map of the occupancy of the most prevalent OTUs (S2 Fig). As with Fig 2, approximately 90% of the assignable bacterial reads are covered by the analysis in S2 Fig, so we expect similar organisms to be found. Because QIIME places OTUs on a taxonomy instead of a tree, taxonomic assignments are often restricted to the genus or family level, even at an 80% confidence level. Unlike with the RNA polyerase reads used in Fig 2, the results were not typically specific enough to make species assignments. To resolve ambiguous cases, we used species assignments that matched Fig 2 if occurring, provided a genus level assignment if several species matched equally well, or simply picked a well-known species to list if the other matches were named only to the genus level. Another useful OTU analysis enabled by 16S data are the rarefaction curves, shown in Fig 3. We plotted the number of 16S sequences that must be sampled to observe a given number of species. We used the R package ‘iNEXT’ [29] for the rarefaction and pooled individual birds for the three groups. The species richness between the three groups significantly differed based on the 95% confidence intervals of the three groups not overlapping (Fig 3). This analysis distinctly shows the higher diversity of the cloacal samples from immature birds, compared to cloacal samples from mature birds or fecal samples.
Fig 3

16S rarefaction curves.

Within-sample richness, estimated by OTU analysis of 16S sequences with QIIME [32]. The rarefaction curves are computed by sampling the number of 16S sequences indicated on the x-axis and counting the number of distinct observed species, plotted on the y-axis. The shaded areas around the lines represent the 95% confidence intervals. None of the confidence intervals overlap, suggesting that each group significantly differs from the other two even when only 10,000 sequences are considered.

16S rarefaction curves.

Within-sample richness, estimated by OTU analysis of 16S sequences with QIIME [32]. The rarefaction curves are computed by sampling the number of 16S sequences indicated on the x-axis and counting the number of distinct observed species, plotted on the y-axis. The shaded areas around the lines represent the 95% confidence intervals. None of the confidence intervals overlap, suggesting that each group significantly differs from the other two even when only 10,000 sequences are considered. We further analyzed the data rolled up into phyla (from Fig 1) to compare the community composition among the three groups (immature cloacal, mature cloacal, and mature fecal). The three groups significantly differed overall in microbial composition based on phylogenetics (PERMANOVA: R2 = 0.49, P < 0.001; Fig 4a). Each group was significantly different than the other two groups after analyzing the pairwise comparisons (immature cloacal vs. mature cloacal: R2 = 0.30, P < 0.001; mature cloacal vs. mature fecal: R2 = 0.56, P = 0.002; immature cloacal vs. mature fecal: R2 = 0.41, P < 0.001), suggesting a unique microbiome for each group.
Fig 4

Plot of non-metric multidimensional scaling (NMDS) of composition of the three groups.

The three groups significantly differ in their microbial composition for both a. phylogenetic classification, and b. functional gene classification.

Plot of non-metric multidimensional scaling (NMDS) of composition of the three groups.

The three groups significantly differ in their microbial composition for both a. phylogenetic classification, and b. functional gene classification. Anticipating the functional composition results described below, Fig 4b provides an analogous analysis across the 962 functional categories provided by Sequedex. The functional gene categories also significantly differed between the three groups (PERMANOVA: R2 = 0.71, P < 0.001; Fig 4b). Each pairwise comparison was statistically significant (immature cloacal vs. mature cloacal: R2 = 0.64, P < 0.001; mature cloacal vs. mature fecal: R2 = 0.81, P < 0.001; immature cloacal vs. mature fecal: R2 = 0.46, P < 0.001). Both the phylogenetic and functional Sequedex output files used in these analyses are provided in spreadsheet form in S1 File. None of the analyses described above provide insight into fine-scale phylogenetic structures within communities. The shotgun metagenomic reads are not in an alignment, while the 16S reads lack the length and resolution to provide such information. For the mature cloacal samples, however, greater than 80% of microbial population appears to be Delftia spp., and it is possible to create alignments of the RNA polymerase gene, a particularly informative phylogenetic marker gene. The branch length for each of the four genes assembled from each sample differs, indicating the presence of either sequencing errors or genetic diversity in Delftia strains present within the sample (S3 Fig). The results also indicate that individual birds can be discriminated from each other, as the reads for each bird are grouped together in a tree, so it is possible to distinguish birds by where their assembled microbiome reads fall in the tree.

Functional composition of microbiomes

As mentioned above, shotgun metagenomics reads analyzed by Sequedex provide an ability to analyze the functional composition of the metagenome in a manner independent of the phylogenetic analysis. Fig 4b illustrates that both analyses provide a similar ability to discriminate among samples. In doing the Sequedex analysis, we chose to use SEED subsystems to characterize the functional complement of genes present in the samples in part because they provide a hierarchical classification of protein function. Although high-level categories are generally within a factor two of representation in the genomic DNA of the community, several differences are evident by inspection, such as the increased protein metabolism in the fecal samples and the increased virulence and transport-related genes in the cloacal samples. At the highest resolution of the SEED classification, several notable subsystems are enriched in the different sample types. Metabolism and cell wall subsystems increase notably in fecal samples, while virulence, motility, membrane transport, and sulfur metabolism increase in the mature cloacal samples (Fig 5b). Nitrogen metabolism is elevated in immature cloacal samples, while phages are greatly elevated in the Bacteroidetes-dominated fecal sample at right, when compared to the Firmicutes-dominated fecal samples (Fig 5c).
Fig 5

A. Functional rollup of read-based classification of the DNA or RNA in each sample. Classification was performed on each read with Sequedex and the SEED subsystems. The legend at right provides the color code for all 28 functional categories, while labels on the plot indicate highly populated categories as well as those that change significantly between types of samples. Complete results for the functional rollups are provided in Supporting Information S1d File. B. Subsystems associated with fecal metabolism. Selected subsystems distinctive of the fecal microbiome are plotted, with the y-axis measuring the abundance of functionally identified reads associated with the particular subsystem, for each of the samples. C. Subsystems associated with cloacal metabolism. Selected subsystems distinctive of the cloacal microbiome are plotted, with the y-axis measuring the abundance of functionally identified reads associated with the particular subsystem, for each of the samples.

A. Functional rollup of read-based classification of the DNA or RNA in each sample. Classification was performed on each read with Sequedex and the SEED subsystems. The legend at right provides the color code for all 28 functional categories, while labels on the plot indicate highly populated categories as well as those that change significantly between types of samples. Complete results for the functional rollups are provided in Supporting Information S1d File. B. Subsystems associated with fecal metabolism. Selected subsystems distinctive of the fecal microbiome are plotted, with the y-axis measuring the abundance of functionally identified reads associated with the particular subsystem, for each of the samples. C. Subsystems associated with cloacal metabolism. Selected subsystems distinctive of the cloacal microbiome are plotted, with the y-axis measuring the abundance of functionally identified reads associated with the particular subsystem, for each of the samples. A birds-eye (so to speak) view of how all 962 SEED subsystems are represented in the thirty samples is shown in Fig 6. While no distinctly ‘immature cloacal’ functional subsystems can be identified, subsystems elevated in fecal samples are primarily associated with anaerobic metabolism and growth, while subsystems elevated in mature cloacal samples are primarily associated with nitrogen metabolism and surviving in the presence of toxins. Highly abundant SEED subsystems found in the California condors include Acetyl-CoA fermentation to butyrate, methionine degradation, glycolysis and gluconeogenesis, and several different translation elongation factor subsystems. The complete breakdown of functional categories across all thirty samples is provided in spreadsheet form in S1 File.
Fig 6

SEED subsystem with significant changes between sample types.

Z-scores for the difference between fecal and immature cloacal samples on the x-axis are plotted against z-scores for the difference between mature and immature cloacal samples on the y-axis for the two thirds of the most-occupied functional categories. For clarity, only the most populated 2/3 of the SEED subsystems are shown, with the filled magenta squares representing the most populated 1/3 of subsystems and the clear squares are the middle group. Using the clusters of samples defined with the PCA analysis, specific subsystems are identified according to their relative representation in fecal vs. immature cloacal samples (x-axis) and mature vs. immature cloacal samples (y-axis) in units of difference divided by the standard deviation estimated across samples. These categories provide greater resolution to the changes identified in the rollups, and complete results are provided in S1d File.

SEED subsystem with significant changes between sample types.

Z-scores for the difference between fecal and immature cloacal samples on the x-axis are plotted against z-scores for the difference between mature and immature cloacal samples on the y-axis for the two thirds of the most-occupied functional categories. For clarity, only the most populated 2/3 of the SEED subsystems are shown, with the filled magenta squares representing the most populated 1/3 of subsystems and the clear squares are the middle group. Using the clusters of samples defined with the PCA analysis, specific subsystems are identified according to their relative representation in fecal vs. immature cloacal samples (x-axis) and mature vs. immature cloacal samples (y-axis) in units of difference divided by the standard deviation estimated across samples. These categories provide greater resolution to the changes identified in the rollups, and complete results are provided in S1d File.

Discussion

In this study we sampled captive California condors to determine individual variation in the gut microbiome of a captive scavenging species. Previous studies have found that the bacterial repertoire in wildlife depends mainly on diet and host phylogeny [30, 31] as well as whether avian samples are fecal or cloacal [32]. At this point in time, wild condors are not available for comparison. However, we predict that wild condors will have a more diverse microbiome than captive individuals. We found that immature condors had much more diverse microbiomes than adult birds. This is unlike poultry and Chinstrap Penguins (Pygoscelis antarctica), in which microbiome diversity increases with age [23, 24]. However, it does follow the results of Teyssier et al. [6] who found cloacal microbiota of Great Tit nestlings decreased in diversity and increased relative abundance of Firmicutes between 8 and 15 days of age. While the process for the differences of the microbial community is unclear, it will be important to continue to compare species for the development of microbial communities with age of individuals. The two methods (16S and metagenomic) largely agree on the ~ 20 species that comprise approximately 89% of the assignable genomic reads found in the condor samples. However, the two methods disagree significantly on the relative abundance of each species in the bacterial community. One of the largest discrepancies is in the prevalence of Delftia in the mature cloacal samples, which the shotgun metagenomics identifies as 80% of mature cloacal samples, but which barely registers in the 16S data (compare Figs 1 and 2 to S2 Fig). Given our ability to assemble multiple copies of the Delftia RNA polymerase gene from these samples, it appears far more likely that the shotgun metagenomic estimate is the more reliable, presumably due to the lack of primer-specific amplification biases. This behavior is consistent with other studies comparing species abundance between targeted 16S sequencing and shotgun metagenomics approaches [33]. The complete distribution of OTUs across the condor samples is provided in S1c File, and a FASTA format file with the representative 16S sequences for each OTU in S1c is provided as S2 File. Grond et al. [12] provides a recent review of studies on avian gut microbiota. In this work, they note the broad agreement across studies of the presence of firmicutes, bacteroidetes, proteobacteria, and actinobacteria, consistent with our findings. In view of the significant discrepancies of 16S profiles compared to shotgun metagenomic, it appears that caution is necessary when making more detailed comparisons across host species and environmental conditions until more studies characterize the repeatability of various sampling methods across samples. Additionally, the increased phylogenetic resolution afforded by 150 base pair reads of phylogenetic marker genes over that of 16S reads would improve comparison across papers from phylum- or genus-level to species-level comparisons. Roggenbuck et al. [14] describe several near-neighbors to pathogens in vultures: Clostridia and Fusobacteria dominated vultures’ fecal and hindgut samples. Some members of these bacterial classes are pathogenic to other vertebrates, and may also be pathogenic to vultures. If so, it is likely that vultures tolerate toxins from pathogenic bacteria obtained from their carrion diet because they benefit from breakdown of carrion by those same bacteria. We found the classes Clostridia and Fusobacteria to dominate fecal samples from California condors, highlighting the importance of the scavenger lifestyle in captivity as well. In a recent microbiome study of the microbes in Large-Billed or Jungle Crow (Corvus macrorhynchos), it was found that Jungle Crows not only acquire pathogenic microbes, but they return the microbes to the environment via feces [34]. If this pattern is similar for wild California condors, they may play an important role in spreading pathogens in the environment. Thus, this species, as well as vultures, could be used as sentinel species to understand the pathogens present in certain ecosystems. The California condors we sampled in this study showed considerable diversity but share a common core of microbes. This could be due to the captive birds being fed the same food but is more likely related to the birds’ shared physiology. Roggenbuck et al. [14] found captive vultures from a zoo house together shared gut microbes with wild vultures while differing from non-vulture birds sharing their captivity and diet. Like vultures, condors are detritivores, and the condors used in this study were all fed the same raw diet. The observed variation in microbiomes could be due to varying preferences for lean or fatty meat types or could be caused by a condor’s social status leading to differences in diet among animals in the same enclosure. In addition, as we were only able to sample DNA for this study the diversity found may be under-representative of that actually present. Much greater differences are expected for mRNA samples, in which only genes being utilized in the particular environment sampled will be transcribed. Fusobacteria are obligate anaerobes that ferment amino acids, peptides, and carbohydrates that are often found in the mucus layers of humans and animals [35]. The phylogeny of the genus Fusobacteria is shown in S1 Fig, and greater than 99% nucleotide identity, was observed for RNA polymerase reads to all of the following species: Fusobacterium varium, which is a gastrointestinal species; F. ulcerans, which has been isolated from tropical ulcers [35]; and F. mortiferum, which is an anaerobic human pathogen that ferments carbohydrates [36]. The surface of the cloaca is a mucus membrane in which commensal or pathogenic bacteria may be found [37]. It provides both the first line of defense and an area where bacteria such as Lactobacillus spp. can attach to the host. Delftia spp. and Comamonas spp. are microorganisms that often comprise the bulk of bacteria in biofilms associated with wastewater treatment and river wetlands [38, 39]. The organisms in the majority of the condor samples that are highlighted here are associated with robust biofilm formation, antibiotic resistance, and eye infections [40-42]. As noted previously, many of the organisms found in meat-eating birds are closely related to bacteria isolated from humans and birds [14]. The bulk of bacteria in the fecal microbiomes are in the classes Clostridia and Fusobacteriia, which follows the results of the study by Videvall et al. [32], who compared fecal and cloacal microbiomes in ostriches (Stuthio camelus). Specifically, for Clostridia, the predominant species of most of the fecal samples in this study, with reads having greater than 98% nucleotide identity to RNA polymerases in reference strains, was Clostridium perfringens. Clostridium perfringens is a poultry pathogen that causes necrotic enteritis if predisposing factors are present [43]. From our functional analyses (see Fig 4), it is clear the functional categories are distinct between fecal and cloacal samples. While the classes Clostridia and Fusobacteriia make up the bulk of the bacteria in the fecal microbiomes, the genus Delftia makes up the bulk of the bacteria in the mature cloacal microbiomes. This variance we see between the fecal and mature cloacal microbiomes is also supported by Videvall et al. [32], who concludes that it is best to use fecal samples as a representative of the colon in birds and that fecal and cloacal samples only represent the lower digestive tract of the colon. Our use of both fecal and cloacal samples provides complementary information on the California condor microbiome. Some organisms identified in California condors are quite close to reference genomes despite being novel species. Examples of these species follow. Ninety-three percent amino acid identity was observed to be Propionibacterium acnes, a non-spore-forming, Gram-positive, opportunistic pathogen that is part of normal human microflora and found in the skin, oral cavity, GI, and genito-urinary tracts [44] Bacteroides fragilis reads were found with greater than 99% nucleotide identity to the RNA polymerase of reference genomes in both fecal and cloacal samples. Bacteroides fragilis is an obligate anaerobe that is a normal component of human gut flora, but is a common cause of abscesses and bacteremia [45-47]. Ninety-seven percent amino acid identity of RNA polymerase reads was observed to Eggerthella YY7918, a species that is often part of the human intestinal flora and has been identified as a cause of severe [48]. This species was found at a much higher abundance in the cloacal samples than the fecal ones. Shigella flexneri is a non-motile facultative anaerobic gram-negative pathogen that causes dysentery in humans [49, 50] and is increasingly antibiotic resistant [51]. Other organisms identified in California condors are not very close to reference genomes. For example, Shingella flexneri was found at levels of 5–10% in the condor fecal and immature cloacal samples. Consistent with Fig 3, the fecal and mature cloacal samples have considerably less within-sample diversity than the immature cloacal samples. While this method is free of some biases, the presence of sequencing errors in 16S data can lead to significant over-estimation of the number of distinct sequences in a sample. The species richness is lower than that observed by comparable studies in some bird species (see [9]), but is consistent with that observed in vultures [14].

Functional composition of microbes

Functional profiles separate the microbial communities in a manner consistent with phylogenetic profiles, and with a similar dispersion among samples (Fig 4). We also showed our ability to identify the determinants of functional classification, both at the coarse-grained level (fecal samples have more protein and carbohydrate metabolism genes, while cloacal samples show more transport and motility genes) and at the fine-grained level (fecal samples show more replication and methionine degradation, while cloacal samples show more nitrate metabolism and anti-toxin systems). Genome inventories of bacteria are often quite different from their near-neighbors, or not-so-near-neighbors, which happen to have been sequenced and deposited in Genbank. Wu et al. 2015 [39] describe the importance of nitrogen metabolism to the biofilm lifestyle of Comomonas, and it is natural to wonder if the Delftia species we see as a major component of the mature cloacal samples contains nitrogen reduction capabilities. The two copies of the alpha, beta, and gamma subunits of nitrate reductase provided in S3 File indicate that this phenotype is present in the Delftia species inhabiting the condors we evaluated. Following the example of Roggenbuck, et al. [41], we looked for Clostridium spp. toxins in our samples, and do present partial sequences of the alpha toxin in S2 File. While more investigation is needed to understand why we did not see the entire gene, we were able to recover significant quantities of colicin sequences, and they are recorded with their sequences. While such detailed analysis is quite laborious, it appears likely to be necessary if functional characterization of microbiomes is to reach a useful level of detail, as described by Grond et al. [12]. Determining the phylogenetic composition of a natural microbial community from the sequence of fragments of DNA is complex both because of the great variability in the mapping between sequence similarity and phylogenetic similarity for different genomic fragments, and because natural microbial communities often contain organisms that are quite distinct from anything that has been cultured, sequenced, and placed in a reference database [28] (see S3 Fig for a phylogenetic tree of the Delftia RNA polymerase genes recovered from cloacal samples compared to those from nearby reference genomes used in the construction of Sequedex’s database). We chose to compare the samples with conserved signature peptides for which the level of phylogenetics has been characterized, in order to address both of these issues. These concepts are discussed at length in the Sequedex paper [28]. Sequedex uses 300,000,000 amino acid signatures of length 10 to assign genomic fragments individually to the 2550 nodes on its one-per-species phylogenetic tree of life. In each case, further investigation can ascertain the phylogenetic composition with more accuracy.

Conclusion

We characterized gut microbiota from healthy captive California condors to facilitate the design of future studies and to demonstrate novel methods that can be applied to other systems. Useful microbiome signatures were identified by comparing samples within and between individual condors, using both standard and novel methods. In sampling captive and wild animals, one must make a number of choices about sample type: sequencing DNA or RNA, the necessary depth of sequencing, sequencing 16S reads or at random, the types of birds (e.g. sick or healthy, captive or wild, young or old) to be differentiated, if sampling will be done in a time series analysis or at isolated points in time, and the number of locations to be sampled from. In this work, we were able to identify distinct species of microbes in cloacal and fecal samples, as well as interpretable differences in the functional complement of genes in the two groups of samples. One of the most interesting findings from this study is the 40% higher diversity in the immature birds compared to adult condors. The Sequedex software package has enabled rapid analysis of such data sets in a few hours on a laptop computer. It provides both readily interpretable output files profiling both phylogeny and function without complex, non-linear processes such as assembly, and the annotation of raw reads to enable follow-up confirmatory analysis. When combined with appropriate sampling of, for example, a sentinel species at the top of the food web, the potential to gather information rich data across the globe will serve an acute need in monitoring climate change impacts on ecosystems in order to guide mitigation efforts. Furthermore, if successful, this work opens the possibility of using microbiome profiles and disease surveillance of other sentinel species to provide urgently needed specific signatures of ecosystem health [52].

Materials and methods

Ethics statement

The methods used in this study were approved by the Los Alamos National Laboratory (LANL) Institutional Animal Care and Use Committee and did not affect the health and welfare of the animals. All State and Federal wildlife and captive breeding permits are held by the Peregrine Fund (Boise, ID). All laboratory procedures for biological samples were approved by the LANL Biosafety Committee.

Sample collection

Our study samples are from 22 captive adults and immature (hatch year) birds from the rearing facility of the Peregrine Fund in Boise, Idaho. Since 1996, the Peregrine Fund has been breeding and raising California condors in captivity for release to the wild. The sampled condors lived in captivity since hatching, were fed a controlled diet of mostly rodents, rabbits, and commercial bird of prey diet, and were housed in a single building. All of the birds were housed together and are the same diet daily. While birds may be housed separately between cages, due to the open structure of the facility and movement of the birds, they can be considered being raised together without isolation from each other in regards to exposure to microbes and eating the same diet. Fecal samples were collected opportunistically from October 2013 to early January 2014, from both adult and immature condors within 30 minutes of the time of defecation. Cloacal samples were collected during routine health checks by the facility personnel and veterinarians. Both fecal (n = 10) and cloacal (n = 20) samples were placed in sterile plastic specimen tubes when collected and stored in a -20°C freezer within 10–30 minutes of being acquired. Cloacal samples were saved on dry flocked swabs (HydraFlock, Puritan Guilford, ME) in the sterile plastic specimen collection tubes with no reagent. Sampling strategy was designed primarily by sequencing costs and number of birds available. Ten condor fecal samples (9 immature birds and 1 adult) and 20 cloacal samples (15 immature birds and 5 adults) were taken for analysis from different individuals for a total of 30 samples sequenced.

DNA isolation

Immediately prior to analysis, frozen fecal samples were diluted with water and vortexed. Cloacal swabs were resuspended in water by vortexing. Lysis buffer from the DNA Fungal/Bacterial DNA isolation kit (Zymo Research, Irvine, CA) was added to the samples, and lysis was performed for 40 seconds in ZR bead bashing tubes (Zymo Research, Irvine, CA) using the Fastprep 24 instrument (MPBio). Cloacal DNA was purified using the same kit (Quick DNA Fungal/Bacterial Miniprep Kit, Zymo Cat. #D6005), as per the manufacturer’s protocol. DNA was quantified using Qubit instruments and ranged from below the level of detection to 60 ng/ul. The integrity of DNA was evaluated using 1% agarose gels with Lambda DNA/HindIII Marker (Thermo Fisher Scientific, Cat. #SM0102).

Library preparation and sequencing

For shotgun metagenomics samples, Illumina libraries were prepared using NEBNext Ultra DNA Library Preparation Kit (New England Biolabs, Cat. #E7370S) per the manufacturer’s protocol. Due to the varying amounts of DNA extracted, the input amount of DNA ranged from 100 ng to less than 1 ng. DNA was fragmented using a Covaris E220, and the ends were made blunt and adapters and indexes were added onto the ends of the fragments prior to PCR enrichment. For samples with inputs below 10 ng of DNA, the number of PCR cycles was increased from the manufacturer’s protocol of 12 to 15. The Illumina metagenomic libraries were cleaned up using AMPure XP beads (Beckman Coulter, Cat. #A63881) before being eluted in DNA Elution Buffer (Zymo Research, Cat. #D3004-4-10). The concentration of the libraries was obtained using the Qubit dsDNA HS Assay (ThermoFisher Scientific, Cat. #Q32854). The average size of the library was determined by the Agilent High Sensitivity DNA Kit (Agilent, Cat. #5067–4626). An accurate library quantification was determined using the Library Quantification Kit–Illumina/Universal Kit (KAPA Biosystems, Cat. #KK4824). The libraries were sequenced on the Illumina NextSeq generating paired-end 151 bp reads. For 16S sequencing, degenerate primers that amplify the V4 region of bacterial rDNA genes were used (515–806 pair). Due to the varying amounts of DNA extracted, the DNA input ranged from 12.5 to less than 1 ng. The first round of PCR amplified the V4 region and added unique tags to the amplicon, using a denaturation temperature of 95°C for 3 minutes, 25 cycles at 95°C for 30 seconds, 55°C for 30 seconds and 72°C for 30 seconds, followed by an extension of 72°C for 5 minutes before holding at 4°C. The second round of PCR added Illumina specific sequencing adapter sequences and used a denaturation temperature of 95°C for 3 minutes, 8 cycles of 95°C for 30 seconds, 55°C for 30 seconds, and 72°C for 30 seconds, followed by an extension of 72°C for 5 minutes before holding at 4°C. The amplicons were cleaned up using AMPure XP beads (Beckman Coulter, Cat. #A63881). A no-template control was processed but did not show a band in the V4 amplicon region and was not further processed. Unique tags allowed for multiple amplicons to be pooled together. The concentration of the amplicon pool was obtained using the Qubit dsDNA HS Assay (ThermoFisher Scientific, Cat. #Q32854). The average size of the library was determined by the Agilent High Sensitivity DNA Kit (Agilent, Cat. #5067–4626). An accurate library quantification was determined using the Library Quantification Kit–Illumina/Universal Kit (KAPA Biosystems, Cat. #KK4824). The amplicon pool was sequenced on the Illumina MiSeq generating paired end 301 bp reads.

Genomic analysis

Shotgun metagenomes were analyzed for phylogenetic and functional content according to the 2550-species tree of life, and 963 SEED functional categories utilized with the Sequedex package [28] (http://sequedex.lanl.gov/). Sequedex is a signature-peptide based analysis that assigns both phylogeny and function independently to each read based on its database of 300 million signature peptides. It avoids many of the common pitfalls of assembly-based, blast-based, or gene-based analysis methods and is extensively described in Berendzen et al. 2012. While techniques exist to attempt to identify individual component taxa in a complex sample [53], we chose in our case to simply recruit reads from these samples to amino acid reference sequences for the RNA polymerase genes, using the reading frame for each read identified by Sequedex. In this case we can assemble a phylogenetic marker gene long enough for detailed phylogenetic analysis, and provide some indication of whether different birds have distinct species present. A direct evaluation of the functional profile of the microbiome was obtained from the read-based functional assignment to the hierarchical, SEED subsystem classification scheme, using Sequedex as well [28]. As with phylogeny, binning of reads into generally-defined functional categories can be misleading. The by-read annotation provide by Sequedex allowed us to assay for not just total representation of particular SEED subsystem, but also greatly facilitated our ability to retrieve specific genes from shotgun metagenomic samples. We demonstrated this ability on five representative genes: the phylogenetic marker rpoB, a metabolic niche gene, nitrate reductase, a beta lactamase, and a Clostridium toxin. The S3 File provides the sequences of example genes. In phylogenetic signature methods, high-significance patterns that are shared across taxa are identified with positions on a pre-calculated tree via a Least Common Ancestor algorithm. In principle, it matters little whether these signatures are genetic or anatomical. However, for genetic signatures, some patterns appear in multiple species for reasons other than phylogeny (e.g., triplet repeats of DNA that may be artifacts of the replication mechanisms, or horizontally transferred genes appear because of common ecological niches). In our analysis, those candidate signatures that are part of a repetitive region of sequence were discarded, and the classifying power of the remaining set of signatures were characterized. Good signature sets have both high selectivity and high sensitivity. We characterized the signature set we used as having similar sensitivity to BLASTX, while having more than an order of magnitude higher selectivity when used for phylogenetic classification of short reads such as those used in this experiment [28]. The functional profiles from Sequedex provided support and additional insight to the niche determinants derived from the microbiology of the particular species of organisms found in each sample. Both phylogenetic assignments and genome inventory analyses were performed on the samples, allowing the bacteria identified to be separated into three probable groups: microbes of importance to digestion, those associated with mucosal environments, and potential pathogens of interest. In this work, we chose to identify specific reads from a phylogenetic marker gene, the RNA polymerase, for further examination using BLASTN against the non-redundant database at NCBI. Sequence reads corresponding to a phylogenetic marker gene, RNA polymerase, were identified from each of these nodes and assigned a taxonomy using NCBI to blast a nucleotide fragment (blastn) against the nucleotide (nt) database. After examination of about a dozen reads in this way, species can be identified for each set of nodes. By examining the SEED subsystems identified in S1e File, as well as the number of reads assigned to particular Pfam families in S1f File, we were able to find samples and genes likely to have high enough prevalence to be assembled and compared to the literature. We provide two examples of beta lactamases in S2 File, where a reference gene was used as a template, and translated peptide sequences were matched to it by identifying exact 10-mer matches to the reference. Two examples were recruited for each reference, in order to provide some idea of the variability of the resulting sequence that is recruited by this method. The analysis method we use here, Sequedex, is an example of a signature-based analysis method [28]. Signature methods have a distinguished history across a range of scientific disciplines as a way of identifying genes of value in situations where application of fundamental theory may be complex. Hidden Markov Models of amino-acid sequences are an example of signature methods, in contrast with BLAST, which is usually used as a profiling method.

Statistical analysis

We compared measures of community structure between the three groups (age and sample type), including species richness, species diversity, and community composition. For species diversity, we used Simpson’s diversity index. We tested for differences in richness and diversity between these groups using one-way ANOVAs followed by Tukey’s post-hoc test to test for pairwise differences between groups. Assumptions of normality and equal variances for parametric tests were met. Community composition among the three groups (age and sample type) were compared using non-metric multidimensional scaling (NMDS) using the metaMDS function in the vegan package (version 2.4–4; [54]) in R (version 3.4.1; [55]). Two separate NMDS ordinations were done; one based on phylogenetics and the other based on functional gene categories. To test for overall differences among groups, Bray-Curtis distances were calculated and then examined using Permutational Multivariate Analysis of Variance (PERMANOVA). The adonis function (vegan) was used with 1000 permutations. To test for differences between each group, pairwise PERMANOVAs were run, each with 1000 permutations, and resulting p-values were compared to a Bonferroni corrected alpha of 0.017. A better understanding of the diversity of organisms present in the samples can be obtained through phylogenetic analysis. We compared separate copies of RNA polymerase genes assembled from either the forward or reverse reads in each sample. The assembled genes are closest to the two Delftia reference genomes, with one of the two being Delftia acidovorans and the other reference not assigned a species name. The observed phylogenetic separation of the genes from the condor microbiome appear to match the reference Delftia at the genus level but not at the species level. The four genes assembled from each sample cluster together on the tree, indicating the organisms from each bird are genetically distinct from each other. The 16S data analysis was performed with QIIME version 1.8 [56] with representative 16S sequences are provided in S2 File, and were selected using the QIIME utility, single_rarefaction.py, and a value for a sequence similarity parameter of 0.94. Annotations of 16S sequences were made by comparing to the 97_otus.fasta database.

Phylogenetic Trees of Four Bacterial Genera in condors.

Close-ups on reference tree of reads recruited for four representative samples, highlighting a region of the phylogenetic tree near (top-left) Fusobacterium mortiferum, (top-right) Clostridium perfringens, (bottom-left) Lactobacillus johnsonii and (bottom-right) Propionibacterium avidum. Selected reads from a phylogenetic marker gene (RNA Polymerase) were examined with Blastn or Blastp against the non-redundant databases at NCBI to identify the most closely related reference organism. (PNG) Click here for additional data file.

Results of 16S OTU analysis.

The fraction of the sample represented in the indicated OTU is shown as a heat map with a factor two ratio between adjacent colors, which run from dark blue to red. The rows are sorted phylogenetically, while the columns are in the same order as in Table 1 and Fig 1, but with the two samples for which the 16S samples failed left out of the heat map. The columns represent different samples in the same order as in Fig 2, and the rows represent the different nodes of the phylogeny. To the left of the heat map, we provide the most specific phylogenetic assignment QIIME made at the 80% confidence level, grouped phylogenetically in the same order as in Fig 2. These phylogenetic assignments correspond to the 16S sequences deposited in the Sequence Read Archive for this study. Representative 16S sequences are provided in S2 File, and are numbered according to the identifiers provided on the left of the heat map above. To the right of the heat map are the species names inferred from the 16S sequence and knowledge of what species were found with the RNA polymerase nucleotide reads identified by Sequedex from the shotgun sequencing data Similar to the process used to make Fig 2, representative sequences for each OTU were compared to the Ribosomal DataBase project using BLASTN to identify the species provided to the right of the Figure. (PNG) Click here for additional data file.

The phylogeny of condors’ microbiomes, comparing genes assembled from each sample to nearby genes from the Sequedex reference tree.

Phylogenetic tree of the RNA polymerase genes from reference genomes phylogenetically near to Delftia, and four assembled RNA polymerase genes from each of five mature cloacal samples. The sample numbers ‘CCP.xx’ refer to Table 1, while the ‘1’ or ‘2’ refer to forward and reverse reads, which were kept separate. Two equivalent assemblies were made to both the forward and reverse reads, and they are referred to as ‘a’ and ‘b’. The alignment used to compute the tree is provided as a Supplementary text file. S3 Fig shows a phylogenetic tree resulting from such analysis. (PNG) Click here for additional data file.

Subsystems associated with stress response.

Selected subsystems distinctive of the stress response and efflux of toxic compounds are plotted, with the y-axis measuring the abundance of functionally identified reads associated with the particular subsystem, for each of the samples. (TIF) Click here for additional data file.

Subsystems associated with fecal metabolism.

Selected subsystems distinctive of the fecal microbiome are plotted, with the y-axis measuring the abundance of functionally identified reads associated with the particular subsystem, for each of the samples. (TIF) Click here for additional data file.

Spreadsheet with six tabs (a-f) of phylogenetic and functional classifications by various metrics.

(a) Phylogenetic rollup using Sequedex to classify shotgun metagenomic reads. (b) Normalized counts of reads assigned to each node of the Sequedex phylogenetic tree. (c) Operational taxonomic unit (OTU) counts for the 16S reads from each sample. (d) Functional rollup using Sequedes to classify shotgun metagenomic reads according to the high-level SEED classifications. (e) Normalized counts of shotgun metagenomic reads assigned to each SEED subsystem. (f) Normalized counts of shotgun metagenomic reads assigned to each Pfam family. (XLSX) Click here for additional data file.

Representative 16S sequences of OTUs, identified by the labels provided in Supplementary information S1c, provided in fasta format.

(XLSX) Click here for additional data file.

Representative genes assembled from open reading frames identified by Sequedex, in a spreadsheet format.

(XLSX) Click here for additional data file.

Supplemental information text for additional descriptions for the figures and data files.

(DOCX) Click here for additional data file.

Transfer Alert

This paper was transferred from another journal. As a result, its full editorial history (including decision letters, peer reviews and author responses) may not be present. 27 Jul 2019 PONE-D-19-18040 California Condor Microbiomes: Bacterial Variety and Functional Properties in Captive-Bred Individuals PLOS ONE Dear Dr. Fair, Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process. Both the reviewers and I think the manuscript has a lot of merit, although a number of changes were recommended.  In particular, there are some places in the manuscript that could be removed to shorten it, such as the description of sequencing in the introduction.  This may help to better focus the paper. We would appreciate receiving your revised manuscript by Sep 10 2019 11:59PM. When you are ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file. If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. To enhance the reproducibility of your results, we recommend that if applicable you deposit your laboratory protocols in protocols.io, where a protocol can be assigned its own identifier (DOI) such that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols Please include the following items when submitting your revised manuscript: A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). This letter should be uploaded as separate file and labeled 'Response to Reviewers'. A marked-up copy of your manuscript that highlights changes made to the original version. This file should be uploaded as separate file and labeled 'Revised Manuscript with Track Changes'. An unmarked version of your revised paper without tracked changes. This file should be uploaded as separate file and labeled 'Manuscript'. Please note while forming your response, if your article is accepted, you may have the opportunity to make the peer review history publicly available. The record will include editor decision letters (with reviews) and your responses to reviewer comments. If eligible, we will contact you to opt in or out. We look forward to receiving your revised manuscript. Kind regards, Suzanne L. Ishaq, PhD Academic Editor PLOS ONE Journal Requirements: When submitting your revision, we need you to address these additional requirements. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at http://www.journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and http://www.journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf 1. Thank you for including your competing interests statement; "The authors have declared that no competing interests exist." We note one ore more authors are affiliated to Viome, and GenerisBio. Please provide an amended Funding Statement declaring this commercial affiliation, as well as a statement regarding the Role of Funders in your study. If the funding organization did not play a role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript and only provided financial support in the form of authors' salaries and/or research materials, please review your statements relating to the author contributions, and ensure you have specifically and accurately indicated the role(s) that these authors had in your study. You can update author roles in the Author Contributions section of the online submission form. Please also include the following statement within your amended Funding Statement. “The funder provided support in the form of salaries for authors [insert relevant initials], but did not have any additional role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript. The specific roles of these authors are articulated in the ‘author contributions’ section.” If your commercial affiliation did play a role in your study, please state and explain this role within your updated Funding Statement. 2. Please also provide an updated Competing Interests Statement declaring this commercial affiliation along with any other relevant declarations relating to employment, consultancy, patents, products in development, or marketed products, etc. Within your Competing Interests Statement, please confirm that this commercial affiliation does not alter your adherence to all PLOS ONE policies on sharing data and materials by including the following statement: "This does not alter our adherence to  PLOS ONE policies on sharing data and materials.” (as detailed online in our guide for authors http://journals.plos.org/plosone/s/competing-interests). If this adherence statement is not accurate and  there are restrictions on sharing of data and/or materials, please state these. Please note that we cannot proceed with consideration of your article until this information has been declared. Please include both an updated Funding Statement and Competing Interests Statement in your cover letter. We will change the online submission form on your behalf. Please know it is PLOS ONE policy for corresponding authors to declare, on behalf of all authors, all potential competing interests for the purposes of transparency. PLOS defines a competing interest as anything that interferes with, or could reasonably be perceived as interfering with, the full and objective presentation, peer review, editorial decision-making, or publication of research or non-research articles submitted to one of the journals. Competing interests can be financial or non-financial, professional, or personal. Competing interests can arise in relationship to an organization or another person. Please follow this link to our website for more details on competing interests: http://journals.plos.org/plosone/s/competing-interests [Note: HTML markup is below. Please do not edit.] Reviewers' comments: Reviewer's Responses to Questions Comments to the Author 1. Is the manuscript technically sound, and do the data support the conclusions? The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented. Reviewer #1: Partly Reviewer #2: Partly ********** 2. Has the statistical analysis been performed appropriately and rigorously? Reviewer #1: Yes Reviewer #2: Yes ********** 3. Have the authors made all data underlying the findings in their manuscript fully available? The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified. Reviewer #1: Yes Reviewer #2: Yes ********** 4. Is the manuscript presented in an intelligible fashion and written in standard English? PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here. Reviewer #1: Yes Reviewer #2: Yes ********** 5. Review Comments to the Author Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters) Reviewer #1: Dear authors, I have had the pleasure of reviewing your paper for the second time, and want to thank you for taking my previous comments into account. The manuscript shows interesting and important results by comparing different sequencing methods in microbiomes of captive condors. I have a few overall comments, and below specific, line by line comments. Overall, it is important to mention early in the manuscript that metagenomic sequencing represents potential functional profiles, and not functions actually occurring at the time of sampling. This is mentioned somewhere in the discussion, but should be moved up. Also, use of the word species for bacteria is complicated. If the authors used OTUs for clustering of 16S reads, they should be referred to as such. I hope my comments help improve the manuscript. Sincerely, Kirsten Grond kirsten.grond@uconn.edu Introduction The introduction is very thorough, which makes it long. If this is not a problem with the journal, leave it like this. If it is, you could possibly move part of the sequencing description in L114-126 to the methods section Results: L139. Add references or discuss this in the Discussion instead. L168-170. Remove this sentence, as this information is in the figure description. L172-174. Back this statement up with alpha diversity metrics & statistics. L177-183. Can be moved to figure description. L188-189. Remove this sentence L198-202. Move to figure description L204-206. Move to methods L207. polymerase L208-211. If matches are only to genus level, mention genus only and do not pick a well-known species instead (if I understood this right). Using a species that is potentially incorrect misrepresents results. L212-216. The authors should not use rarefaction curves to say anything about diversity. Alpha diversity metrics with appropriate statistical tests should be used. L217-220. Not necessary to describe what was done. Just report results. L222. Test statistic for permanova is R2. L222-224. Instead of using permanova for pairwise comparisons I would suggest using TukeyHSD. When eliminating a variable from permanovas/changing order of variables the results change. L234-235. Seems more like a methodological discussion point than a result L235-236. If the sequencing was deep enough (>100x coverage) and assembled, the authors can check the alignment rate of the raw reads back to the contigs or bins, and the alignment rate would tell if they are getting fine scaled phylogenies. For example, an alignment rate of >90% would be good and likely include most members of the community. A low alignment rate of <50% would indicate incomplete coverage and phylogenies will be missing a large portion of the communities. L238-244. Move to methods. L244-246. Move to figure caption or delete. Only report the results from the analyses here. L249. Based on Delftia strain ID? L251-254. I am unsure what the goal of this paragraph is. State the main findings of figure 4b instead? L257-258. Remove sentence. L262-263. Move this topic to discussion L264-266. What do the authors mean by polarized extremes? L271. Birds-eye! :D L272-275. Move to figure description L283-289. Seems more appropriate in methods. Also the first mention of the genes investigated, which should definitely be in the methods. Discussion The discussion is thorough, but very long. It contains several paragraphs that are part method or can be omitted. I indicated below which sections can potentially be removed. L301. The authors should be consistent in terminology. Metagenomics is also used as a term for 16S in the manuscript, which could be confusing. L302-304. Sentence needs clarification L306-308. Do you know anything about Delftia genome size? If they have large genomes, that could also explain the difference in detection between the methods. Also, sequencing depth is much larger in the metagenomics. L313-315. I don’t think this sentence is necessary. L321-327. Good points! L329-330. Some members of the classes are pathogenic, not the whole class. L330-332. Merge these sentences to improve flow. L333. Fusobacteria L332-348. Super cool that physiology is so selective! L364-365. Remove sentence L371-373. This is a good thing to mention, but would fit better in a paragraph discussing the differences between fecal and cloacal samples. L374-376. Would not report a math on 93% similarity. At most that is a genus level match. L387-397. This describes methods and can be removed from the discussion L398-403. Rewrite this paragraph with alpha diversity results. L405-409. Can be removed L414. Grained L430-431. Remove sentence L440-441. Remove sentence or explain the concepts. L442-452. This would be useful to move to methods, where signatures are explained. L455-463. This section should be in the Results as it states results but does not explain/discuss them. L464-466. Methods L467-478. This paragraph does not appear to add much to the discussion, and could be removed in the interest of shortening this section. Conclusion L480. Sentence reads funky. L485-486. Use as first sentence of the paragraph and include main finding. L500-502. Remove sentence. More samples are always needed, but the authors have a nice sample set here and the last sentence of the manuscript should not take away from that. Methods L542-543. Potential PCR bias in low DNA samples because of the different cycle numbers? Is cycle number a significant predictor for microbiome diversity/community composition? L554-560. Add degree symbol for C L569-570. Need a lot more information on analysis parameters here. Which programs were used for trimming, assembly, annotation, references etc. Did the authors align the raw reads to the TOL database and use those annotations for taxa? If so, there likely are a large number of false ID’s. To avoid this problem, the authors could assemble their reads, and align to that assembly. Any reads that don’t align to those contigs can be annotated for function separately, but should not be trusted for taxa info. L600. Did the authors look if results changes using other distance matrices? Maybe add to supplement? L606. Is Delftia a genus or species? Which genes did you look at? L611. Expand the 16S analyses section. Which OTU-picking strategy and reference database did you use? What % was used for OTU cutoff? Figures Overall: - Non-gradient colors would make the graph easier to read Figure 1 - The labeling on the graph is confusing. There are places with two bacteria per color, and it is unclear whether this is a mistake or if the class is split between them. Would recommend to change the colors (or overlay with pattern) and add separate legend for the most abundant species. - Add label to y-axis. - In the caption it is mentioned that species were identified with among others16S data, which does not have the resolution for species identification. Figure 2 - Why does not every row have something assigned to it? It is difficult to see which species corresponds with what data in the way this figure is presented. Also, expand the figure so full bacterial names are on one line and italicize them. I still don’t understand why the condor reads are in this figure. What does that add to the paper? Figure 3 - Move to supplement. Not essential for main manuscript. Figure 5A - label y-axis Figure 5C - Move the legend to not overlap with the bars in the graph. Figure 6 - increase symbol size and make text more readable. Figure S3. - What do the different colors mean? Reviewer #2: To authors: This paper utilizes two analyses methods to compare the microbiome of cloacal and fecal samples of captive California condors. The strengths of this text are 1. It provides baseline knowledge of how a scavenging bird’s microbiome looks like. 2. It demonstrates that the different analyses approaches can potentially produce dissimilar results. Simple and straightforward writing is not a strength of this text and authors should make an effort to improve the readability of the paper. I hope that this manuscript motivates further research into host-associated microbes in wild scavenger birds. General comments: 1. My main and only big criticism is that you could have used additional variables to explain the observed patterns. It is well known that sex and breeding status con affect the microbiome. Is there a chance you could rerun some of the analyses to further explain your results including at least sex of the bird? 2. Your discussion would benefit if you provide further information on the bacterial transmission/colonization processes of these birds. Specific comments. (L=line) 1. L43: Provide a broader conclusion 2. L49: Should be decade 3. Figure 1: Why are there random bacterial names on top of the bacterial groups? They make the figure harder to read. 4. L180-181: This belongs in the figure legend. 5. L199: Willing to make? I don’t understand. 6. Figure 3: Title is not necessary 7. Figure 3: Why are all the rarefaction curves cut at the same number of sampled sequences? 8. Figure 3: Could you provide the variance in the curves? 9. L258: delete the first are 10. L296-300: Can you provide hypotheses as to why immature condors have a more diverse microbiome? 11. L360-362: Could you provide a reference here? ********** 6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files. If you choose “no”, your identity will remain anonymous but your review may still be made public. Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy. Reviewer #1: Yes: Kirsten Grond Reviewer #2: No [NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files to be viewed.] While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email us at figures@plos.org. Please note that Supporting Information files do not need this step. 22 Oct 2019 Responses to the specific and thoughtful review comments: Introduction The introduction is very thorough, which makes it long. If this is not a problem with the journal, leave it like this. If it is, you could possibly move part of the sequencing description in L114-126 to the methods section. We worked to shorten the introduction to make it more concise, while trying to maintain the flow of the introduction for the work. Results: L139. Add references or discuss this in the Discussion instead. References have been added. L168-170. Remove this sentence, as this information is in the figure description. This sentence has been removed. L172-174. Back this statement up with alpha diversity metrics & statistics. More statistics regarding species richness and true ‘alpha’ diversity have been added to the manuscript as suggested. For alpha diversity, we now refer to Simpson’s diversity and provide appropriate statistics. L177-183. Can be moved to figure description. These sentences have been moved to the figure description as suggested. L188-189. Remove this sentence This sentence has been removed. L198-202. Move to figure description These sentences have been moved to the figure description. L204-206. Move to methods This sentence is defining labels on figure S2, and has been moved to the caption for that figure. L207. Polymerase This misspelling has been corrected. L208-211. If matches are only to genus level, mention genus only and do not pick a well-known species instead (if I understood this right). Using a species that is potentially incorrect misrepresents results. For many genera, the 16S sequences match to a small subset of the species in the genus. Rather than discard all of that information, we selected an interpretable representative of what matched, but warned the reader that the resolution was not always unique to a species, and provided the Fasta file keyed to the OTU representative sequence. It should also be noted that even when only one species from the database is matching, the sequence potentially matches other species, not in the database. So, we feel the added information from picking a representative species from among 2 or 3 (which was typical) offsets the information lost by simply providing the genus. Listing all matching species would be prohibitively cumbersome. L212-216. The authors should not use rarefaction curves to say anything about diversity. Alpha diversity metrics with appropriate statistical tests should be used. We reanalyzed the refraction curves and left Figure 3 in the main text as it seemed more useful with the additional confidence intervals around the curves. The rarefaction curves are computed by sampling the number of 16S sequences indicated on the x-axis and counting the number of distinct observed species, plotted on the y-axis. The shaded areas around the lines represent the 95% confidence intervals. None of the confidence intervals overlap, suggesting that each group significantly differs from the other two even when only 10,000 sequences are considered. In addition, it is more appropriate to state the rarefaction curves are for species richness and this was changed throughout the paper. L217-220. Not necessary to describe what was done. Just report results. This section has been edited so that methods are not included. L222. Test statistic for permanova is R2. The test statistic has been corrected. L222-224. Instead of using permanova for pairwise comparisons I would suggest using TukeyHSD. When eliminating a variable from permanovas/changing order of variables the results change. TukeyHSD will not work on distance matrices or on the output of NMDS analyses. The intent of these analyses is to survey microbial communities in different sample types to see if they differ in terms of their composition. Here, we follow the practice of microbial community ecology even though it may not be perfect. These pairwise PERMANOVAs are common in microbiome studies. We do not have to do pairwise comparisons since there is some resistance to such practices. If the reviewers would rather us not do pairwise comparisons, we would be happy to remove them. Doing the pairwise differences will tell us which groups are different from the others. Even though re-ordering the variables will change the specific results (compared to when all are included), we are still testing the hypothesis that each pair of groups is significantly different from one another. We compare our p-values to a corrected alpha level to avoid Type I errors. L234-235. Seems more like a methodological discussion point than a result It could also fit in the methods, but where it is now it acts as a transition to a description of what our methods couldn’t do and what we did because of that lack. L235-236. If the sequencing was deep enough (>100x coverage) and assembled, the authors can check the alignment rate of the raw reads back to the contigs or bins, and the alignment rate would tell if they are getting fine scaled phylogenies. For example, an alignment rate of >90% would be good and likely include most members of the community. A low alignment rate of <50% would indicate incomplete coverage and phylogenies will be missing a large portion of the communities. We agree this makes sense, but we did not want to get sidetracked assembling complex communities, only to see a slightly different qualitative heuristic. We did not think it made sense for us to assemble things, and we acknowledge it is qualitative. L238-244. Move to methods. As suggested, these sentences have been moved to the methods. L244-246. Move to figure caption or delete. Only report the results from the analyses here. This has been moved to the figure caption. L249. Based on Delftia strain ID? Based on where microbiome reads for each bird fall within the phylogenetic tree – some explanation has also been added to the sentence. L251-254. I am unsure what the goal of this paragraph is. State the main findings of figure 4b instead? This paragraph serves as a transition from the phylogeny-based analysis to a function-based analysis. Figure 4B is mentioned because the two data types are distinct in that figure. L257-258. Remove sentence. This sentence has been removed. L262-263. Move this topic to discussion As suggested, this topic has been moved to the discussion. L264-266. What do the authors mean by polarized extremes? The polarized extremes portion of the sentence has been removed. L271. Birds-eye! :D L272-275. Move to figure description This information has been moved to the figure description. L283-289. Seems more appropriate in methods. Also, the first mention of the genes investigated, which should definitely be in the methods. As suggested, this has been moved to the methods. Discussion The discussion is thorough, but very long. It contains several paragraphs that are part method or can be omitted. I indicated below which sections can potentially be removed. L301. The authors should be consistent in terminology. Metagenomics is also used as a term for 16S in the manuscript, which could be confusing. The manuscript has been read through and edits made to keep metagenomics and 16S as separate, consistent terms. L302-304. Sentence needs clarification This sentence has been removed. L306-308. Do you know anything about Delftia genome size? If they have large genomes, that could also explain the difference in detection between the methods. Also, sequencing depth is much larger in the metagenomics. Delftia acidovorans does have a large genome (6.9 mega-bases) but it also has five copies of the ribosome, which is above average for bacteria. Both 16S and metagenomic analysis are by-read fractions, so the ratio of Delftia to others will be independent of sequencing depth. Also, Delftia is <3% of the sample based on the OUT analysis (supplementary table 1) but is 80% of the sample by metagenomics (Figure 1). This ratio is much larger than is explainable by genome size or ribosomal copy number effects. L313-315. I don’t think this sentence is necessary. This sentence has been removed. L321-327. Good points! L329-330. Some members of the classes are pathogenic, not the whole class. Good point. The wording has been edited. L330-332. Merge these sentences to improve flow. As suggested, the sentences have been merged. L333. Fusobacteria This word has been corrected. L332-348. Super cool that physiology is so selective! L364-365. Remove sentence This sentence has been removed. L371-373. This is a good thing to mention, but would fit better in a paragraph discussing the differences between fecal and cloacal samples. A paragraph mentioning differences between fecal and cloacal samples has been added. L374-376. Would not report a math on 93% similarity. At most that is a genus level match. Here we are trying to show that some of our unknown organisms are close to reference genomes. However, you are correct that it is misleading in how it was written. It has been edited for clarity. L387-397. This describes methods and can be removed from the discussion This paragraph was removed from the paper. L398-403. Rewrite this paragraph with alpha diversity results. We rewrote this paragraph to address the new results we did for species richness. L405-409. Can be removed This section has been removed. L414. Grained This word has been corrected. L430-431. Remove sentence This sentence has been removed. L440-441. Remove sentence or explain the concepts. Added Ben’s description and a reference to Berendzen et al. 2012 L442-452. This would be useful to move to methods, where signatures are explained. This has been moved to methods, as suggested. L455-463. This section should be in the Results as it states results but does not explain/discuss them. Good point. This section has been moved to Results. L464-466. Methods This sentence has been moved to methods. L467-478. This paragraph does not appear to add much to the discussion, and could be removed in the interest of shortening this section. As suggested, this paragraph has been removed. Conclusion L480. Sentence reads funky. The sentence has been edited to read better. L485-486. Use as first sentence of the paragraph and include main finding. This sentence has been used as the first sentence of the paragraph and has been modified to include our main findings. L500-502. Remove sentence. More samples are always needed, but the authors have a nice sample set here and the last sentence of the manuscript should not take away from that. This sentence has been removed. Methods L542-543. Potential PCR bias in low DNA samples because of the different cycle numbers? Is cycle number a significant predictor for microbiome diversity/community composition? We looked at if cycle number was related in any way to the results for the microbiome communities. All of the samples were run together. However, with a few of the samples, there was less DNA extracted than the majority of the samples, which had ample DNA. L554-560. Add degree symbol for C A degree symbol has been added. L569-570. Need a lot more information on analysis parameters here. Which programs were used for trimming, assembly, annotation, references etc. Did the authors align the raw reads to the TOL database and use those annotations for taxa? If so, there likely are a large number of false ID’s. To avoid this problem, the authors could assemble their reads, and align to that assembly. Any reads that don’t align to those contigs can be annotated for function separately, but should not be trusted for taxa info. We used Sequedex extensively. Sequedex avoids all of the complexities and sources of error mentioned by the reviewer. We will add another sentence to the paragraph mentioned to clear up confusion by briefly describing Sequedex and pointing the reader to an extensive description of it (Berendzen et al. 2012). L600. Did the authors look if results changes using other distance matrices? Maybe add to supplement? Bray-Curtis distances are the most robust for community data and are commonly used for microbiome studies. We ran the NMDS and PERMANOVAs again with other types of distance matrices and they all showed the same results. We do not see a need to add additional figures to supplemental data. L606. Is Delftia a genus or species? Which genes did you look at? Delftia is a genus. We looked at the genes in two Delftia reference genomes, one for Delftia acidovorans, and the other for Delftia spp., where the reference was not assigned a species name. L611. Expand the 16S analyses section. Which OTU-picking strategy and reference database did you use? What % was used for OTU cutoff? We addressed this concern in the text of the paper. The 16S data analysis was performed with QIIME version 1.8 [42] with representative 16S sequences are provided in Supplementary Information SI2, and were selected using the QIIME utility, single_rarefaction.py, and a value for a sequence similarity parameter of 0.94. Annotations of 16S sequences were made by comparing to the 97_otus.fasta database. Figures Overall: - Non-gradient colors would make the graph easier to read For each of the figure comments below, we work to improve the figure quality per the recommendations. Figure 1 - The labeling on the graph is confusing. There are places with two bacteria per color, and it is unclear whether this is a mistake or if the class is split between them. Would recommend to change the colors (or overlay with pattern) and add separate legend for the most abundant species. - Add label t o y-axis. - In the caption it is mentioned that species were identified with among others16S data, which does not have the resolution for species identification. Figure 2 - Why does not every row have something assigned to it? It is difficult to see which species corresponds with what data in the way this figure is presented. Also, expand the figure so full bacterial names are on one line and italicize them. I still don’t understand why the condor reads are in this figure. What does that add to the paper? Figure 3 - Move to supplement. Not essential for main manuscript. We actually redid this rarefaction analysis and curve and added confidence intervals. We left his figure in the main section as we felt it added more value now with the new analysis and confidence intervals. Figure 5A - label y-axis Figure 5C - Move the legend to not overlap with the bars in the graph. Figure 6 - increase symbol size and make text more readable. We increased the font size and readability of this figure. We increase the symbol size but then they obstructed the view of the other symbols and labels. Instead, we focused on making the overall figure better in the overall presentation. Figure S3. - What do the different colors mean? Reviewer #2: To authors: This paper utilizes two analyses methods to compare the microbiome of cloacal and fecal samples of captive California condors. The strengths of this text are 1. It provides baseline knowledge of how a scavenging bird’s microbiome looks like. 2. It demonstrates that the different analyses approaches can potentially produce dissimilar results. Simple and straightforward writing is not a strength of this text and authors should make an effort to improve the readability of the paper. I hope that this manuscript motivates further research into host-associated microbes in wild scavenger birds. General comments: 1. My main and only big criticism is that you could have used additional variables to explain the observed patterns. It is well known that sex and breeding status con affect the microbiome. Is there a chance you could rerun some of the analyses to further explain your results including at least sex of the bird? Yes, we agree that there are other variables that may explain some observed patterns. However, regarding sex, the number of males and females are very unbalanced. Because there are 3 types (groups) of samples, we would have to do a test of sex 3 times. The sample sizes for females are way too low compared to males for each group. For example, there are 9 male fecal samples compared to only 1 female. For cloacal immature samples, 12 are males and only 3 are females. Because of this unbalanced design, we chose not to run statistics to avoid mistaken inferences. 2. Your discussion would benefit if you provide further information on the bacterial transmission/colonization processes of these birds. We added text to describe the living condition and co-habitation of the birds. Due to the movement of the birds and openness of the facility, this opportunity is as close as you can get to a controlled experiment for raising birds together with the same exact diet in a laboratory. Due to the openness of the facility, no birds were considered isolated and therefore, have different exposure to microbes than other birds. Specific comments. (L=line) 1. L43: Provide a broader conclusion This was also mentioned by the other reviewer. We now state our two conclusions in the paper, which are 1) We characterized gut microbiota from healthy captive California condors to facilitate the design of future studies 2) demonstrate novel methods that can be applied to other systems 2. L49: Should be decade This word has been corrected. 3. Figure 1: Why are there random bacterial names on top of the bacterial groups? They make the figure harder to read. The bacterial names are explained in the figure caption: most of the phyla are dominated by a couple species, and these species were listed to aid in interpretation. 4. L180-181: This belongs in the figure legend. This description has been moved to the figure legend. 5. L199: Willing to make? I don’t understand. The wording of this sentence has been edited to remove the personification. 6. Figure 3: Title is not necessary We deleted this figure as suggested. 7. Figure 3: Why are all the rarefaction curves cut at the same number of sampled sequences? This was because this was all done to roughly the same depth of sequencing, and statistical significance drops at some point. 8. Figure 3: Could you provide the variance in the curves? Yes, we added confidence intervals to this figure and they added some additional support in the figure to show how the three groups did indeed differ. Adding confidence intervals was one of the reasons we left the figure in the primary section. 9. L258: delete the first are The entire sentence has been removed. 10. L296-300: Can you provide hypotheses as to why immature condors have a more diverse microbiome? This is definitely one of the most interesting results of the study since the few previous studies of the development of the microbiome in birds have reported the opposite. Without additional studies, our thinking and hypothesis is that it takes some time for biological communities to compete with each other and settle down in this species. This may be due to the more acidic environment of the condors compared to other species. We added some text in the discussion as it is an important point to bring up for the scientific community to potentially investigate in the future. 11. L360-362: Could you provide a reference here? As suggested, references have been provided. Submitted filename: Response to Reviewers.docx Click here for additional data file. 14 Nov 2019 California Condor Microbiomes: Bacterial Variety and Functional Properties in Captive-Bred Individuals PONE-D-19-18040R1 Dear Dr. Fair, We are pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it complies with all outstanding technical requirements. Within one week, you will receive an e-mail containing information on the amendments required prior to publication. When all required modifications have been addressed, you will receive a formal acceptance letter and your manuscript will proceed to our production department and be scheduled for publication. Shortly after the formal acceptance letter is sent, an invoice for payment will follow. To ensure an efficient production and billing process, please log into Editorial Manager at https://www.editorialmanager.com/pone/, click the "Update My Information" link at the top of the page, and update your user information. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org. If your institution or institutions have a press office, please notify them about your upcoming paper to enable them to help maximize its impact. If they will be preparing press materials for this manuscript, you must inform our press team as soon as possible and no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org. With kind regards, Suzanne L. Ishaq, PhD Academic Editor PLOS ONE Additional Editor Comments (optional): Reviewers' comments: Reviewer's Responses to Questions Comments to the Author 1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation. Reviewer #1: All comments have been addressed Reviewer #2: All comments have been addressed ********** 2. Is the manuscript technically sound, and do the data support the conclusions? The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented. Reviewer #1: Yes Reviewer #2: Yes ********** 3. Has the statistical analysis been performed appropriately and rigorously? Reviewer #1: Yes Reviewer #2: Yes ********** 4. Have the authors made all data underlying the findings in their manuscript fully available? The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified. Reviewer #1: Yes Reviewer #2: Yes ********** 5. Is the manuscript presented in an intelligible fashion and written in standard English? PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here. Reviewer #1: Yes Reviewer #2: Yes ********** 6. Review Comments to the Author Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters) Reviewer #1: Overall, I am happy with how the authors have addressed my comments. I have reviewed this manuscript for a different journal before reviewing it for PLoS, and the authors have done a great job at addressing all comments. Please congratulate them from me on their effort and a great manuscript. Reviewer #2: (No Response) ********** 7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files. If you choose “no”, your identity will remain anonymous but your review may still be made public. Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy. Reviewer #1: Yes: Kirsten Grond Reviewer #2: No 19 Nov 2019 PONE-D-19-18040R1 California Condor Microbiomes: Bacterial Variety and Functional Properties in Captive-Bred Individuals Dear Dr. Fair: I am pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department. If your institution or institutions have a press office, please notify them about your upcoming paper at this point, to enable them to help maximize its impact. If they will be preparing press materials for this manuscript, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org. For any other questions or concerns, please email plosone@plos.org. Thank you for submitting your work to PLOS ONE. With kind regards, PLOS ONE Editorial Office Staff on behalf of Dr. Suzanne L. Ishaq Academic Editor PLOS ONE
  47 in total

1.  Phospho-beta-glucosidase from Fusobacterium mortiferum: purification, cloning, and inactivation by 6-phosphoglucono-delta-lactone.

Authors:  J Thompson; S A Robrish; C L Bouma; D I Freedberg; J E Folk
Journal:  J Bacteriol       Date:  1997-03       Impact factor: 3.490

2.  Developmental microbial ecology of the crop of the folivorous hoatzin.

Authors:  Filipa Godoy-Vitorino; Katherine C Goldfarb; Eoin L Brodie; Maria A Garcia-Amado; Fabian Michelangeli; Maria G Domínguez-Bello
Journal:  ISME J       Date:  2010-02-04       Impact factor: 10.302

3.  Gut microbiome of the critically endangered New Zealand parrot, the kakapo (Strigops habroptilus).

Authors:  David W Waite; Peter Deines; Michael W Taylor
Journal:  PLoS One       Date:  2012-04-18       Impact factor: 3.240

Review 4.  Exploring the avian gut microbiota: current trends and future directions.

Authors:  David W Waite; Michael W Taylor
Journal:  Front Microbiol       Date:  2015-07-03       Impact factor: 5.640

5.  Wild griffon vultures (Gyps fulvus) as a source of Salmonella and Campylobacter in Eastern Spain.

Authors:  Clara Marin; Maria-Dolores Palomeque; Francisco Marco-Jiménez; Santiago Vega
Journal:  PLoS One       Date:  2014-04-07       Impact factor: 3.240

6.  Seroepidemiologic Survey of Potential Pathogens in Obligate and Facultative Scavenging Avian Species in California.

Authors:  Mary H Straub; Terra R Kelly; Bruce A Rideout; Curtis Eng; Janna Wynne; Josephine Braun; Christine K Johnson
Journal:  PLoS One       Date:  2015-11-25       Impact factor: 3.240

Review 7.  The Microbiome of Animals: Implications for Conservation Biology.

Authors:  Simon Bahrndorff; Tibebu Alemu; Temesgen Alemneh; Jeppe Lund Nielsen
Journal:  Int J Genomics       Date:  2016-04-18       Impact factor: 2.326

8.  Comparative Gut Microbiota of 59 Neotropical Bird Species.

Authors:  Sarah M Hird; César Sánchez; Bryan C Carstens; Robb T Brumfield
Journal:  Front Microbiol       Date:  2015-12-21       Impact factor: 5.640

9.  Development of the Chick Microbiome: How Early Exposure Influences Future Microbial Diversity.

Authors:  Anne L Ballou; Rizwana A Ali; Mary A Mendoza; J C Ellis; Hosni M Hassan; W J Croom; Matthew D Koci
Journal:  Front Vet Sci       Date:  2016-01-20

10.  Dynamics of Gut Microbiota Diversity During the Early Development of an Avian Host: Evidence From a Cross-Foster Experiment.

Authors:  Aimeric Teyssier; Luc Lens; Erik Matthysen; Joël White
Journal:  Front Microbiol       Date:  2018-07-09       Impact factor: 5.640
View more
  4 in total

Review 1.  Clinical challenges treating Stenotrophomonas maltophilia infections: an update.

Authors:  Maria F Mojica; Romney Humphries; John J Lipuma; Amy J Mathers; Gauri G Rao; Samuel A Shelburne; Derrick E Fouts; David Van Duin; Robert A Bonomo
Journal:  JAC Antimicrob Resist       Date:  2022-05-05

2.  Correction: California condor microbiomes: Bacterial variety and functional properties in captive-bred individuals.

Authors:  Lindsey Jacobs; Benjamin H McMahon; Joel Berendzen; Jonathan Longmire; Cheryl Gleasner; Nicolas W Hengartner; Momchilo Vuyisich; Judith D Cohn; Marti Jenkins; Andrew W Bartlow; Jeanne M Fair
Journal:  PLoS One       Date:  2020-03-16       Impact factor: 3.240

3.  Exposure to Anthropogenic Areas May Influence Colonization by Zoonotic Microorganisms in Scavenging Birds.

Authors:  Guillermo María Wiemeyer; Pablo Ignacio Plaza; Carla Paola Bustos; Alejandra Jimena Muñoz; Sergio Agustín Lambertucci
Journal:  Int J Environ Res Public Health       Date:  2021-05-14       Impact factor: 3.390

4.  Advances in the Microbiology of Stenotrophomonas maltophilia.

Authors:  Joanna S Brooke
Journal:  Clin Microbiol Rev       Date:  2021-05-26       Impact factor: 50.129
  4 in total

北京卡尤迪生物科技股份有限公司 © 2022-2023.