Our understanding of microbial evolution and diversification has been enriched by experimental studies of bacterial isolates in the laboratory, but it remains a challenge to study evolution in the context of more complex communities (Lenski, 2017). Ongoing advances in culture-independent technologies have allowed us to study bacteria in the complex and dense communities in which they naturally occur (Garud and Pollard, 2020). Within a community, individual players engage in many negative and positive ecological interactions. Negative interactions can originate from competition for resources and biomolecular warfare (Hibbing et al., 2010; Mitri and Foster, 2013), while positive interactions can stem from secreted metabolites that are used by other members of the community (cross-feeding) (Venturelli et al., 2018). These ecological interactions can create new niches and selective pressures, leading to eco-evolutionary feedbacks whose nature are yet to be fully understood.

Ecological interactions can yield positive or negative effects on the diversification of a focal species. Under the ‘diversity begets diversity’ (DBD) hypothesis, higher levels of community diversity increase the rate of speciation (or diversification, more generally) due to positive feedback mechanisms such as niche construction (Calcagno et al., 2017; Schluter and Pennell, 2017). Competition for limited niche space could also drive DBD if species diversify into new niches to avoid competition (Meyer and Kassen, 2007; Mitri and Foster, 2013; Schluter, 2000). By contrast, the ‘ecological controls’ (EC) hypothesis posits that competition for a limited number of niches at high levels of community diversity results in a negative effect on further diversification. Metabolic models predict that DBD may initially spur diversification due to cross-feeding, but the diversification rate eventually slows and reaches a plateau as metabolic niches are filled (San Roman and Wagner, 2021). These theoretical predictions are largely supported by our previous study involving 16S rRNA gene amplicon sequencing data from the Earth Microbiome Project, in which we observed a generally positive relationship (which we call the diversity slope; Figure 1) between community diversity and focal-taxon diversity at most taxonomic levels, reaching a plateau at the highest levels of diversity (Madi et al., 2020).

Figure 1

Download asset Open asset

In this previous study, we found stronger support for DBD in the animal gut relative to more diverse microbiomes such as soils and sediments, which were closer to a plateau of diversity (Madi et al., 2020). While diversity slopes were generally positive at taxonomic levels as fine as amplicon sequence variants (akin to species or strains) within a genus, they were most positive at higher levels such as classes or phyla. A recent experiment on soil bacteria also found evidence of DBD at the family level, likely driven by niche construction and metabolic cross-feeding (Estrela et al., 2022). It therefore remains unclear if the predictions of DBD hold primarily at these higher taxonomic levels, involving the ecological process of community assembly, or if they also apply at the finer intra-species level. Within-host intra-species diversity can arise by co-colonization of a host by genetically distinct strains belonging to the same species or evolutionary diversification of a lineage via de novo mutation and gene gain/loss events within a host.

Such fine-scale strain-level variation has important functional and ecological consequences; among other things, strains are known to engage in interactions that cannot be predicted from their species identity alone (Goyal et al., 2022). Although closely-related bacteria are expected to have broadly similar niche preferences, finer-scale niches may differ below the species level (Martiny et al., 2015). For example, the acquisition of a carbohydrate-active enzyme by Bacteroides plebeius allows it to exploit a new dietary niche in the guts of people consuming nori (seaweed) (Hehemann et al., 2010), and single nucleotide adaptations permit Enterococcus gallinarum translocation across the intestinal barrier resulting in inflammation (Yang et al., 2022). Despite their potential phenotypic effects, it is unknown if such fine-scale genetic changes are favored by higher community diversity (due, for example, to niche construction, as predicted by DBD) or suppressed (due to competition for limited niche space, as predicted by EC). Competition could also lead to DBD if focal species evolve new niche preferences to avoid extinction (Mitri and Foster, 2013; Schluter, 2000) – an idea with some support in experimental microcosms (Meyer and Kassen, 2007) but largely unexplored in natural communities.

Here, we investigate the relationship between intra-species genetic diversity and community diversity in the human gut microbiome, a well-studied system in which we previously found support for DBD at higher taxonomic levels. We use static and temporal shotgun metagenomic data from a large panel of healthy adult hosts from the Human Microbiome Project (HMP) (Lloyd-Price et al., 2017; Human Microbiome Project Consortium, 2012) as well as from four healthy individuals sampled almost daily over the course of 1 year (Poyet et al., 2019). Using metagenomic data allows us to track change in single nucleotide variation, strain diversity, and gene gain or loss events within relatively abundant species in the microbiome, and to study how these measures of intra-species diversity are associated with community diversity. Although such analyses of natural diversity cannot fully control for unmeasured confounding environmental factors, they are an important complement to controlled experimental and theoretical studies which lack real-world complexity.

We investigated the relationship between community diversity and within-species genetic diversity in human gut microbiota using two shotgun metagenomic datasets. First, we analyzed data from a panel of 249 healthy hosts (Lloyd-Price et al., 2017; Human Microbiome Project Consortium, 2012), in which stool samples were collected one to three times from each host at approximately 6-month intervals. Second, we analyzed data from four individuals sampled more densely over the course of ~18 months (Poyet et al., 2019). In both cases, we only consider intra-species diversity of relatively abundant species that are well sampled in these metagenomic datasets (Methods).

We examined several metrics of community diversity and intra-species diversity and calculated the slope of their relationship, defined as the diversity slope (Figure 1). We note that intra-species diversity can arise within hosts via de novo point mutation, gene gain or loss, or the coexistence of genetically distinct strains that diverged before colonizing the host. To quantify community diversity, we calculated Shannon diversity and richness at the species level. Shannon diversity is relatively insensitive to sampling effort (Madi et al., 2020; Walters and Martiny, 2020) but richness can be underestimated in low sample sizes. We therefore computed richness on data rarefied to an equal number of reads per sample, yielding generally similar results to unrarefied data (described below). In all cases, we included the number of reads per sample (coverage) as a covariate in our models, as this could affect estimates of both community diversity and intra-species diversity. To quantify intra-species diversity, we used a reference genome-based approach to call single nucleotide variants (SNVs) and gene copy number variants (CNVs) within each focal species and computed polymorphism rates, measured as the fraction of synonymous nucleotide sites in a species’ core genome with intermediate allele frequencies (between 0.2 and 0.8) within a host (Methods). We also repeated the analysis on nonsynonymous sites, as these are subject to stronger selective constraints. As an additional metric of intra-species diversity, we inferred the number of strains within each species using StrainFinder applied to all polymorphic sites (including those outside the 0.2–0.8 frequency range) (Smillie et al., 2018).

Community diversity is positively associated with intra-species polymorphism in the human gut microbiome

As an exploratory visualization, we began by plotting the relationship between community diversity and intra-species polymorphism rate calculated at synonymous sites in cross-sectional HMP metagenomes for the nine most prevalent species (Figure 2A and B). The slope of this relationship (the diversity slope; Figure 1) provides an indicator of the evidence for DBD (positive slope) or EC (flat or negative slope). The relationship between polymorphism rate and community diversity was mostly positive in the top nine most prevalent species in HMP hosts (Figure 2A and B). These nine species are used as a simple illustration of the diversity slope, not as a formal hypothesis-testing framework.

Figure 2 with 2 supplements see all

Download asset Open asset

To generalize across species and to formally test the predictions of DBD, we fit generalized additive models (GAMs) to the HMP data. Using GAMs, we are able to model non-linear relationships and account for random variation in the strength of the diversity slope across bacterial species, the uneven number of samples per host, and the non-independence of samples from the same host (Methods; see Supplementary file 1a and Supplementary file 2 section 1 for additional model details). These GAMs included 69 focal species with sufficient coverage to quantify within-species polymorphism (Methods); the results therefore apply to relatively abundant species in the human gut microbiome. GAMs showed an overall positive association between within-species polymorphism and Shannon diversity (Figure 2C, GAM, p=0.031, Chi-square test) as well as between within-species polymorphism and community richness after controlling for coverage as a covariate (Figure 2D, GAM, p=0.017, Chi-square test) or rarefying samples to an equal number of reads (Figure 2E, GAM, p=2.63e-04, Chi-square test). The random effect of species identity is highly significant in all models, indicating that each bacterial species has its own characteristic diversity slope (Supplementary file 1a). It appears that synonymous polymorphism reaches a plateau at high levels of community richness, which is particularly evident when using rarefied data (Figure 2E). Using the same GAMs applied to nonsynonymous polymorphism, we found no significant associations between diversity and within-species polymorphism rate (GAM, p>0.05, Chi-square test) (Supplementary file 1b, Supplementary file 2 section 4). This could be due to lower statistical power, since there are fewer nonsynonymous than synonymous sites, or it could reflect a true difference in the diversity slope between these site categories.

These generally positive correlations between focal species polymorphism and species-level measures of community diversity also hold when community diversity is measured at higher taxonomic levels; specifically, synonymous polymorphism rate was significantly positively associated with Shannon diversity calculated at the genus and family levels (GAMs, p<0.05, Chi-square test) (Figure 2—figure supplement 1, Supplementary file 1c). However, synonymous polymorphism rate was not significantly associated with Shannon diversity calculated at the highest taxonomic levels (order, class, and phylum, GAMs, p>0.05, Chi-square test). The positive correlation between polymorphism rate and richness held at all taxonomic levels (GAMs, p<0.05, Chi-square test) (Figure 2—figure supplement 1, Supplementary file 1c, Supplementary file 2 sections 2 and 3). When estimated at nonsynonymous sites, polymorphism rate was not significantly correlated with Shannon diversity at any taxonomic level (GAMs, p>0.05, Chi-square test), but was positively correlated with richness at the highest levels (phyla, class, and order, p=3e-04, p=0.017, and p=6.11e-04, respectively, Chi-square test from GAMs) (Figure 2—figure supplement 2, Supplementary file 1d, Supplementary file 2 sections 5 and 6). Even when not statistically significant, the diversity slopes were generally positive at all taxonomic levels for both synonymous and nonsynonymous polymorphism (Figure 2—figure supplements 1 and 2). Overall, these results are consistent with the predictions of DBD at most taxonomic levels. However, slightly different relationships are observed when considering different measures of community diversity (Shannon or richness) and different components of within-species diversity (nonsynonymous or synonymous).

Different measures of community diversity have contrasting associations with intra-species strain diversity

Within host polymorphism rates span several orders of magnitude (10^–5/bp to 10^–2/bp), largely due to the fact that strain content is variable across hosts. As previously argued (Garud et al., 2019), with conservatively high estimates for mutation rate (μ~10⁻⁹) (Sung et al., 2012), generation times (~10/day) (Poulsen et al., 1995), and time since colonization (<100 years), polymorphism rates of ~10^–2/bp or more are inconsistent with within-host diversification of a single colonizing lineage. Therefore, hosts with relatively high intra-host polymorphism rates are likely colonized by mixtures of multiple strains that diverged long before colonizing a host. Moreover, recent work suggests that the numbers and genetic composition of strains colonizing a host can vary from host to host (Garud et al., 2019; Olm et al., 2017; Russell and Cavanaugh, 2017; Truong et al., 2017; Verster et al., 2017). The associations between polymorphism and community diversity (Figure 2) are likely driven by a combination of de novo mutation and co-colonization by multiple strains.

To separate these two sources of diversity and to explicitly account for the strain structure within hosts, we inferred the number of strains per focal species with StrainFinder (Smillie et al., 2018) (Methods) and used strain number as another quantifier of intra-species diversity. The relationship between community diversity and strain number varied depending on the focal species and the measure of community diversity. For example, the inferred number of Bacteroides vulgatus strains increased with community diversity, while Bacteroides uniformis strain count decreased or remained flat (Figure 3A and B). Expanding upon these examples, we used generalized linear mixed models (GLMMs) to investigate the relationship between the number of strains per focal species and community diversity, while taking into account coverage per sample as a covariate and variation between species, hosts and samples as random effects (Methods). GLMMs are a special case of GAMs that can handle overdispersed, zero-truncated count data such as strain counts. The number of strains per focal species was positively correlated with community Shannon diversity (GLMM, p=3.58e-07, likelihood ratio test [LRT]) (Figure 3C, Supplementary file 1e, Supplementary file 2 section 7.1). This suggests that the positive correlation between polymorphism rate and Shannon diversity (Figure 2) is due at least in part to strain diversity.

Figure 3 with 1 supplement see all

Download asset Open asset

By contrast, species richness was negatively correlated with strain number (GLMM, p=1.67e-06, LRT) (Figure 3D, Supplementary file 1e, Supplementary file 2 section 7.2). The negative relationship with richness was unlikely to be confounded by sequencing depth, since the same result was obtained using rarefied data (Figure 3E, Supplementary file 1e, Supplementary file 2 section 7.3). The negative strain number-richness relationship also held at all other taxonomic ranks (GLMM, p<0.05, LRT), while the strain number-Shannon diversity relationship was generally positive (Figure 3—figure supplement 1, Supplementary file 1f, Supplementary file 2 sections 8 and 9). These effects also appear to be species-specific: for example, the number of B. vulgatus strains per host is positively correlated with both Shannon diversity and richness (consistent with DBD predictions) whereas Bacteroides ovatus has no relationship with Shannon diversity but a negative correlation with richness (consistent with EC; Figure 2A and B). Together, these results reveal that different components of community diversity can have contrasting effects on the diversity slope.

Community Shannon diversity is a predictor of intra-species polymorphism and gene loss in time series data

Our analyses thus far have considered only individual time points, which represent static snapshots of the dynamic processes of community assembly and evolution in the microbiome. To interrogate these phenomena over time, we analyzed 160 HMP subjects who were sampled two to three times ~6 months apart. Under a DBD model, we expect community diversity at an earlier time point to result in higher within-species polymorphism at a future time point. To test this expectation, we defined ‘polymorphism change’ as the difference between polymorphism rates at the two time points (Methods). We also investigated the effects of community diversity on gene loss and gain events within a focal species, as such changes in gene content are known to occur frequently within host gut microbiomes (Garud et al., 2019; Groussin et al., 2021; Yaffe and Relman, 2020; Zhao et al., 2019). Here, a gene was considered absent if its copy number (c) was <0.05 and present if 0.6 ≤c ≤ 1.2. As in the cross-sectional analyses above, we also controlled for sequencing depth of the sample and excluded genes with aberrant coverage or presence in multiple species (Methods).

In HMP samples, polymorphism change showed no significant relationships with community diversity at the earlier time point, whether it was estimated with Shannon index or species richness (GAM, p>0.05) (Supplementary file 2 section 10.1). These results suggest that DBD is negligible or undetectable over ~6 month time lags in the human gut. By contrast, we found that gene loss in a focal species between two consecutive time points was positively correlated with community diversity at the earlier time point (Figure 4; GLMM, p=0.028, p=0.034, and p=0.049, LRT for Shannon, richness and rarefied richness, respectively) (Supplementary file 1g, Supplementary file 2 section 10.3). Gene gains did not show any significant relationships with community diversity (GLMM, p>0.05). Selection for gene loss in more diverse communities is a prediction of the Black Queen hypothesis (BQH), provided that higher community diversity results in more redundant gene functions that compensate for losses in a focal species (Morris et al., 2012). Most species in HMP samples lost fewer than 10 genes over ~6 months – consistent with de novo deletion events of a few genes – but occasionally hundreds of genes were lost from a host, suggesting that strains with smaller genomes were selected in more diverse communities (Figure 4A and B).

Figure 4

Download asset Open asset

To study these dynamics at higher temporal resolution, we analyzed shotgun metagenomic data from four more frequently sampled healthy individuals from a previous study (Poyet et al., 2019). Stool from donor am was sequenced over 18 months with a median of 1 day between samples; an over 12 months (median 2 days between samples); ao over 5 months (median 1 day between samples); and ae over 7 months (median 2 days between samples). In this data, we tracked both polymorphism change and gene gains and losses between two successive time points in 15 species with a minimal marker gene coverage of 10 in at least 10 samples. These include seven species of Bacteroides, two Eubacterium, two Faecalibacterium, two Ruminococcus, as well as Alistipes putredinis and Parabacteroides merdae.

Using the Poyet dataset, we asked whether community diversity in the gut microbiome at one time point could predict polymorphism change at a future time point by fitting GAMs with the change in polymorphism rate as a function of the interaction between community diversity at the first time point and the number of days between the two time points. Shannon diversity at the earlier time point was correlated with increases in polymorphism (consistent with DBD) up to ~150 days (~4.5 months) into the future (Figure 5—figure supplement 1), but this relationship became weaker and then inverted (consistent with EC) at longer time lags (Figure 5A, Supplementary file 1h, GAM, p=0.023, Chi-square test). The diversity slope is approximately flat for time lags between 4 and 6 months, which could explain why no significant relationship was found in HMP, where samples were collected every ~6 months. No relationship was observed between community richness and changes in polymorphism (Supplementary file 1h, GAM, p>0.05).

Figure 5 with 1 supplement see all

Download asset Open asset

We next asked if community diversity at one time point could predict gene gains or losses at future time points by fitting GLMMs (analogous to the GAMs above, but more appropriate for gain/loss count data). Our method does not explicitly distinguish between gene gain/loss arising from recombination or deletion versus replacement of strains with different gene content. We found that community Shannon diversity predicted future gene loss in a focal species, and this effect became stronger with longer time lags (Figure 5B, Supplementary file 1i, GLMM, p=0.006, LRT for the effect of the interaction between the initial Shannon diversity and time lag on the number of genes lost). The model predicts that increasing Shannon diversity from its minimum to its maximum would result in the loss of 0.075 genes from a focal species after 250 days. In other words, about one of the 15 focal species considered would be expected to lose a gene in this time frame.

Higher Shannon diversity was also associated with fewer gene gains, and this relationship also became stronger over time (Figure 5C, Supplementary file 1i, GLMM, p=1.11e-09, LRT). We found a similar relationship between community species richness and gene gains, although the relationship was slightly positive at shorter time lags (Figure 5D, Supplementary file 1i, GLMM, p=3.41e-04, LRT). No significant relationship was observed between richness and gene loss (Supplementary file 1i, GLMM, p>0.05). Taken together with the HMP results (Figure 4), these longer time series reveal how the sign of the diversity slope can vary over time and how community diversity is generally predictive of reduced focal species gene content.

How eco-evolutionary feedbacks shape biological communities is an open question that, to date, has received substantial experimental and theoretical attention but is challenging to address in nature. In our previous study using 16S rRNA amplicon sequences from the Earth Microbiome Project, we found generally positive diversity slopes that eventually flattened at high levels of community diversity (Madi et al., 2020). This pattern is generally consistent with the predictions of DBD during the early stages of community assembly, but at later stages becomes more consistent with EC as niches become filled. Based on the time series metagenomic data analyzed here, the predictions of DBD also tend to hold over short time scales but fail over longer time scales of several months. Whether this leads to a terminal plateau of diversity, or whether ecological disturbances lead to cycles of DBD and EC, deserves further study.

In our previous study, the animal gut microbiome had one of the highest positive diversity slopes, making it an ideal candidate for investigating eco-evolutionary interactions at greater intra-species resolution using metagenomic data. In this follow-up study, we investigate the same phenomenon at a subspecies level, with results that are broadly consistent with the predictions of DBD giving way to EC over long time scales. We note that experiments supporting DBD have generally been conducted over short time scales ranging from 2 to 20 days (Estrela et al., 2022; Jousset et al., 2016), consistent with the importance of DBD early in community assembly. We also identify several nuances and caveats to this general conclusion, which are discussed below in detail.

Another recent study also found evidence for eco-evolutionary feedbacks in the HMP, in the form of a positive relationship between evolutionary modifications or strain replacements in a focal species and community diversity (Good and Rosenfeld, 2022). Using a model, they further showed that these eco-evolutionary dynamics could be explained by resource competition and did not require the cross-feeding interactions previously invoked (Estrela et al., 2022; San Roman and Wagner, 2021; San Roman and Wagner, 2018) to explain DBD at higher taxonomic levels. This could be because cross-feeding operates at the family or genus level and is less relevant at finer evolutionary scales.

There are several noteworthy caveats to our study. First, using metagenomic data from human microbiomes allowed us to study genetic diversity, but limited us to considering only relatively abundant species with genomes that were well covered by short sequence reads. Deeper or more targeted sequencing may permit us to determine whether the same patterns hold for rarer members of the microbiome. However, it is notable that the majority of the dozens of species across the two datasets analyzed support DBD, suggesting that the phenomenon may generalize.

Second, we cannot establish causal relationships without controlled experiments. We are therefore careful to conclude that positive diversity slopes are consistent with the predictions of DBD, and negative slopes with EC, but unmeasured environmental drivers could be at play. For example, increased dietary diversity could simultaneously select for higher community diversity and also higher intra-species diversity. In our previous study, we found that positive diversity slopes persisted even after controlling for potential abiotic drivers such as pH and temperature (Madi et al., 2020), but a similar analysis was not possible here due to a lack of metadata. Neutral processes can account for several ecological patterns such as species-area relationships (Hubbell, 2001), and must be rejected in favor of niche-centric models like DBD or EC. Using neutral models without DBD or EC, we found generally flat or negative diversity slopes due to sampling processes alone and that positive slopes were hard to explain with a neutral model (Madi et al., 2020). These models were intended mainly for 16S rRNA gene sequence data, but we expect the general conclusions to extend to metagenomic data. Nevertheless, further modeling and experimental work will be required to fully exclude a neutral explanation for the diversity slopes we report in the human gut microbiome.

Based on controlled experiments (Estrela et al., 2022) and modeling studies (San Roman and Wagner, 2021), DBD is a plausible causal explanation for positive diversity slopes in the gut microbiome. Although they also note that causality is difficult to establish, Good and Rosenfeld, 2022 suggest the importance of focal species evolution as a driver of changes in community structure, as shown in an experimental study of Pseudomonas in compost communities (Padfield et al., 2020). Clearly, further work is needed to establish the extent and relative rates of eco-evolutionary feedback in both directions. How these feedbacks among bacteria are influenced by abiotic factors and by interactions with fungi, archaea, and phages also deserves further study.

Third, the diversity slope changes depending on which component of within-species diversity or community diversity is considered. Notably, the number of strains within a focal species is positively correlated with Shannon diversity, but inversely correlated with species richness, suggesting that the ability of strains to colonize a host may be associated with higher community evenness rather than total species count. Higher evenness might maximize the chance of inter-species interactions, whereas higher richness might be driven by rare species that are less likely to interact. Although Shannon diversity is considered to be more robust and informative than richness in estimating bacterial diversity (He et al., 2013; Reese and Dunn, 2018), we observe the same contrasting results between Shannon diversity and richness when community diversity is calculated at higher taxonomic levels, suggesting that this pattern is not due to artifacts such as sequencing effort.

Our measures of intra-species diversity included both synonymous and nonsynonymous SNVs, inferred strain richness, and gene content. Synonymous nucleotide variation was consistently and positively associated with both community richness and Shannon diversity at all taxonomic levels (although not always with statistical significance). Nonsynonymous variation also tended to track positively with both measures of community diversity but was only statistically significantly associated with phylum and class richness. This suggests that evolutionarily older, less selectively constrained synonymous mutations and more recent nonsynonymous mutations that affect protein structure both track similarly with measures of community diversity. Nonetheless, a parsimonious explanation for possible differences between the two classes is that while they are affected similarly, we have more statistical power to identify correlations in the more numerous synonymous mutations. This merits further investigation.

Metagenomes from the same individual sampled over time allowed us to detect gene gain and loss events. In both HMP and Poyet et al. time series, community diversity was predictive of future gene loss in a focal species. This phenomenon is not explicitly predicted by either DBD or EC but it is compatible with aspects of the BQH, with some caveats. BQH predicts that a focal species will be less likely to encode genes with functions provided by other members of the surrounding community if such functions are ‘leaky’ and available as diffusible public goods (Morris et al., 2012). The BQH could also act as a driver of polymorphism within a species (Morris et al., 2014). Gene loss may be adaptive, provided that there is a cost to encoding and expressing the relevant genes (Albalat and Cañestro, 2016; Koskiniemi et al., 2012; Simonsen, 2022). The tendency for reductive genome evolution in bacteria is well established (Albalat and Cañestro, 2016; Koskiniemi et al., 2012; Puigbò et al., 2014). Genome reduction is a particular hallmark of endosymbiotic bacteria, which depend on their hosts for many metabolic gene products (McCutcheon and Moran, 2011; Nikoh et al., 2011). It has been shown that uncultivated bacteria from the gut have undergone considerable genome reduction, which may be an adaptive process that results from reliance on public goods (Nayfach et al., 2019). In the gut microbiome, the BQH has been invoked to explain the distribution of genes involved in vitamin B metabolism (Sharma et al., 2019) and iron acquisition (Vatanen et al., 2019).

Our findings in human gut metagenomes are compatible with the BQH under the assumption that increasing community diversity also increases the availability of leaky gene products – which may not be the case if genomes in the gut microbiome are functionally redundant, as inferred in a recent study (Tian et al., 2020). This study found that species in the gut microbiome were highly redundant at the level of annotated metabolic pathways (KEGG orthologs) and that more functionally redundant microbiomes were more resistant to colonization by fecal transplants. Relatively low redundancy microbiomes could therefore be more easily colonized but might also require migrants to encode more gene functions in order to persist. Importantly, functional redundancy may be high at the level of well-annotated metabolic functions, but low at the finer level of individual gene families, as demonstrated in marine microbiomes (Galand et al., 2018) but not yet studied explicitly in the gut. Here, we report that genome reduction in the gut is higher in more diverse gut communities. This could be due to de novo gene loss, preferential establishment of migrant strains encoding fewer genes, or a combination of the two. The mechanisms underlying this correlation remain unclear and could be due to biotic interactions – including metabolic cross-feeding as posited by some models (Estrela et al., 2022; San Roman and Wagner, 2021; San Roman and Wagner, 2018) but not others (Good and Rosenfeld, 2022) – or due to unknown abiotic drivers of both community diversity and gene loss. Finally, we measured community diversity from the phylum to the species level, not below. We therefore did not investigate how the BQH could extend to maintain gene content variation within a species, as has been shown experimentally in Escherichia coli (Morris et al., 2014). This could be an avenue for future work.

In our previous analysis of lower-resolution 16S rRNA amplicon sequences, we reported a tendency for focal genera with larger genomes to have higher diversity slopes, perhaps because they experience stronger DBD (Madi et al., 2020). At face value, this tendency seems at odds with the BQH, which predicts genome reduction in more diverse communities. This apparent contradiction may be reconciled by considering eco-evolutionary dynamics on different time scales. A recent study used phylogenetic and metabolic reconstructions to show that gene gains often drive metabolic dependencies among bacteria (Goyal, 2021), potentially explaining why genera with larger maximum genome size could experience stronger DBD. Our earlier study only had the genetic resolution to consider focal taxa down to the genus level, and by using the maximum genome size observed in a public database we did not capture the dynamic process of gene gain and loss within a species, as was possible in the current metagenomic study. It is therefore possible that on longer (ecological) time scales, larger genomes have more metabolic interactions and thus experience stronger DBD, while genome reduction in more diverse communities occurs on shorter (evolutionary) time scales.

In summary, we demonstrate how metagenomic data can be used to test the predictions of eco-evolutionary theory, including DBD, EC, and the BQH. It remains to be seen whether the distinct eco-evolutionary processes proposed by DBD and the BQH operate orthogonally or whether they interact. If BQH leads to gene losses that remain polymorphic rather than being lost entirely from the species (Morris et al., 2014) – or invasions of strains with fewer genes that remain incomplete and do not replace the resident strain – this could be viewed as a form of diversification and perhaps a special case of DBD. Here, we considered gene loss as a directional process; we did not attempt to distinguish between directional changes in gene copy number and the complete extinction of a gene, which is difficult to show using metagenomic data. Future work could attempt to resolve this point and to potentially combine DBD and BQH into a unified theory.

The raw sequencing reads for the metagenomic samples used in this study were downloaded from Human Microbiome Project Consortium, 2012 and Lloyd-Price et al., 2017 (URL: https://aws.amazon.com/datasets/human-microbiome-project/); and Poyet et al., 2019 (NCBI accession number PRJNA544527). All computer code for this paper is available at https://github.com/Naima16/DBD_in_gut_microbiome, (copy archived at swh:1:rev:eef9dc1643f279e4bb2f6d7e19c4fd7d5acc384f; Madi et al., 2023).

1. 1. Human Microbiome Project

   (2013) Registry of Open Data on AWS

   ID human-microbiome-project. Human Microbiome Project Data Set.

   https://aws.amazon.com/datasets/human-microbiome-project/
2. 1. Poyet M
   2. Groussin M
   3. Gibbons SM
   4. Avila-Pacheco J
   5. Jiang X
   6. Kearney SM
   7. Perrotta AR
   8. Berdy B
   9. Zhao S
   10. Lieberman TD
   11. Swanson PK
   12. Smith M
   13. Roesemann S
   14. Alexander JE
   15. Rich SA
   16. Livny J
   17. Vlamakis H
   18. Clish C
   19. Bullock K
   20. Deik A
   21. Scott J
   22. Pierce KA
   23. Xavier RJ
   24. Alm EJ

   (2019) NCBI BioProject

   ID PRJNA544527. BIO-ML: The Broad Institute-OpenBiome Microbiome Library enables mechanistic studies with isolates, genomes, and longitudinal metagenomics and metabolomic data.

   https://www.ncbi.nlm.nih.gov/bioproject/PRJNA544527

1. 1. Albalat R
   2. Cañestro C

   (2016) Evolution by gene loss

   Nature Reviews. Genetics 17:379–391.

   https://doi.org/10.1038/nrg.2016.39

   • PubMed
   • Google Scholar
2. Software

   1. Bolker B

   (2023) Mixedmodels-misc, version swh:1:rev:3b8b732667b6a261a3384f5cade8e78b68230f3a

   Software Heritage.

   https://archive.softwareheritage.org/swh:1:dir:8e83cdf33e3ea4f240774111de2c8757bd581a73;origin=https://github.com/bbolker/mixedmodels-misc;visit=swh:1:snp:168324f13d9af4e20cbb2bdcd87f1389e8c3831c;anchor=swh:1:rev:3b8b732667b6a261a3384f5cade8e78b68230f3a
3. Preprint

   1. Brooks ME
   2. Kristensen K
   3. van Benthem KJ
   4. Magnusson A
   5. Berg CW
   6. Nielsen A
   7. Skaug HJ
   8. Mächler M
   9. Bolker BM

   (2017) Modeling Zero-Inflated Count Data with GlmmTMB

   bioRxiv.

   https://doi.org/10.1101/132753

   • Google Scholar
4. 1. Calcagno V
   2. Jarne P
   3. Loreau M
   4. Mouquet N
   5. David P

   (2017) Diversity spurs diversification in ecological communities

   Nature Communications 8:15810.

   https://doi.org/10.1038/ncomms15810

   • PubMed
   • Google Scholar
5. 1. Chaumeil PA
   2. Mussig AJ
   3. Hugenholtz P
   4. Parks DH

   (2019) GTDB-tk: a toolkit to classify genomes with the genome taxonomy database

   Bioinformatics 36:1925–1927.

   https://doi.org/10.1093/bioinformatics/btz848

   • PubMed
   • Google Scholar
6. 1. Edgar RC

   (2010) Search and clustering orders of magnitude faster than blast

   Bioinformatics 26:2460–2461.

   https://doi.org/10.1093/bioinformatics/btq461

   • PubMed
   • Google Scholar
7. Preprint

   1. Estrela S
   2. Diaz-Colunga J
   3. Vila JCC
   4. Sanchez-Gorostiaga A
   5. Sanchez A

   (2022) Diversity Begets Diversity under Microbial Niche Construction

   bioRxiv.

   https://doi.org/10.1101/2022.02.13.480281

   • Google Scholar
8. 1. Galand PE
   2. Pereira O
   3. Hochart C
   4. Auguet JC
   5. Debroas D

   (2018) A strong link between marine microbial community composition and function challenges the idea of functional redundancy

   The ISME Journal 12:2470–2478.

   https://doi.org/10.1038/s41396-018-0158-1

   • PubMed
   • Google Scholar
9. 1. Garud NR
   2. Good BH
   3. Hallatschek O
   4. Pollard KS

   (2019) Evolutionary dynamics of bacteria in the gut microbiome within and across hosts

   PLOS Biology 17:e3000102.

   https://doi.org/10.1371/journal.pbio.3000102

   • PubMed
   • Google Scholar
10. 1. Garud NR
    2. Pollard KS

    (2020) Population genetics in the human microbiome

    Trends in Genetics 36:53–67.

    https://doi.org/10.1016/j.tig.2019.10.010

    • PubMed
    • Google Scholar
11. Preprint

    1. Good BH
    2. Rosenfeld LB

    (2022) Eco-Evolutionary Feedbacks in the Human Gut Microbiome

    bioRxiv.

    https://doi.org/10.1101/2022.01.26.477953

    • Google Scholar
12. 1. Goyal A

    (2021) Horizontal gene transfer drives the evolution of dependencies in bacteria

    SSRN Electronic Journal 1:3951496.

    https://doi.org/10.2139/ssrn.3951496

    • Google Scholar
13. 1. Goyal A
    2. Bittleston LS
    3. Leventhal GE
    4. Lu L
    5. Cordero OX

    (2022) Interactions between strains govern the eco-evolutionary dynamics of microbial communities

    eLife 11:e74987.

    https://doi.org/10.7554/eLife.74987

    • PubMed
    • Google Scholar
14. 1. Groussin M
    2. Poyet M
    3. Sistiaga A
    4. Kearney SM
    5. Moniz K
    6. Noel M
    7. Hooker J
    8. Gibbons SM
    9. Segurel L
    10. Froment A
    11. Mohamed RS
    12. Fezeu A
    13. Juimo VA
    14. Lafosse S
    15. Tabe FE
    16. Girard C
    17. Iqaluk D
    18. Nguyen LTT
    19. Shapiro BJ
    20. Lehtimäki J
    21. Ruokolainen L
    22. Kettunen PP
    23. Vatanen T
    24. Sigwazi S
    25. Mabulla A
    26. Domínguez-Rodrigo M
    27. Nartey YA
    28. Agyei-Nkansah A
    29. Duah A
    30. Awuku YA
    31. Valles KA
    32. Asibey SO
    33. Afihene MY
    34. Roberts LR
    35. Plymoth A
    36. Onyekwere CA
    37. Summons RE
    38. Xavier RJ
    39. Alm EJ

    (2021) Elevated rates of horizontal gene transfer in the industrialized human microbiome

    Cell 184:2053–2067.

    https://doi.org/10.1016/j.cell.2021.02.052

    • PubMed
    • Google Scholar
15. Book

    1. Hardin JW
    2. Hilbe JM

    (2018)

    Generalized Linear Models and Extensions

    Stata Press.

    • Google Scholar
16. 1. He Y
    2. Zhou BJ
    3. Deng GH
    4. Jiang XT
    5. Zhang H
    6. Zhou HW

    (2013) Comparison of microbial diversity determined with the same variable tag sequence extracted from two different PCR amplicons

    BMC Microbiology 13:208.

    https://doi.org/10.1186/1471-2180-13-208

    • PubMed
    • Google Scholar
17. 1. Hehemann JH
    2. Correc G
    3. Barbeyron T
    4. Helbert W
    5. Czjzek M
    6. Michel G

    (2010) Transfer of carbohydrate-active enzymes from marine bacteria to japanese gut microbiota

    Nature 464:908–912.

    https://doi.org/10.1038/nature08937

    • PubMed
    • Google Scholar
18. 1. Hibbing ME
    2. Fuqua C
    3. Parsek MR
    4. Peterson SB

    (2010) Bacterial competition: surviving and thriving in the microbial jungle

    Nature Reviews. Microbiology 8:15–25.

    https://doi.org/10.1038/nrmicro2259

    • PubMed
    • Google Scholar
19. Book

    1. Hubbell SP

    (2001)

    The Unified Neutral Theory of Biodiversity and Biogeography

    Princeton University Press.

    • Google Scholar
20. 1. Human Microbiome Project Consortium

    (2012) A framework for human microbiome research

    Nature 486:215–221.

    https://doi.org/10.1038/nature11209

    • PubMed
    • Google Scholar
21. 1. Jousset A
    2. Eisenhauer N
    3. Merker M
    4. Mouquet N
    5. Scheu S

    (2016) High functional diversity stimulates diversification in experimental microbial communities

    Science Advances 2:e1600124.

    https://doi.org/10.1126/sciadv.1600124

    • PubMed
    • Google Scholar
22. 1. Korem T
    2. Zeevi D
    3. Suez J
    4. Weinberger A
    5. Avnit-Sagi T
    6. Pompan-Lotan M
    7. Matot E
    8. Jona G
    9. Harmelin A
    10. Cohen N
    11. Sirota-Madi A
    12. Thaiss CA
    13. Pevsner-Fischer M
    14. Sorek R
    15. Xavier RJ
    16. Elinav E
    17. Segal E

    (2015) Growth dynamics of gut microbiota in health and disease inferred from single metagenomic samples

    Science 349:1101–1106.

    https://doi.org/10.1126/science.aac4812

    • PubMed
    • Google Scholar
23. 1. Koskiniemi S
    2. Sun S
    3. Berg OG
    4. Andersson DI

    (2012) Selection-driven gene loss in bacteria

    PLOS Genetics 8:e1002787.

    https://doi.org/10.1371/journal.pgen.1002787

    • PubMed
    • Google Scholar
24. 1. Langmead B
    2. Salzberg SL

    (2012) Fast gapped-read alignment with bowtie 2

    Nature Methods 9:357–359.

    https://doi.org/10.1038/nmeth.1923

    • PubMed
    • Google Scholar
25. 1. Lenski RE

    (2017) Experimental evolution and the dynamics of adaptation and genome evolution in microbial populations

    The ISME Journal 11:2181–2194.

    https://doi.org/10.1038/ismej.2017.69

    • PubMed
    • Google Scholar
26. 1. Lloyd-Price J
    2. Mahurkar A
    3. Rahnavard G
    4. Crabtree J
    5. Orvis J
    6. Hall AB
    7. Brady A
    8. Creasy HH
    9. McCracken C
    10. Giglio MG
    11. McDonald D
    12. Franzosa EA
    13. Knight R
    14. White O
    15. Huttenhower C

    (2017) Strains, functions and dynamics in the expanded human microbiome project

    Nature 550:61–66.

    https://doi.org/10.1038/nature23889

    • PubMed
    • Google Scholar
27. 1. Madi N
    2. Vos M
    3. Murall CL
    4. Legendre P
    5. Shapiro BJ

    (2020) Does diversity beget diversity in microbiomes?

    eLife 9:e58999.

    https://doi.org/10.7554/eLife.58999

    • PubMed
    • Google Scholar
28. Software

    1. Madi N
    2. Chen D
    3. Wolff R
    4. Shapiro BJ
    5. Garud N

    (2023) DBD_in_gut_microbiome, version swh:1:rev:eef9dc1643f279e4bb2f6d7e19c4fd7d5acc384f

    Software Heritage.

    https://archive.softwareheritage.org/swh:1:dir:f2dcf8b9282fd2dc051f93679fbef1d81ed120b9;origin=https://github.com/Naima16/DBD_in_gut_microbiome;visit=swh:1:snp:d74119eef913a5a8e0c62e0c5440670fed2a2593;anchor=swh:1:rev:eef9dc1643f279e4bb2f6d7e19c4fd7d5acc384f
29. 1. Martiny JBH
    2. Jones SE
    3. Lennon JT
    4. Martiny AC

    (2015) Microbiomes in light of traits: a phylogenetic perspective

    Science 350:aac9323.

    https://doi.org/10.1126/science.aac9323

    • PubMed
    • Google Scholar
30. 1. McCutcheon JP
    2. Moran NA

    (2011) Extreme genome reduction in symbiotic bacteria

    Nature Reviews. Microbiology 10:13–26.

    https://doi.org/10.1038/nrmicro2670

    • PubMed
    • Google Scholar
31. 1. Meyer JR
    2. Kassen R

    (2007) The effects of competition and predation on diversification in a model adaptive radiation

    Nature 446:432–435.

    https://doi.org/10.1038/nature05599

    • PubMed
    • Google Scholar
32. 1. Mitri S
    2. Foster KR

    (2013) The genotypic view of social interactions in microbial communities

    Annual Review of Genetics 47:247–273.

    https://doi.org/10.1146/annurev-genet-111212-133307

    • PubMed
    • Google Scholar
33. 1. Morris JJ
    2. Lenski RE
    3. Zinser ER

    (2012) The black queen hypothesis: evolution of dependencies through adaptive gene loss

    MBio 3:e00036-12.

    https://doi.org/10.1128/mBio.00036-12

    • PubMed
    • Google Scholar
34. 1. Morris JJ
    2. Papoulis SE
    3. Lenski RE

    (2014) Coexistence of evolving bacteria stabilized by a shared black queen function

    Evolution; International Journal of Organic Evolution 68:2960–2971.

    https://doi.org/10.1111/evo.12485

    • PubMed
    • Google Scholar
35. 1. Nayfach S
    2. Rodriguez-Mueller B
    3. Garud N
    4. Pollard KS

    (2016) An integrated metagenomics pipeline for strain profiling reveals novel patterns of bacterial transmission and biogeography

    Genome Research 26:1612–1625.

    https://doi.org/10.1101/gr.201863.115

    • PubMed
    • Google Scholar
36. 1. Nayfach S
    2. Shi ZJ
    3. Seshadri R
    4. Pollard KS
    5. Kyrpides NC

    (2019) New insights from uncultivated genomes of the global human gut microbiome

    Nature 568:505–510.

    https://doi.org/10.1038/s41586-019-1058-x

    • PubMed
    • Google Scholar
37. 1. Nikoh N
    2. Hosokawa T
    3. Oshima K
    4. Hattori M
    5. Fukatsu T

    (2011) Reductive evolution of bacterial genome in insect gut environment

    Genome Biology and Evolution 3:702–714.

    https://doi.org/10.1093/gbe/evr064

    • PubMed
    • Google Scholar
38. 1. Olm MR
    2. Brown CT
    3. Brooks B
    4. Firek B
    5. Baker R
    6. Burstein D
    7. Soenjoyo K
    8. Thomas BC
    9. Morowitz M
    10. Banfield JF

    (2017) Identical bacterial populations colonize premature infant gut, skin, and oral microbiomes and exhibit different in situ growth rates

    Genome Research 27:601–612.

    https://doi.org/10.1101/gr.213256.116

    • PubMed
    • Google Scholar
39. 1. Padfield D
    2. Vujakovic A
    3. Paterson S
    4. Griffiths R
    5. Buckling A
    6. Hesse E

    (2020) Evolution of diversity explains the impact of pre-adaptation of a focal species on the structure of a natural microbial community

    The ISME Journal 14:2877–2889.

    https://doi.org/10.1038/s41396-020-00755-3

    • PubMed
    • Google Scholar
40. 1. Poulsen LK
    2. Licht TR
    3. Rang C
    4. Krogfelt KA
    5. Molin S

    (1995) Physiological state of Escherichia coli BJ4 growing in the large intestines of streptomycin-treated mice

    Journal of Bacteriology 177:5840–5845.

    https://doi.org/10.1128/jb.177.20.5840-5845.1995

    • PubMed
    • Google Scholar
41. 1. Poyet M
    2. Groussin M
    3. Gibbons SM
    4. Avila-Pacheco J
    5. Jiang X
    6. Kearney SM
    7. Perrotta AR
    8. Berdy B
    9. Zhao S
    10. Lieberman TD
    11. Swanson PK
    12. Smith M
    13. Roesemann S
    14. Alexander JE
    15. Rich SA
    16. Livny J
    17. Vlamakis H
    18. Clish C
    19. Bullock K
    20. Deik A
    21. Scott J
    22. Pierce KA
    23. Xavier RJ
    24. Alm EJ

    (2019) A library of human gut bacterial isolates paired with longitudinal multiomics data enables mechanistic microbiome research

    Nature Medicine 25:1442–1452.

    https://doi.org/10.1038/s41591-019-0559-3

    • PubMed
    • Google Scholar
42. 1. Puigbò P
    2. Lobkovsky AE
    3. Kristensen DM
    4. Wolf YI
    5. Koonin EV

    (2014) Genomes in turmoil: quantification of genome dynamics in prokaryote supergenomes

    BMC Biology 12:66.

    https://doi.org/10.1186/s12915-014-0066-4

    • PubMed
    • Google Scholar
43. 1. Reese AT
    2. Dunn RR

    (2018) Drivers of microbiome biodiversity: a review of general rules, feces, and ignorance

    MBio 9:e01294-18.

    https://doi.org/10.1128/mBio.01294-18

    • PubMed
    • Google Scholar
44. 1. Russell SL
    2. Cavanaugh CM

    (2017) Intrahost genetic diversity of bacterial symbionts exhibits evidence of mixed infections and recombinant haplotypes

    Molecular Biology and Evolution 34:2747–2761.

    https://doi.org/10.1093/molbev/msx188

    • PubMed
    • Google Scholar
45. 1. San Roman M
    2. Wagner A

    (2018) An enormous potential for niche construction through bacterial cross-feeding in a homogeneous environment

    PLOS Computational Biology 14:e1006340.

    https://doi.org/10.1371/journal.pcbi.1006340

    • PubMed
    • Google Scholar
46. 1. San Roman M
    2. Wagner A

    (2021) Diversity begets diversity during community assembly until ecological limits impose a diversity ceiling

    Molecular Ecology 30:5874–5887.

    https://doi.org/10.1111/mec.16161

    • PubMed
    • Google Scholar
47. Book

    1. Schluter D

    (2000)

    The Ecology of Adaptive Radiation

    Oxford Series in Ecology and Evolution.

    • Google Scholar
48. 1. Schluter D
    2. Pennell MW

    (2017) Speciation gradients and the distribution of biodiversity

    Nature 546:48–55.

    https://doi.org/10.1038/nature22897

    • PubMed
    • Google Scholar
49. 1. Sharma V
    2. Rodionov DA
    3. Leyn SA
    4. Tran D
    5. Iablokov SN
    6. Ding H
    7. Peterson DA
    8. Osterman AL
    9. Peterson SN

    (2019) B-Vitamin sharing promotes stability of gut microbial communities

    Frontiers in Microbiology 10:1485.

    https://doi.org/10.3389/fmicb.2019.01485

    • PubMed
    • Google Scholar
50. 1. Shonkwiler JS

    (2016) Variance of the truncated negative binomial distribution

    Journal of Econometrics 195:209–210.

    https://doi.org/10.1016/j.jeconom.2016.09.002

    • Google Scholar
51. 1. Simonsen AK

    (2022) Environmental stress leads to genome streamlining in a widely distributed species of soil bacteria

    The ISME Journal 16:423–434.

    https://doi.org/10.1038/s41396-021-01082-x

    • PubMed
    • Google Scholar
52. 1. Smillie CS
    2. Sauk J
    3. Gevers D
    4. Friedman J
    5. Sung J
    6. Youngster I
    7. Hohmann EL
    8. Staley C
    9. Khoruts A
    10. Sadowsky MJ
    11. Allegretti JR
    12. Smith MB
    13. Xavier RJ
    14. Alm EJ

    (2018) Strain tracking reveals the determinants of bacterial engraftment in the human gut following fecal microbiota transplantation

    Cell Host & Microbe 23:229–240.

    https://doi.org/10.1016/j.chom.2018.01.003

    • PubMed
    • Google Scholar
53. 1. Sung W
    2. Ackerman MS
    3. Miller SF
    4. Doak TG
    5. Lynch M

    (2012) Drift-barrier hypothesis and mutation-rate evolution

    PNAS 109:18488–18492.

    https://doi.org/10.1073/pnas.1216223109

    • PubMed
    • Google Scholar
54. 1. Tian L
    2. Wang XW
    3. Wu AK
    4. Fan Y
    5. Friedman J
    6. Dahlin A
    7. Waldor MK
    8. Weinstock GM
    9. Weiss ST
    10. Liu YY

    (2020) Deciphering functional redundancy in the human microbiome

    Nature Communications 11:6217.

    https://doi.org/10.1038/s41467-020-19940-1

    • PubMed
    • Google Scholar
55. 1. Truong DT
    2. Tett A
    3. Pasolli E
    4. Huttenhower C
    5. Segata N

    (2017) Microbial strain-level population structure and genetic diversity from metagenomes

    Genome Research 27:626–638.

    https://doi.org/10.1101/gr.216242.116

    • PubMed
    • Google Scholar
56. 1. Vatanen T
    2. Plichta DR
    3. Somani J
    4. Münch PC
    5. Arthur TD
    6. Hall AB
    7. Rudolf S
    8. Oakeley EJ
    9. Ke X
    10. Young RA
    11. Haiser HJ
    12. Kolde R
    13. Yassour M
    14. Luopajärvi K
    15. Siljander H
    16. Virtanen SM
    17. Ilonen J
    18. Uibo R
    19. Tillmann V
    20. Mokurov S
    21. Dorshakova N
    22. Porter JA
    23. McHardy AC
    24. Lähdesmäki H
    25. Vlamakis H
    26. Huttenhower C
    27. Knip M
    28. Xavier RJ

    (2019) Genomic variation and strain-specific functional adaptation in the human gut microbiome during early life

    Nature Microbiology 4:470–479.

    https://doi.org/10.1038/s41564-018-0321-5

    • PubMed
    • Google Scholar
57. 1. Venturelli OS
    2. Carr AC
    3. Fisher G
    4. Hsu RH
    5. Lau R
    6. Bowen BP
    7. Hromada S
    8. Northen T
    9. Arkin AP

    (2018) Deciphering microbial interactions in synthetic human gut microbiome communities

    Molecular Systems Biology 14:e8157.

    https://doi.org/10.15252/msb.20178157

    • PubMed
    • Google Scholar
58. 1. Verster AJ
    2. Ross BD
    3. Radey MC
    4. Bao Y
    5. Goodman AL
    6. Mougous JD
    7. Borenstein E

    (2017) The landscape of type VI secretion across human gut microbiomes reveals its role in community composition

    Cell Host & Microbe 22:411–419.

    https://doi.org/10.1016/j.chom.2017.08.010

    • PubMed
    • Google Scholar
59. 1. Walters KE
    2. Martiny JBH

    (2020) Alpha-, beta-, and gamma-diversity of bacteria varies across habitats

    PLOS ONE 15:e0233872.

    https://doi.org/10.1371/journal.pone.0233872

    • PubMed
    • Google Scholar
60. 1. Wexler AG
    2. Goodman AL

    (2017) An insider’s perspective: bacteroides as a window into the microbiome

    Nature Microbiology 2:17026.

    https://doi.org/10.1038/nmicrobiol.2017.26

    • PubMed
    • Google Scholar
61. Preprint

    1. Wolff R
    2. Shoemaker WR
    3. Garud NR

    (2021) Ecological Stability Emerges at the Level of Strains in the Human Gut Microbiome

    bioRxiv.

    https://doi.org/10.1101/2021.09.30.462616

    • Google Scholar
62. 1. Wu D
    2. Jospin G
    3. Eisen JA

    (2013) Systematic identification of gene families for use as “markers” for phylogenetic and phylogeny-driven ecological studies of bacteria and archaea and their major subgroups

    PLOS ONE 8:e77033.

    https://doi.org/10.1371/journal.pone.0077033

    • PubMed
    • Google Scholar
63. 1. Yaffe E
    2. Relman DA

    (2020) Tracking microbial evolution in the human gut using hi-C reveals extensive horizontal gene transfer, persistence and adaptation

    Nature Microbiology 5:343–353.

    https://doi.org/10.1038/s41564-019-0625-0

    • PubMed
    • Google Scholar
64. 1. Yang Y
    2. Nguyen M
    3. Khetrapal V
    4. Sonnert ND
    5. Martin AL
    6. Chen H
    7. Kriegel MA
    8. Palm NW

    (2022) Within-Host evolution of a gut pathobiont facilitates liver translocation

    Nature 607:563–570.

    https://doi.org/10.1038/s41586-022-04949-x

    • PubMed
    • Google Scholar
65. 1. Yassour M
    2. Jason E
    3. Hogstrom LJ
    4. Arthur TD
    5. Tripathi S
    6. Siljander H
    7. Selvenius J
    8. Oikarinen S
    9. Hyöty H
    10. Virtanen SM
    11. Ilonen J
    12. Ferretti P
    13. Pasolli E
    14. Tett A
    15. Asnicar F
    16. Segata N
    17. Vlamakis H
    18. Lander ES
    19. Huttenhower C
    20. Knip M
    21. Xavier RJ

    (2018) Strain-level analysis of mother-to-child bacterial transmission during the first few months of life

    Cell Host & Microbe 24:146–154.

    https://doi.org/10.1016/j.chom.2018.06.007

    • PubMed
    • Google Scholar
66. 1. Zhao S
    2. Lieberman TD
    3. Poyet M
    4. Kauffman KM
    5. Gibbons SM
    6. Groussin M
    7. Xavier RJ
    8. Alm EJ

    (2019) Adaptive evolution within gut microbiomes of healthy people

    Cell Host & Microbe 25:656–667.

    https://doi.org/10.1016/j.chom.2019.03.007

    • PubMed
    • Google Scholar
67. 1. Zheng W
    2. Zhao S
    3. Yin Y
    4. Zhang H
    5. Needham DM
    6. Evans ED
    7. Dai CL
    8. Lu PJ
    9. Alm EJ
    10. Weitz DA

    (2022) High-throughput, single-microbe genomics with strain resolution, applied to a human gut microbiome

    Science 376:eabm1483.

    https://doi.org/10.1126/science.abm1483

    • PubMed
    • Google Scholar

Decision letter after peer review:

Thank you for submitting your article "Community diversity is associated with intra-species genetic diversity and gene loss in the human gut microbiome" for consideration by eLife. Your article has been reviewed by 3 peer reviewers, and the evaluation has been overseen by a Reviewing Editor and Detlef Weigel as the Senior Editor. The following individual involved in review of your submission has agreed to reveal their identity: Djordje Bajić (Reviewer #1).

The reviewers have discussed their reviews with one another, and the Reviewing Editor has drafted this to help you prepare a revised submission.

Essential revisions:

Three main weaknesses of the work were pointed out that would require extensive work to address, which would be beyond the scope of the current paper. The reviewers are requesting instead that these weaknesses be made very clear in the manuscript. The first is that DBD is only one possible process that can explain the observed patterns, and alternative hypotheses are not sufficiently outlined. The second is that the sequencing depth of the analyzed samples is little for gut microbiomes and hence allows only to look at a fraction of the strain-level diversity present. Third, it is not clear what the mechanism would link the BQH and DBD. Based on these points, it would be important to

(i) tone down the conclusions,

(ii) elaborate on certain concepts,

(iii) discuss these limitations of your study, and

(iv) consider whether any further analyses of your data might address these points (more detail below).

Reviewer #2 (Recommendations for the authors):

The study is purely correlative, some of the correlations are weak or not significant, only a subset of the analysed species shows a significant positive correlation, and when changing the method to assess the diversity not the same species are being identified to have a positive correlation (diversity-SNV analysis) or opposite trends are being observed (diversity-strain number). This means that the detected patterns are not very robust. The authors refer to a recent pre-print (Estrela et al) that provides experimental support for DBD. However, the environmental conditions in that study were very different from the one found in the gut, in that only a single carbon source was provided. DBD is only one possible process that may explain the observed patterns (see above). Therefore, I recommend the authors tune down their conclusions, discuss their findings more critically, i.e. offer alternative explanations, or argue better why these alternative explanations are less likely than DBD to explain the observed patterns.

From the discussion, it is not clear what would explain the DBD at finer taxonomic scale. It would be great to have a bit more insights about how the authors believe that diversity at e.g. the species, genus, or even family level could influence strain-level diversity within an individual species. What type of niches can be created by the presence of additional families? And how can these niches be occupied by distinct strains of the same species? Although it is true that some strains may be ecologically distinct, I would still assume that most strains occupy similar niches and are competing for similar nutrients. Moreover, wouldn't we expect that there is so much functional redundancy in the gut microbiome already that a slight increase in taxonomic diversity would not necessarily create new metabolic niches from which closely related strains with similar metabolic capacities would profit? I think it would help to explain a bit more how the DBD should work out at the strain level and to what extent there is evidence that the increased strain-level diversity can really be adaptation to new niches created by other microbes in the system.

The HMP1-1 and Poyet samples were rarified to 20 and 5 mio reads. This seems very little sequencing depth considering the high amount of genus/species level diversity in the human gut microbiome and the fact that the authors want to look at strain-level diversity. I have my doubts that this amount of data will allow to quantitatively assess diversity at the strain level in these samples, which may in part explain the not very robust correlations observed. How much of the total diversity is assessed with this number of reads? Rarefaction curves of SNVs or genes per species discovered when sub-setting the datasets across samples would be helpful. In addition to the low overall sequencing depth, only polymorphisms with a frequency of 0.2-0.8 were considered, which further limits the number of strains that can be detected. What was the idea behind applying this cut-off?

Why was only one species analysed across the entire Poyet dataset? If this is a general pattern it should be observable for more species than just B. vulgatus. While I acknowledge that read coverage may not be high enough for all samples for other species, there should be enough consecutive timepoints with sufficient coverage for some species. Unless there is a good reason why such analysis can only be done for B. vulgatus, I strongly recommend to extend this analysis to other community members to find further support for DBD-like patterns.

Along the same lines, the Poyet dataset would offer an opportunity to follow how diversity of B. vulgatus changes over more timepoints than just two in response to the species/genus/family diversity (e.g. over the entire 18 months?). This could provide interesting new insights that may help to understand how diversity changes over time at different levels. Why not look at changes in diversity across all timepoints? Would we expect that DBD continues over a longer period of time or would we see some type of negative feedback, because of ecological control kicking in at one point leading to oscillations of diversity?

The figures with the correlations should be improved. Specifically, Figure 2, 3, and 5 include too many data points of different species on top of each other. It is impossible to look at the distribution of the data of individual species and appreciate the existence of correlations. In panel A of Figure 2, I see only one data point for Dialister invisus. Why? Is the color legend missing next to panel B in Figure 3?

For the StrainFinder analysis, it was assumed that species for which no site passed the 20x threshold are presented by a single strain. This seems wrong in my opinion. Such data should be excluded as the shallow sequencing simply does not allow assessing the number of strains of that species in these samples.

Out of 68 species only 15 or 18 show a significant slope. This means that there are more species that display either EC or no correlation (as EC may not always detectable). This takes away from the conclusion that just DBD explains the patterns of diversity found in the gut microbiome, and should be acknowledged.

Line 148: What is the overlap of the 15 and 18 detected positive correlations between Shannon and richness analysis? From comparing the legends of Figure 2A and B, it seems that the overlap is not great. It would be helpful if the authors could state the overlap in their paper and discuss it.

Line 147: Significance of the correlations should be corrected for multiple testing. Would the identified correlations still be significant?

How does the method to identify gene loss/gene gain differentiate between gene loss and strain-level dynamics? That is, we do not know whether a certain genome lost genes or a strain lacking those genes happened to dominate the community. If such strain happened to migrate into the community for example, this would not be evidence for the BQH.

Reviewer #3 (Recommendations for the authors):

The paper is very well written and generally well-integrated with previous work by the group and by others.

I found the main novelty of the present work to be the usage of time series data to enquire about how present microbiome community diversity may influence within species polymorphism at a future time point, motivated by the mechanism underlying the Black Queen Hypothesis (BQH) put forward ten years ago.

There are however several points that I think need to be revised:

1. It was not clear to me in the authors' introduction and discussion if and how the DBD hypothesis is integrated with the BQH.

Do the authors consider these two hypotheses to be independent? What sort of mechanisms do the authors envisage to drive a positive association between community diversity and polymorphism? And is the mechanism underlying the BQH assumed to result in the fixation of a gene loss within the focal species or may it also result in polymorphism as in Morris, Papoulis and Lenski 2014 Coexistence of evolving bacteria stabilized by a shared black queen function?

I think clarifying these points would strengthen the manuscript.

2. Another issue that was not clear to me was why the authors compute the polymorphism rates with only synonymous sites. If there is a reason to exclude non-synonymous sites it should be mentioned in the manuscript. In addition, the abstract and the conclusions should precisely state that the significant correlations are with polymorphism rates at synonymous sites.

3. The manuscript emphasizes the finding of a positive correlation in several species but does not emphasize that for the majority of species no correlation was found. Thus, the conclusion that DBD prevails (pg. 7 line 150 and abstract) looks a bit exaggerated.

Essential revisions:

Three main weaknesses of the work were pointed out that would require extensive work to address, which would be beyond the scope of the current paper. The reviewers are requesting instead that these weaknesses be made very clear in the manuscript. The first is that DBD is only one possible process that can explain the observed patterns, and alternative hypotheses are not sufficiently outlined. The second is that the sequencing depth of the analyzed samples is little for gut microbiomes and hence allows only to look at a fraction of the strain-level diversity present. Third, it is not clear what the mechanism would link the BQH and DBD. Based on these points, it would be important to

(i) tone down the conclusions,

(ii) elaborate on certain concepts,

(iii) discuss these limitations of your study, and

(iv) consider whether any further analyses of your data might address these points (more detail below).

We thank the reviewers and editors for their thorough and thoughtful assessment of our paper and we have worked to address these essential revisions. First, we have extensively re-written this manuscript to be more balanced in evaluating alternative hypotheses to DBD, including ecological controls (EC) and abiotic drivers of diversity. Second, we clarify that our results apply only to relatively abundant species with sufficient sequencing depth to study sub-species diversity. We explicitly state that further work will be required to extend these results to rarer species (please see Discussion). Third, we acknowledge it is not clear which mechanism would link DBD and BQH – or if these models operate orthogonally. This is also an avenue for future work, which we mention in the Discussion.

As suggested, we:

i) Tone down our conclusions and mention alternative explanations for the observed patterns;

ii) Elaborate on certain concepts, including why sub-species genetic diversity might alterniches and thus be impacted by community diversity, and also how DBD or EC could arise by de novo mutation and/or strain migration across hosts;

iii) Include several paragraph of caveats and limitations of our study (please see Discussion); and

iv) We comprehensively analyze 15 species from four hosts instead of only Bacteroides vulgatus from a single host from Poyet et al. to reinforce the correlation between community diversity and gene loss over time. We also add new analyses to determine whether the DBD relationship differs between non-synonymous and synonymous nucleotide variants. Finally, we closely evaluated the pathway-level analysis and found that the gene-level analyses were more robust. We have therefore removed the pathway analyses.

Below we elaborate on these points and include passages from the revised text.

Reviewer #2 (Recommendations for the authors):

The study is purely correlative, some of the correlations are weak or not significant, only a subset of the analysed species shows a significant positive correlation, and when changing the method to assess the diversity not the same species are being identified to have a positive correlation (diversity-SNV analysis) or opposite trends are being observed (diversity-strain number). This means that the detected patterns are not very robust. The authors refer to a recent pre-print (Estrela et al) that provides experimental support for DBD. However, the environmental conditions in that study were very different from the one found in the gut, in that only a single carbon source was provided. DBD is only one possible process that may explain the observed patterns (see above). Therefore, I recommend the authors tune down their conclusions, discuss their findings more critically, i.e. offer alternative explanations, or argue better why these alternative explanations are less likely than DBD to explain the observed patterns.

While we still think that DBD provides a plausible mechanism to explain the observed patterns, we are now careful to acknowledge alternative mechanisms. We more prominently list possible caveats and limitations in the Discussion. We also give more weight in the Results to contrasting patterns that we observe in the data. For example:

“The negative strain number-richness relationship also held at all other taxonomic ranks (GLMM, P<0.05, LRT) (Figure S3, Table S6, Supplementary File 1 section 9), while the strain number-Shannon diversity relationship was generally positive (Figure S3, Supplementary File 1 section 8). These effects also appear to be species-specific: for example, the number of Bacteroides vulgatus strains per host is positively correlated with both Shannon diversity and richness (consistent with DBD predictions) whereas B. ovatus has no relationship with Shannon diversity but a negative correlation with richness (consistent with EC; Figure 2A, B). Together, these results reveal that different components of community diversity can have contrasting effects on the diversity slope.”

From the discussion, it is not clear what would explain the DBD at finer taxonomic scale. It would be great to have a bit more insights about how the authors believe that diversity at e.g. the species, genus, or even family level could influence strain-level diversity within an individual species. What type of niches can be created by the presence of additional families? And how can these niches be occupied by distinct strains of the same species? Although it is true that some strains may be ecologically distinct, I would still assume that most strains occupy similar niches and are competing for similar nutrients. Moreover, wouldn't we expect that there is so much functional redundancy in the gut microbiome already that a slight increase in taxonomic diversity would not necessarily create new metabolic niches from which closely related strains with similar metabolic capacities would profit? I think it would help to explain a bit more how the DBD should work out at the strain level and to what extent there is evidence that the increased strain-level diversity can really be adaptation to new niches created by other microbes in the system.

Thank you for raising this important point. We have now added the following Introduction paragraph to outline the reasons we might expect strain-level diversity to affect niche preferences, and thus be relevant to DBD:

“Such fine-scale strain-level variation has important functional and ecological consequences; among other things, strains are known to engage in interactions that cannot be predicted from their species identity alone (Goyal et al., 2022). Although closely-related bacteria are expected to have broadly similar niche preferences, finer-scale niches may differ below the species level (Martiny et al., 2015). For example, the acquisition of a carbohydrate-active enzyme by Bacteroides plebeius allows it to exploit a new dietary niche in the guts of people consuming nori (seaweed) (Hehemann et al., 2010), and single nucleotide adaptations permit Enterococcus gallinarum translocation across the intestinal barrier resulting in inflammation (Yang et al., 2022). Despite their potential phenotypic effects, it is unknown if such fine-scale genetic changes are favored by higher community diversity (due for example to niche construction, as predicted by DBD) or suppressed (due to competition for limited niche space, as predicted by EC). Competition could also lead to DBD if focal species evolve new niche preferences to avoid extinction (Mitri and Foster, 2013; Schluter, 2000) – an idea with some support in experimental microcosms (Meyer and Kassen, 2007) but largely unexplored in natural communities.”

Regarding functional redundancy, we have added the following Discussion paragraph:

“Our findings in human gut metagenomes are compatible with the BQH under the assumption that increasing community diversity also increases the availability of leaky gene products – which may not be the case if genomes in the gut microbiome are functionally redundant, as inferred in a recent study (Tian et al., 2020). This study found that species in the gut microbiome were highly redundant at the level of annotated metabolic pathways (KEGG orthologs) and that more functionally redundant microbiomes were more resistant to colonization by fecal transplants. Relatively low-redundancy microbiomes could therefore be more easily colonized, but might also require migrants to encode more gene functions in order to persist. Importantly, functional redundancy may be high at the level of well-annotated metabolic functions, but low at the finer level of individual gene families, as demonstrated in marine microbiomes (Galand et al., 2018) but not yet studied explicitly in the gut. Here we report that genome reduction in the gut is higher in more diverse gut communities. This could be due to de novo gene loss, preferential establishment of migrant strains encoding fewer genes, or a combination of the two. The mechanisms underlying this correlation remain unclear and could be due to biotic interactions – including metabolic cross-feeding as posited by some models (Estrela et al., 2022; San Roman and Wagner, 2021, 2018) but not others (Good and Rosenfeld, 2022) – or due to unknown abiotic drivers of both community diversity and gene loss.”

The HMP1-1 and Poyet samples were rarified to 20 and 5 mio reads. This seems very little sequencing depth considering the high amount of genus/species level diversity in the human gut microbiome and the fact that the authors want to look at strain-level diversity. I have my doubts that this amount of data will allow to quantitatively assess diversity at the strain level in these samples, which may in part explain the not very robust correlations observed. How much of the total diversity is assessed with this number of reads? Rarefaction curves of SNVs or genes per species discovered when sub-setting the datasets across samples would be helpful. In addition to the low overall sequencing depth, only polymorphisms with a frequency of 0.2-0.8 were considered, which further limits the number of strains that can be detected. What was the idea behind applying this cut-off?

We thank the reviewer for their concerns about potentially under-calling SNVs and genes when subsetting datasets. However, we only rarefied the data when ascertaining species richness, since richness is most affected by sampling effort. We realize that it may have been confusing as to when rarefaction was applied to the data and therefore have modified the Methods to be clearer about our approach as follows:

“SNV and gene content variation within a focal species were ascertained only from the full dataset and not the rarefied dataset.”

As mentioned above, we now clearly mention the caveat that our results apply only to relatively abundant focal species for which we are able to estimate sub-species diversity. For example, Strain Finder only estimates the number of strains within species with sufficient coverage. We also include sequence coverage (number of total reads in the metagenome) in all of our GLMM and GAM analyses, so the effects of sampling effort are taken into account throughout.

We thank the reviewer for asking about the 0.2-0.8 thresholds, as it provides us an opportunity to clarify our methodology. First we wish to emphasize that these thresholds of 0.2 – 0.8 were not used for strain detection. Rather, all SNVs were used for strain detection with StrainFinder, provided that the site had 20x or greater coverage. We clarify this in the main text as well as the methods.

The 0.2-0.8 threshold was applied when computing polymorphism rates, as in Garud, Good et al. 2019, where we analyzed the same HMP dataset. This measure of polymorphism rate is similar to a more traditional measure of heterozygosity H=E[2f(1−f)], in which intermediate frequency alleles contribute the most weight. Our approach is preferable, however, because it is more robust to low-frequency sequencing errors that can overwhelm the average in H. We clarify our citations to our previous paper in the Methods section, where our rationale is explained.

Why was only one species analysed across the entire Poyet dataset? If this is a general pattern it should be observable for more species than just B. vulgatus. While I acknowledge that read coverage may not be high enough for all samples for other species, there should be enough consecutive timepoints with sufficient coverage for some species. Unless there is a good reason why such analysis can only be done for B. vulgatus, I strongly recommend to extend this analysis to other community members to find further support for DBD-like patterns.

We now expand the analysis to 15 species and 4 individuals as we elaborate in our response above.

Along the same lines, the Poyet dataset would offer an opportunity to follow how diversity of B. vulgatus changes over more timepoints than just two in response to the species/genus/family diversity (e.g. over the entire 18 months?). This could provide interesting new insights that may help to understand how diversity changes over time at different levels. Why not look at changes in diversity across all timepoints? Would we expect that DBD continues over a longer period of time or would we see some type of negative feedback, because of ecological control kicking in at one point leading to oscillations of diversity?

Thank you for this very useful suggestion. As described above, the analysis of these time series now includes time lag in the model. We find that the positive correlation between community diversity and focal species polymorphism at a future time point holds on shorter time lags, then flattens and becomes negative at longer time lags (Figure 5 and S4). The relationship between community diversity and gene loss is always positive, with slight variation depending on the time lag. It is intriguing to think about possible oscillations of DBD and EC, and we hope this can be explored in future work. We now mention this in the first Discussion paragraph:

“Based on the time series metagenomic data analyzed here, the predictions of DBD also tend to hold over short time scales, but fail over longer time scales of several months. Whether this leads to a terminal plateau of diversity, or whether ecological disturbances lead to cycles of DBD and EC, deserves further study.”

The figures with the correlations should be improved. Specifically, Figure 2, 3, and 5 include too many data points of different species on top of each other. It is impossible to look at the distribution of the data of individual species and appreciate the existence of correlations. In panel A of Figure 2, I see only one data point for Dialister invisus. Why? Is the color legend missing next to panel B in Figure 3?

We agree that the original figures were difficult to read and we have remade them completely. The new figures 2, 3, and 4 now show each of the nine most prevalent focal species as a separate panel, to avoid overlapping points, and also show the GLMM or GAM model fits in lower panels to illustrate the broad patterns supported by all the data.

For the StrainFinder analysis, it was assumed that species for which no site passed the 20x threshold are presented by a single strain. This seems wrong in my opinion. Such data should be excluded as the shallow sequencing simply does not allow assessing the number of strains of that species in these samples.

We thank the reviewer for this suggestion, and we have incorporated it into our revised analysis. Now, for the purposes of strain finding, we consider absent from a given sample any species in which there are fewer than 100 polymorphic sites with greater than 20x coverage. 9% of species/sample pairs used in the original analysis were excluded after applying this cutoff.

100 sites was chosen as a conservative threshold for strain detection. Prior work has shown that when multiple strains of a species are present within a single host, fixed genetic differences between strains tend to greatly outnumber genetic differences between individuals belonging to the same strain (Yassour et al. 2016; Garud, Good et al. 2019; Zheng et al. 2022). In particular, while tens of thousands of sites or more typically segregate between strains, only hundreds to a few thousands of sites typically segregate within a strain. Additionally, assuming a conservative sequencing error rate of 10^-3 per base pair, thousands of sites may be spuriously marked as polymorphic due to technical error—such polymorphisms will, however, be found at low frequency, and are explicitly accounted for by StrainFinder. In total, genetic differences between strains should account for a large proportion of the total polymorphisms present within a sample when multiple strains coexist. Therefore, we expect that a significant majority of the (minimum) 100 polymorphic sites will be informative as to the presence of multiple strains.

Out of 68 species only 15 or 18 show a significant slope. This means that there are more species that display either EC or no correlation (as EC may not always detectable). This takes away from the conclusion that just DBD explains the patterns of diversity found in the gut microbiome, and should be acknowledged.

Our original intention in reporting these simple linear correlations for each species was to give illustrative examples, not to establish statistical significance. We have now removed these individual correlations for each species, which were not corrected for multiple tests, nor did they include important covariates like coverage. We now focus entirely on the more robust GLMM or GAMs, which include these covariates and provide better statistical descriptions of the entire dataset in question.

Line 148: What is the overlap of the 15 and 18 detected positive correlations between Shannon and richness analysis? From comparing the legends of Figure 2A and B, it seems that the overlap is not great. It would be helpful if the authors could state the overlap in their paper and discuss it.

As mentioned in the response above, these individual linear correlations have been removed.

Line 147: Significance of the correlations should be corrected for multiple testing. Would the identified correlations still be significant?

As mentioned above, these correlations were meant to be anecdotal and illustrative. Since they are not statistically robust, we have removed them in favor of the more appropriate GLMMs and GAMs. This obviates the need for multiple test corrections, since the entire dataset is used to test a single overarching hypothesis.

How does the method to identify gene loss/gene gain differentiate between gene loss and strain-level dynamics? That is, we do not know whether a certain genome lost genes or a strain lacking those genes happened to dominate the community. If such strain happened to migrate into the community for example, this would not be evidence for the BQH.

Indeed, our method does not distinguish between these possibilities, and we now clarify this as follows in the Results:

“Our method does not explicitly distinguish between gene gain/loss arising from recombination or deletion versus replacement of strains with different gene content.”

However, we disagree with the reviewer that gene loss due to strain dynamics rather than de novo gene loss would not be evidence for BQH. We think that selection for strains encoding fewer genes in more diverse communities (with the proviso that these more diverse communities encode more public goods) is also compatible with the general predictions of BQH. This is now mentioned in the Discussion as follows:

“Here we report that genome reduction in the gut is higher in more diverse gut communities. This could be due to de novo gene loss, preferential establishment of migrant strains encoding fewer genes, or a combination of the two.”

Reviewer #3 (Recommendations for the authors):

The paper is very well written and generally well-integrated with previous work by the group and by others.

I found the main novelty of the present work to be the usage of time series data to enquire about how present microbiome community diversity may influence within species polymorphism at a future time point, motivated by the mechanism underlying the Black Queen Hypothesis (BQH) put forward ten years ago.

Thank you for these kind words.

There are however several points that I think need to be revised:

1. It was not clear to me in the authors' introduction and discussion if and how the DBD hypothesis is integrated with the BQH.

Do the authors consider these two hypotheses to be independent? What sort of mechanisms do the authors envisage to drive a positive association between community diversity and polymorphism? And is the mechanism underlying the BQH assumed to result in the fixation of a gene loss within the focal species or may it also result in polymorphism as in Morris, Papoulis and Lenski 2014 Coexistence of evolving bacteria stabilized by a shared black queen function?

I think clarifying these points would strengthen the manuscript.

We thank the reviewer for their thoughtful comments. In the Discussion, we now elaborate further on how our results are consistent with DBD and BQH. However, we fully acknowledge that future work is needed to integrate DBD and BQH into a single unified model. We have now written a new final Discussion paragraph on this topic:

“In summary, we demonstrate how metagenomic data can be used to test the predictions of eco-evolutionary theory, including DBD, EC, and the BQH. It remains to be seen whether the distinct eco-evolutionary processes proposed by DBD and the BQH operate orthogonally or whether they interact. If BQH leads to gene losses that remain polymorphic rather than being lost entirely from the species (Morris et al., 2014) – or invasions of strains with fewer genes that remain incomplete and do not replace the resident strain – this could be viewed as a form of diversification and perhaps a special case of DBD. Here we considered gene loss as a directional process; we did not attempt to distinguish between directional changes in gene copy number and the complete extinction of a gene, which is difficult to show using metagenomic data. Future work could attempt to resolve this point and to potentially combine DBD and BQH into a unified theory.”

Additionally, we appreciate the reviewer mentioning the possibility that BQH could lead to polymorphism within a species. We acknowledge this could be true especially if multiple strains of a species colonize a host. In the Discussion we now mention this possibility needs to be investigated further:

“Finally, we measured community diversity from the phylum to the species level, not below. We therefore did not investigate how the BQH could extend to maintain gene content variation within a species, as has been shown experimentally in E. coli (Morris et al., 2014). This could be an avenue for future work.”

2. Another issue that was not clear to me was why the authors compute the polymorphism rates with only synonymous sites. If there is a reason to exclude non-synonymous sites it should be mentioned in the manuscript. In addition, the abstract and the conclusions should precisely state that the significant correlations are with polymorphism rates at synonymous sites.

We thank the reviewer for this comment. We have now added an analysis of nonsynonymous sites, which largely follow the same patterns observed for synonymous sites. This can be seen by comparing Figures S1 (synonymous) and S2 (non-synonymous), which both show generally positive associations between community diversity and focal species polymorphism. The trends are very similar for both synonymous and non-synonymous polymorphism, although generally less statistically significant for the non-synonymous sites. This is to be expected since non-synonymous sites are less numerous, providing a smaller sample size that could reduce statistical power. This is now reported in the Results, and in the Discussion as follows:

“Nonsynonymous variation also tended to track positively with both measures of community diversity but was only statistically significantly associated with phylum and class richness. This suggests that evolutionarily older, less selectively constrained synonymous mutations and more recent nonsynonymous mutations that affect protein structure both track similarly with measures of community diversity. Nonetheless, a parsimonious explanation for possible differences between the two classes is that while they are affected similarly, we have more statistical power to identify correlations in the more numerous synonymous mutations. This merits further investigation.”

3. The manuscript emphasizes the finding of a positive correlation in several species but does not emphasize that for the majority of species no correlation was found. Thus, the conclusion that DBD prevails (pg. 7 line 150 and abstract) looks a bit exaggerated.

The simple linear correlations for each species were meant to be illustrative, but were not properly representative of statistical significance. We have now removed these individual correlations for each species, which were not corrected for multiple tests, nor did they include important covariates like coverage. We now focus on the more robust GLMM or GAMs, which include these covariates and provide better statistical descriptions of the entire dataset in question. This now shows that a positive relationship between community diversity and focal species polymorphism is the predominant pattern, although of course this varies by time scale and by species. We make this clear in the Results in this passage, among others:

“These effects also appear to be species-specific: for example, the number of Bacteroides vulgatus strains per host is positively correlated with both Shannon diversity and richness (consistent with DBD predictions) whereas B. ovatus has no relationship with Shannon diversity but a negative correlation with richness (consistent with EC; Figure 2A, B). Together, these results reveal that different components of community diversity can have contrasting effects on the diversity slope.”

We have now toned down the conclusion that “DBD prevails” and have rewritten the Abstract to include the following text, which we think is more balanced:

“We find that both intra-species polymorphism and strain number are positively correlated with community Shannon diversity. This trend is consistent with DBD, although we cannot exclude abiotic drivers of diversity. Shannon diversity is also predictive of increases in polymorphism over time scales up to ~4-6 months, after which the diversity slope flattens and then becomes negative—consistent with DBD eventually giving way to EC. Also supporting a complex mixture of DBD and EC, the number of strains per focal species is positively associated with Shannon diversity but negatively associated with richness.”

Author details

1. Naïma Madi

   Département de sciences biologiques, Université de Montréal, Montréal, Canada

   Contribution

   Conceptualization, Software, Formal analysis, Investigation, Visualization, Methodology, Writing – original draft, Writing – review and editing

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0001-8793-7389

2. Daisy Chen

   1. Computational and Systems Biology, University of California, Los Angeles, Los Angeles, United States
   2. Bioinformatics and Systems Biology Program, University of California, San Diego, San Diego, United States

   Contribution

   Data curation, Formal analysis, Methodology, Writing – review and editing

   Contributed equally with

   Richard Wolff

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0001-6516-7029

3. Richard Wolff

   Department of Ecology and Evolutionary Biology, University of California, Los Angeles, Los Angeles, United States

   Contribution

   Formal analysis, Investigation, Methodology, Writing – review and editing

   Contributed equally with

   Daisy Chen

   Competing interests

   No competing interests declared

4. B Jesse Shapiro

   1. Département de sciences biologiques, Université de Montréal, Montréal, Canada
   2. McGill Genome Centre, McGill University, Montreal, Canada
   3. Quebec Centre for Biodiversity Science, Montreal, Canada
   4. McGill Centre for Microbiome Research, Montreal, Canada
   5. Department of Microbiology and Immunology, McGill University, Montreal, Canada

   Contribution

   Conceptualization, Supervision, Funding acquisition, Investigation, Visualization, Writing – original draft, Project administration, Writing – review and editing

   For correspondence

   jesse.shapiro@mcgill.ca

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0001-6819-8699

5. Nandita R Garud

   1. Department of Ecology and Evolutionary Biology, University of California, Los Angeles, Los Angeles, United States
   2. Department of Human Genetics, University of California, Los Angeles, Los Angeles, United States

   Contribution

   Conceptualization, Data curation, Supervision, Funding acquisition, Investigation, Methodology, Writing – original draft, Project administration, Writing – review and editing

   For correspondence

   ngarud@ucla.edu

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0003-4217-4407

• 1,350

  Page views

• 199

  Downloads

• 3

  Citations

Article citation count generated by polling the highest count across the following sources: Crossref, PubMed Central, Scopus.