Differences in the fitness effects of traded resources shape traits and persistence in multi-mutualist communities

Renuka Agarwal; Anne E Curé; Kari A Segraves; David M Althoff

doi:10.1371/journal.pone.0340707

. 2026 Feb 3;21(2):e0340707. doi: 10.1371/journal.pone.0340707

Differences in the fitness effects of traded resources shape traits and persistence in multi-mutualist communities

Renuka Agarwal ^1,^*, Anne E Curé ¹, Kari A Segraves ², David M Althoff ^1,^*

Editor: Karthik Raman³

PMCID: PMC12867262 PMID: 41632805

Abstract

Mutualistic interactions, where species reciprocally benefit from each other, are crucial for ecosystem stability and biodiversity. These interactions often involve species that experience different fitness effects for the traded resources or services. Because mutualisms rely on positive feedback between partners, such asymmetries can strongly influence evolutionary outcomes. Differences in fitness effects create divergent selective pressures, shaping trait evolution and determining the persistence of mutualisms. The strength of these effects can also vary depending on the availability of traded resources from other sources. Despite their importance, the evolutionary role of fitness asymmetries in mutualism has received little attention, beyond recognizing that some species may be more dependent on their partners than others. This study investigates how asymmetry in the fitness effects of a traded resource influences the persistence and phenotypic trait evolution of species in multi-mutualist guilds. To test this, we constructed synthetic multi-mutualist communities by combining, reproductively isolated and genetically modified strains of Saccharomyces cerevisiae that engage in a nutritional mutualism by trading adenine and lysine. One guild of four strains cannot produce lysine but overproduces adenine while the other guild cannot produce adenine but overproduces lysine. Lysine overproducers survive periods of low adenine better than adenine overproducers survive low lysine. Over a four-week evolution experiment we observed that strain persistence was strongly influenced by the availability of external resources. Communities in media containing traded resources supported the survival of all strains, whereas obligate conditions led to a significant extinction, especially for adenine overproducers. We observed distinct evolutionary trajectories of traits under obligate versus supplemented conditions. Phenotypic assays revealed that costs and benefits evolved differently depending on the essentiality of the traded resource and nutrient supplementation. These results demonstrate that asymmetries in the fitness effects of traded resources can influence evolutionary outcomes, species persistence, and community stability in multi-mutualist communities.

Introduction

Species engage in a range of interactions, from parasitism to competition to mutualism [1]. Among these, mutualism stands out as reciprocally beneficial interaction in which two or more partners gain fitness benefits [2]. Such relationships are essential for the stability and functioning of ecosystems, as they involve exchange of essential resources that contribute to the foundation of many communities [3–5]. Mutualisms can also expand to involve a network of interactions with multiple partner species [6–7]. From the well-known associations between plants and pollinators to complex symbioses between microbes and their hosts, mutualism facilitates key processes such as nutrient cycling, pollination, and defense against predators [2]. Beyond their ecological importance, mutualistic interactions provide valuable insights into evolutionary biology, particularly in understanding how reciprocity evolves and is maintained among species [8–10]. Mutualistic interactions have been studied by focusing on both individual pairs of interacting species as well as guilds or networks of potential mutualistic partner species [2]. Mutualisms are highly diverse, occurring across a wide range of environments and are influenced by both abiotic and biotic factors. For instance, the coexistence of phylogenetically diverse bacteria within a single coral [11], the co-occurrence of multiple endosymbionts in the insect guts [12], and the diversity of arbuscular mycorrhizal fungi in the roots of plants [13]. As most mutualistic interactions in nature are multi-partner, it is highly important to understand the complex interplay between ecological and evolutionary dynamics that govern these multi-species mutualisms.

The direct and indirect interaction feedbacks that can occur in multi-mutualist communities [14] are influenced by the availability of external sources of commodities essential for mutualistic interactions, as well as by changes in species richness driven by shifts in the abiotic or biotic environment [15]. The availability of alternative resources can change the intensity of selection pressures, thereby modifying the degree of dependency and number of partners within mutualistic communities. Both these factors can shape the evolution of mutualistic traits [16]. As alternative sources of traded resources or services become more available, partners can persist to some extent independently, and the benefits of interactions become more flexible [17]. In contrast, extreme dependency in mutualistic relationships can drive specialization, where the survival of one mutualist species is solely dependent on its partner. In such cases, stronger selection pressures on mutualistic traits are expected [18].

The intensity of selection on mutualists may be influenced by the relative contribution of traded resources to each partner’s fitness [17]. This concept is analogous to the life-dinner principle in predator-prey interactions, where the fitness cost to prey is much higher than to predators [19]. For example, in pollination mutualisms, if a pollinator does not visit a flower, it only misses a single nectar meal, whereas the plant loses the opportunity to fertilize ovules. Similarly, endosymbionts that fail to locate a host may perish, while long-lived host can survive without endosymbionts until colonized. Such asymmetries in the importance of mutualistic resources are expected to generate differential selective pressures on partners and may alter the rate of trait evolution. Asymmetries may also magnify competition for traded resources among members of mutualistic guild, adding further selective pressures as mutualists adapt both to their partners and to guild members competing for the same resources [20]. Testing these effects of asymmetry is a next important step in closing a major gap in our understanding of mutualism [16].

The complexity of multi-mutualist communities and fitness asymmetries among partner species raises the question of how do asymmetries in the fitness consequences of traded resources change mutualism dynamics? This question is central to understanding the long-term stability and evolutionary trajectories of mutualistic systems, particularly in the context of environmental change [21–23]. To address this question, we used a synthetic nutritional mutualistic system based on brewer’s yeast, Saccharomyces cerevisiae, in which we can modify the availability of the traded mutualistic resources, adenine and lysine. These nutritional resources differ in their relative importance for yeast survival and reproduction. Specifically, the importance of lysine to yeast cells is greater than for adenine [24–25], creating an inherent asymmetry in partner benefits. We assembled communities consisting of four reproductively isolated yeast strains on each side of the mutualism and maintained them for four weeks. We surveyed mutualistic strain persistence and key traits associated with production of traded resources and population growth.

Materials and methods

Synthetic model of multi-mutualist communities

We used genetically modified strains of S. cerevisiae [25] to study how the availability of traded resources and fitness asymmetries influences the persistence of mutualistic strains and the evolution of their phenotypic traits. The strains were derived from adenine- and lysine-overproducing parental strains WY833 and WY811, respectively [25]. To generate the lysine overproducing mutualists (LysOP), WY811 was crossed with SY9913 (Euroscarf 40007), and selected progeny were subsequently crossed to introduce pairwise combinations of ura3Δ, his3Δ, and leu2Δ auxotrophies used for selective tracking in co-culture. Similarly, the adenine overproducing mutualists (AdeOP) were engineered by crossing WY833 with a progenitor strain derived from WY833 × SY9913 crosses (genotype ade8Δ0 his3Δ1 leu2Δ0 ura3Δ0 lys21WT), followed by additional backcrossing to obtain the desired auxotrophic marker combinations. For a subset of strains, trp1Δ was introduced using PCR-based gene disruption with SY9914 (Euroscarf 40008) as a template to enable positive selection on hygromycin plates. All strains share a common laboratory genetic background and differ only at the engineered biosynthetic and auxotrophic loci, ensuring genetic comparability across treatments. These strains are genetically distinct, reproductively isolated, and functionally analogous to species. They engage in a cross-feeding nutritional mutualism, where each strain overproduces a resource necessary for the survival of its mutualist partner. The two types of overproducing guilds were adenine overproducers (AdeOP) that produce excess amounts of adenine but cannot produce their own lysine and lysine overproducers (LysOP) that produce excess amounts of lysine but cannot produce their own adenine. AdeOP strains release adenine immediately into the media, whereas the LysOP strains release a small amount of lysine while alive but release large amounts of lysine following cell death and lysis [24,26].

Adenine and lysine are essential nutrients for yeast cells, but they differ in their physiological effects when limited. Adenine is actively transported into yeast cells and incorporated into ribonucleotides [27]. A shortage of adenine prevents cell division within about two hours when ATP, ADP, and AMP have been mostly consumed [28]. During this period, yeast also start accumulating storage carbohydrates like trehalose and glycogen and mount a strong stress response that leads to increased survival [29]. Thus, adenine starved cells can survive for long periods of time and have carbohydrates stores that can be mobilized when conditions improve. Lysine, on the other hand, is also actively transported into yeast cells but is used as a nitrogen source for protein production [30]. Yeast cells that are starved for lysine begin programmed cell death and can start dying within 24 h [31]. Additionally, lysine starvation can lead to reduced resource use efficiency of glucose which limits growth [32] and causes increases in the presence of reactive oxygen species within the cell that leads to cell damage [31]. Because of this difference in the importance of adenine and lysine to cell functioning, the AdeOP strains have reduced starvation resistance relative to the LysOP strains [24–25] and will experience increased oxidative stress after lysine starvation. Together, these differences in resource release dynamics, starvation resistance, and physiological responses establish a strong asymmetry between the mutualistic partners, providing a powerful system to experimentally test how resource availability and fitness imbalances shape the persistence, and evolutionary trajectories of mutualistic interactions.

We used 16 genotypes of yeast strains in different combinations to create symmetrical multi-species communities of four AdeOP strains and four LysOP strains (S1 Table). The combination of genotypes allowed us to separate the strains via selective plating. As the AdeOPs and the LysOPs mutualists do not produce lysine and adenine, respectively, they form an obligate mutualism when cultured in media lacking these nutrients. We set up communities in two environments: one without external adenine and lysine, where persistence depended entirely on partner exchange and one supplemented with low levels of adenine (4 × 10 ⁻ ³ g/L) and lysine (18 × 10 ⁻ ³ g/L), allowing yeast to obtain nutrients both from the media and from partner exchange. With low nutrient supplementation, the mutualisms become less dependent, as strains can partially meet their nutritional needs for adenine and lysine from the environment.

Setting up the communities and evolving the strains

We used experimental evolution to explore the evolutionary changes of mutualists in response to each other. To initiate the evolution experiment, we cultured single colonies overnight in liquid YPD (10 g/L yeast extract, 20 g/L peptone, and 20 g/L dextrose). Yeast strains were then rinsed with sterile water and diluted to a standard density of 0.0125 OD_600. Mutualistic communities were assembled by combining four AdeOPs and four LysOPs into 2 ml cultures, with a starting density of 0.1 OD₆₀₀ (0.0125 OD₆₀₀ of each interacting strain) for each community (Fig 1). To establish a mutualism in the absence of external resources, we propagated mutualistic cultures in synthetic dextrose (SD) medium lacking adenine and lysine (1.5 g/L Difco yeast nitrogen base without amino acids or ammonium sulfate, 5 g/L ammonium sulfate, and 20 g/L dextrose, with additional amino acids). To establish a mutualism in the presence of external nutrients, the communities were grown in SD medium with 0.1X adenine and 0.2X lysine (1.5 g/L Difco yeast nitrogen base without amino acids or ammonium sulfate, 5 g/L ammonium sulfate, and 20 g/L dextrose, 4 x 10⁻³ g/L adenine and 18 x 10⁻⁴ g/L lysine and with supplemental amino acids), respectively, in a 48-well plate. We made 96 unique combinations of eight yeast strain communities. Each combination was replicated twice in media without external resources, resulting in a total of 192 independent cultures. For media containing low levels of adenine and lysine, there were 96 independent cultures. We chose this replication scheme because previous work [25] reported high rates of strain extinction under obligate conditions, and our goal was to maximize the number of communities that retained all eight strains. All cultures were incubated in their respective liquid medium on a rotating wheel in 48-well deep-well plates at 30°C. Cultures were grown for 48 hours in 2 ml of medium and typically reached a mean OD₆₀₀ of approximately 8 before transfer. At each transfer, 200 µl of the culture (10% of the total volume) was inoculated into 1.8 ml of fresh medium. This relatively high transfer volume ensured that the communities did not experience strong population bottlenecks and thus maintained a large effective population size throughout the experiment. As a result, selection was expected to act efficiently on phenotypic and genetic variation within the communities. The transfer regime did, however, did serve to reduce the number of cells per unit volume and did reduce the amount of traded resources right after a transfer event. At the end of each week, we checked for the presence of each strain by spot plating each community on eight selective plates that each only allowed growth for a single strain.

Fig 1 — **(a)** Schematic representation of the engineered strains used in this study and how they were combined to assemble different community types. Two metabolic overproduction backgrounds were used: an adenine-overproducing strain unable to synthesize lysine (AdeOP) and a lysine-overproducing strain unable to synthesize adenine (LysOP). Communities were constructed by pairing four AdeOP strains with four LysOP strains, generating multiple unique cross-feeding consortia **(b)** Experimental media conditions. Communities of *Saccharomyces cerevisiae* overproduction strains growing in media with and without external resources present. The communities were grown in two types of media: no external resources added (light yellow background) and with some external resources added (darker yellow background). Replicate communities growing in these media types were maintained for four weeks.

The experiment was run for four weeks. We froze the cultures that were used to initiate the experiment (designated as ‘ancestral’) and the cultures at the end of four weeks (designated as ‘evolved’). The cultures were frozen in 25% glycerol and stored at −80°C.

Assessing trait evolution

After four weeks of evolution, we assessed if any traits important to the mutualism had changed in the evolved strains compared to the ancestral strains. For evolved strains, we wanted to use strains which persisted in the most species rich communities. In the communities that were growing with external resources, we used strains which came from communities where all eight members persisted after four weeks. However, this was not possible for the communities that were evolving in media without external resources due to higher extinctions (Fig 3, S2 Table). In these cases, we had to use communities in which some strains went extinct.

Fig 3 — Each colored line represents a distinct genotype within the AdeOP or LysOP guild (strain IDs shown in the legend; full genotypes listed in S1 Table). Error bars indicate the standard error of the proportion of replicate communities in which the strain persisted (calculated as √(p(1 − p)/n), where p is the persistence proportion and n is the number of replicate communities). Communities with no external resources in the media lost strains more quickly and more strains over time. This effect was more pronounced for adenine overproducers.

To measure trait evolution, we compared the ancestral strains to their evolved counterparts. We first separated each evolved strain from the others in their community by striking them out on selective plates. We then grew single colony for each strain in YPD for 24 h. This culture was then spot plated and grown in media without external resources to ensure no other strains were present. Each evolved strain was then frozen as above and used as the starting culture for phenotyping assays. Each ancestral genotype was replicated eight times and the evolved genotypes from the two media were replicated 6–8 times.

Measuring and analyzing strain persistence

The goal behind the analysis of strain persistence dynamics across different media types is to understand how environmental conditions influence mutualist strain survival. We converted the persistence data into a binomial factor, with 1 indicating presence and 0 indicating extinction. The proportion of each strain surviving in all communities was calculated at each week for different media type. Since proportion data are bounded between 0 and 1, we applied an arcsine square root transformation to stabilize variances and meet the assumptions of parametric testing. We performed a two-way ANOVA in R (v. 2023.06.2 + 561) to test for significant differences in strain persistence across media types, type of overproducer strain (AdeOP & LysOP), and their interaction.

Starvation resistance measurement

We measured starvation resistance to assess the adaptive responses of yeast strains to amino acid deprivation. We compared differences in starvation resistance between AdeOP and LysOP strains by growing ancestral and evolved strains separately in the adenine and lysine deprived media. To initiate the assay, swabs containing five colonies of each ancestral or evolved strain were grown overnight in YPD. Cultures were then adjusted to a starting density of 0.1 OD₆₀₀ in media lacking adenine and lysine. We counted the number of live cells at 0 h, 48 h and 72 h for the AdeOPs and 0 h, 96 h and 120 h for the LysOPs (based on Vidal et al. [25]). Cells were counted after spreading an aliquot on a YPD plate that was maintained at 30°C. To assess differences in the proportion of surviving colonies among the different media types, we conducted a one-way analysis of variance (ANOVA) using R (v. 2023.06.2 + 561). The proportion of colonies was used as the response variable, and lineage category (“ancestral”, “with external resources”, and “without external resources”) was treated as a fixed factor. For AdeOPs, the change in the proportion of live colonies was examined after 48 h and 72 h whereas for LysOP strains after 96 h and 120 h.

Overproduction measurements

We chose to measure overproduction to quantify the cost of the mutualism. We grew individual cultures of each evolved and ancestral strain overnight in YPD by using a swab of five colonies from the selective plates. These cultures were then washed with sterile water and diluted to 0.1 OD₆₀₀ in a specific medium. Ancestral and evolved strains of AdeOPs were diluted in SD medium containing 1% lysine (90 mg/L) and LysOPs were grown in SD medium containing 1% adenine (40 mg/L). These cultures were grown individually for 24 h at 30°C on a rotating wheel. After 24 h, we used a small amount of culture to measure the optical density and 800 µl of the culture was centrifuged and then filtered to remove cell debris.

Due to differences in how adenine and lysine are produced, we needed to use two different methods to collect each nutrient. We collected the adenine filtrate by spinning down 800 µl of culture for 1 min at 3700 rpm and then filtering it through a AcroPrep Advance filter plate (Pall Corporation). Collecting lysine was more complicated because lysine is released mainly after the cell death [26]. To collect the lysine filtrate from the LysOP cultures, we took 800 µl of each culture and heat shocked at 100°C for 5 mins, briefly vortexed, and then placed the samples into a dry ice-ethanol bath for 2 mins. We then thawed the samples, and vortexed again for 2 mins. The culture was then spun down for 1 min at 3700 rpm and the sample was filtered using same approach as for AdeOPs. We did not heat shock the adenine overproducers so that we would have a more accurate estimate of the adenine excreted without the addition of adenine from disrupted cells. Thus, separate extraction protocols were used to accurately represent biologically available metabolite levels for each overproducer type.

To measure the amount of adenine or lysine produced by the overproducing ancestral and evolved strains, we used the filtrate to grow a test strain (SY9915, MATa ade2Δ0 ade8Δ0 his3Δ1 leu2Δ0 lys2Δ0 trp1Δ63 ura3Δ0) that cannot produce adenine and lysine. The test strain was prepared by growing overnight in YPD and then diluted in sterile water after washing. We measured adenine overproduction by adding 50 µl of filtrate from AdeOP cultures and 0.1 OD₆₀₀ culture of test strain into 1000 µl of 2X SD media with 1% lysine but lacking adenine. Similarly, to estimate lysine overproduction, we added 100 µl of filtrate from LysOP culture and 0.1 OD₆₀₀ culture of test strain to 1000 µl of 2X SD media with 1% adenine but lacking lysine. The volumes were standardized to 2000 µl by adding sterile water. We took the OD₆₀₀ for the test strain after 24 h and used the corrected OD (OD₆₀₀ corrected = OD₆₀₀ of test strain/ OD₆₀₀ of overproducer strain) for further analysis. We tested for differences in overproduction between media types with a linear mixed effect model using the lmer function [33] in R, with media type as a fixed effect and the strain ID as a random effect.

Resource use efficiency and growth rate measurements

We measured resource use efficiency to assess each strain’s ability to use traded resources, and thus quantify the benefit gained from mutualism. We define resource use efficiency (RUE) as how much biomass is generated from a limited amount of mutualistic resource. We did this by measuring strain growth when placed alone in medium with a small amount of adenine and lysine. Strains that evolved increased RUE, can turn the same amounts of adenine or lysine into more biomass as compared to the ancestral strain. To measure RUE, we grew individual strains in YPD followed by washing with sterile water and resuspending them in SD medium with 0.05% adenine and lysine. 0.1 OD₆₀₀ cultures of each strain were incubated for 24 h at 30°C. At 24 h, we measured the OD₆₀₀ of the culture. We then compared the yield of ancestral and evolved strains. While the cultures were growing for 24 h we also took OD₆₀₀ measurements at 4 and 6 h to estimate growth rate. We calculated growth rate, r, as the number of doublings during exponential growth between 4–6 h:

r = ln ({OD}_{600}^{t 1} / {OD}_{600}^{t 0}) / ln (2)

We tested for differences in RUE and growth between media types with a linear mixed effect model using the lmer function [33] in R, with media type as a fixed effect and the strain ID as a random effect.

Results

Media and mutualist type jointly determine strain persistence

With one exception, all the mutualistic communities persisted for four weeks. Strain richness, however, was strongly influenced by media conditions and mutualist type. In obligate conditions (no external resources added), 62% of the communities retained all four LysOP strains while only 8% of communities retained all four AdeOP strains (Fig 2). There was also one community that lost all AdeOP strains. When external resources were available in the media, persistence rates increased as all four AdeOP and LysOP strains survived by the end of four weeks (Fig 2). Furthermore, AdeOP strains started going extinct earlier in the obligate media. Extinction of strains began in two weeks for AdeOPs and in three weeks for LysOPs when in obligate media (Fig 3). A two-way ANOVA revealed a significant main effect of media type (F₁,₅₄ = 42.54, p < 0.001). In contrast, the main effect of overproducer type was not significant (F₁,₁₆ = 0.00, p = 1.00). Importantly, a significant interaction between media type and overproducer type was detected (F₁,₅₄ = 15.78, p < 0.001), indicating that the effect of media on persistence depends on the overproduction classification (mutualist type).

Fig 2 — Communities with no external mutualistic resources in the media lost more strains over time. AdeOP strains were lost more frequently than LysOP strains under obligate and when external nutrients were present in the media. Adding nutrients to the media increased individual strain persistence for both mutualist types.

Starvation resistance remained unchanged across evolutionary environments

Starvation resistance is an important trait that allows strains to survive when resources are limited. Adenine and lysine become highly limiting after each transfer event in media lacking external adenine or lysine but are less limiting in supplemented media. Overall, AdeOPs exhibited lower starvation resistance compared to LysOPs. After 48 h of starvation, only 58% (± 3 SE) of the ancestral AdeOPs survived lysine starvation (Fig 4). AdeOP mutualists that evolved in media without external resources, as well as those evolved in media supplemented with additional adenine and lysine, showed similar levels of starvation resistance at 48 h, with survival rates of 58% (± 7 SE) and 58% (± 4 SE), respectively (Fig 4). At 72 h, 26% (± 1.9 SE) of the ancestral AdeOPs survived. At this time point, mutualists that evolved in media without external resources, exhibited a survival rate of 30% (± 5.5 SE), and those in externally supplemented media showed a survival rate of 30% (± 3 SE). There was no significant effect of media type on the proportion of surviving colonies after 48 h (F₂,₇₇ = 0.074, p = 0.929) and 72 h (F₂,₇₇ = 0.401, p = 0.671).

Fig 4 — a) Adenine overproducing (AdeOP) strains perish at a faster rate than b) Lysine overproducing strains (LysOP). The dotted lines at 48 and 72 h for LysOP strains are estimates since mortality was first measured at 96 h.

Similarly, no differences in starvation resistance were observed among LysOPs that evolved in different environments at either 96 or 120 h. At 96 h, 44% (± 2 SE) of the ancestral LysOPs survived adenine starvation. LysOP mutualists that evolved in media without any supplemented resources had a survival rate of 40% (± 3 SE), and those evolved in media containing extra adenine and lysine had a survival rate of 43% (± 3 SE) (Fig 4). At 120 h, 32% (± 1.3 SE) of the ancestral LysOPs survived. At this time point, LysOPs that evolved in media lacking extra resources had a survival rate of 30% (± 2 SE), while those evolved in supplemented media showed a survival rate of 31% (± 2 SE). There was no significant effect of media type on the proportion of surviving colonies after 96 h (F₂,₇₉ = 0.388, p = 0.679) and 120 h (F₂,₇₉ = 0.18, p = 0.835).

Resource availability increased lysine, but not adenine, overproduction

The presence and absence of external traded resources as well as mutualist type had significant effects on the evolution of overproduction of adenine and lysine. For AdeOPs, average overproduction did not change significantly from ancestral strains, however, the direction of change was dependent on the media treatments. Overproduction of adenine by evolved strains slightly decreased in obligate conditions but slightly increased when external adenine and lysine were present in the media as compared to the ancestral (F_2,74 = 4.11, p = 0.02) (Fig 5A (a)). As a result, adenine overproduction was significantly different between the obligate and external mutualistic resources treatments. On average, overproduction by adenine mutualists reduced by 1% in the obligate environment, while it increased by 10% in the environment with external resources. Unlike AdeOPs, the lysine mutualists evolved to be the better overproducers of lysine than the ancestral strains (Fig 5B (b)). On average, lysine overproduction increased by 4% when external adenine and lysine were present in the media and by 16% in obligate media. Strains evolved in media with external resources overproduced lysine more than the ancestral strains, but less than the strains that evolved in media without external resources present (F_2,88 = 4.66, p = 0.012).

Fig 5 — Changes in A) overproduction B) resource use efficiency C) growth rate in a) adenine overproducers (AdeOP) and b) lysine overproducers (LysOP) of *Saccharomyces cerevisiae* after 4 weeks of evolution in media with and without external resources present. Statistical analyses were performed using linear mixed-effects models, with significance of fixed effects assessed using F tests (lmerTest package in R). Letters denote statistically significant differences at P < 0.05. Overproduction and resource use efficiency in both overproducer types changed depending upon the media type in which communities were maintained. Growth rate did not change significantly for either overproduction type.

Reduced efficiency evolved without growth rate compensation

Adenine mutualists evolved to be less efficient in resource use than ancestral strains and strains that evolved in media with external resources present (F_2,74 = 20.48, p < 0.0001) (Fig 5B (a)). On average, resource use efficiency decreased by 13% for AdeOPs from the environment with no external mutualistic resources, but only 3% from the environment with external resources. Strains that evolved in media with no external resources were significantly different from the ancestral strains and those evolved in the environment with external traded resources, while the strains that evolved in the environment with external resources did not differ from the ancestral. Overall LysOP strains showed a similar, but smaller change in resource use efficiency compared to AdeOP strains (F_2,88 = 5.03, p = 0.0085) (Fig 5B (b)). There was a 5% reduction in the resource use efficiency of the lysine mutualists that evolved in the media without external resource, and a 3% reduction for those evolved in the environment with external traded resources. Lysine strains that evolved in no external resources were significantly different from the ancestral, while the strains that evolved in the environment with external traded resources did not differ from the ancestral. There were no significant differences in growth rate for either mutualist type from either evolution environment (Fig 5C (a, b)).

Discussion

Mutualisms usually involve species from different kingdoms or phylogenetically divergent lineages [2]. As such, it is unlikely that the goods and services provided by each partner species will have the same effect on a mutualist’s fitness [17]. This dichotomy in fitness effects can shift the intensity of selection on traits important to facilitating mutualism for each partner. We tested this idea by forming guilds of reproductively isolated strains of brewer’s yeast that provided lysine or adenine to their partner guild. Lysine limitation is more costly for yeast physiology than adenine limitation, thus there is an inherent difference on fitness under limited resources. We also examined how the degree of reliance of partner guilds on one another for acquiring these nutrients influenced strain persistence and trait evolution for traits important to the mutualism.

In the synthetic yeast mutualism used in this study, strains differ in their ability to survive without an exchanged resource. Strains that cannot produce their own lysine (AdeOPs) persist in communities less than strains that cannot produce their own adenine (LysOPs) (Fig 2). This dichotomy in the importance of traded resources had cascading effects on individual strain persistence over the course of the experiment. AdeOP strains were more likely to go extinct in communities experiencing an obligate mutualism with LysOP strains (Figs 2–3). All four LysOP strains were retained in over 60% of communities in contrast to all four AdeOP strains being retained in less than 10% of the communities. Our results demonstrate that asymmetries in the fitness effects of traded resources lead to differential persistence of mutualists in these synthetic communities. In this system, lysine scarcity was more detrimental to AdeOP strains than adenine scarcity was to LysOP strains. Consequently, this suggests that the AdeOPs experienced stronger selection pressures and were more prone to extinction when lysine was not available externally. This aligns with the theoretical predictions that mutualists with higher dependency on a traded resource will experience greater vulnerability under obligate conditions [16].

Interestingly, when traded resources were externally supplemented, the disparity in survival between AdeOP and LysOP strains disappeared. All strains persisted in each community over the four-week period (Fig 3). This finding shows how environmental context, specifically the availability of essential nutrients, modulates the intensity of mutualistic dependence and, consequently, mutualist survival. For instance, in plant-mycorrhizal mutualisms, plants rely heavily on fungal partners for phosphorus uptake under low-nutrient conditions but reduce this reliance when soil nutrients are abundant [34–35]. Similarly, legumes form more nodules with nitrogen-fixing rhizobia in nitrogen-deficient soils but downregulate this mutualism when nitrogen is plentiful [36]. In the yeast multi-mutualist communities, the external supplementation of nutrients relaxed the evolutionary pressure that would otherwise act differentially on strains based on their resource needs. Our results confirm that the strength of mutualistic interactions is deeply context-dependent, with environmental nutrient availability acting as a key driver of mutualist persistence.

In addition to differential persistence, we also observed evolutionary changes in traits linked to mutualism in our experimental system. Resource overproduction, resource use efficiency, and growth traits evolved differently across treatments. Notably, LysOP strains significantly increased lysine overproduction under conditions without external resources, whereas AdeOP strains showed only moderate changes in adenine overproduction (Fig 5A). This suggests that the evolutionary response was greater in the mutualist guild whose partner was more vulnerable to resource limitation. LysOPs evolved increased overproduction under obligate mutualism likely in response to smaller population sizes of their AdeOPs. Increased overproduction could potentially stabilize the interaction by supporting larger populations of their more vulnerable partners. Such compensatory evolution has been observed in nature, where one partner adjusts its contribution to facilitate the mutualistic relationship, especially under asymmetrical dependence. Foster and Wenseleers [37] proposed that reciprocity can be stabilized when more capable or less-dependent partners invest disproportionately in the mutualism, especially when the partner’s survival directly affects their own benefit. Harcombe [38] also demonstrated that bacteria evolved to increase the production of essential nutrients when paired with partners that could not survive without them, thereby enhancing community stability. In the yeast multi-mutualist communities, AdeOPs appeared to be more vulnerable to lysine limitation that LysOPs were to adenine limitation. Despite being under strong selective pressure due to their heightened sensitivity to lysine limitation, AdeOPs exhibited a decrease in overproduction and resource use efficiency (Figs 5A and 5B). One potential reason for this is that the surviving AdeOP strains may have evolved strategies to conserve internal resources or reallocate energy away from overproduction toward survival. Consistent with this interpretation, strain-level overproduction assays indicated that all genotypes within each guild evolved in a similar direction, rather than a subset reducing their contribution while relying on others (Fig S1). This pattern suggests a coordinated guild-wide shift in metabolic strategy rather than the emergence of within-guild cheaters. Because AdeOP strains often occurred at lower abundances and were more prone to stochastic extinction during serial transfers, selection may have favored survival-oriented traits over efficient resource production. In this context, maintaining viability under severe resource limitation may have had greater fitness benefits than maximizing the mutualistic exchange. In environments where survival is more important than mutualistic benefit delivery, such as under extreme nutrient limitation, selection may favor metabolic reallocation toward basic maintenance and stress tolerance. Additionally, the serial propagation transfer protocol also reduced traded resources in the environment, especially for the obligate conditions, thus favoring strains that were better at growing faster rather than being more efficient [39]. This often comes at the expense of RUE [40]. A contrasting pattern was observed in pairwise yeast mutualisms, where Wang et al. [41] reported that mutualism promoted higher resource use efficiency and alleviated the tradeoff between growth and efficiency. Our results therefore suggest that in more complex multi-mutualist communities, asymmetries in the importance of exchanged resources can shift selective pressures, favoring survival-oriented strategies in more vulnerable partners rather than increased mutualistic investment. Supporting this interpretation, Vidal et al. [18] showed that many genes for AdeOP strains evolved in pairwise communities showed positive selection for intracellular transport and stress tolerance. Similar kind of adaptive resource conservation has been observed in microbial systems where organisms shift from growth to survival strategies under chronic stress [42–43]. Alternatively, the observed phenotypes could be a result of genetic drift in small populations experiencing frequent bottlenecks. In experimental evolution systems, bottlenecks reduce effective population sizes, allowing random changes in allele frequencies to fix traits irrespective of their fitness consequences [44]. In our yeast mutualism, the frequent 48 h transfer events reduced populations by 90%, which may have contributed to sudden shift in allele frequencies. Additionally, the reduced availability of nutrients when each community is transferred may have selectively favored genotypes reducing overproduction or resource use efficiency and increasing stress tolerance and survival. These genotypes may have persisted due to lineage competition, even though they are not beneficial for the mutualism in terms of overproduction.

The differences in both persistence and trait evolution show the potential for mutualist guilds to experience divergent evolutionary trajectories based on the relative fitness consequences of traded resources. While members of the guild of AdeOP strains were more prone to extinction, the ones that persisted evolved differently than members of the LysOP guild. Our findings align with previous work showing that ecological dependence and asymmetries in mutualism can lead to divergent outcomes in evolutionary dynamics [16–17]. In mutualistic networks, where multiple species interact, such asymmetries can shape community composition, stability, and coevolutionary outcomes. For example, in plant-mycorrhizal networks, some fungi have many interactions and support a range of host plants, while others are more specialized [45–46]. The stability of such networks may depend on the persistence of these mutualists, which in turn is shaped by their dependence on traded resources.

In conclusion, our study demonstrates that fitness asymmetries in traded mutualistic resources play a key role in shaping both the ecological persistence and evolutionary trajectories of mutualists. Mutualists that suffer greater fitness consequences from lack of traded resources are more likely to go extinct under obligate conditions, while their partners may evolve increased overproduction to compensate for such partner loss. This pattern changes, when traded resources are available from external sources. These results suggest that the balance of mutualistic interactions is sensitive to environmental context and the fitness value of exchanged goods. This work contributes to a growing understanding of how mutualisms evolve and persist, particularly in the context of multi-species interactions and environmental perturbations. Future studies should explore whether similar patterns occur in more complex, natural mutualistic systems and examine the long-term consequences of asymmetric dependency for coevolution and community structure. This system also provides a strong foundation for future modeling work aimed at quantifying how extinction risk and survival-based selection pressures shape evolutionary trajectories in multi-partner mutualisms.

Supporting information

S1 Table. Yeast strains used in this study.

(DOCX)

pone.0340707.s001.docx^{(16.1KB, docx)}

S2 Table. Community of yeast mutualists used for phenotyping.

“X” indicates the absence of a strain. In no external resource (obligate) media, only three communities retained all eight strains by the end of the four-week period. Therefore, we also considered communities with fewer than eight strains.

(DOCX)

pone.0340707.s002.docx^{(17.8KB, docx)}

S1 Fig. Changes in strain-level overproduction of adenine or lysine in a) adenine overproducers (AdeOP) or b) lysine overproducers (LysOP) of Saccharomyces cerevisiae after 4 weeks of evolution in media with and without external resources present.

(TIF)

pone.0340707.s003.tif^{(15.9MB, tif)}

Acknowledgments

The authors thank M. Vidal and C. Liu for their valuable discussion.

Data Availability

The data underlying the results presented in this study are publicly available through Dryad at the following DOI: https://doi.org/10.5061/dryad.kwh70rzjq.

Funding Statement

Grants from the U.S. National Science Foundation (DEB 2137554). The funders had no role in study design, data collection and analysis, decision to publish, or the preparation of the manuscript.

References

1.Price PW. Species interactions and the. Plant animal interactions: an evolutionary approach. 2002. [Google Scholar]
2.Bronstein JL. Mutualism. Oxford University Press; 2015. [Google Scholar]
3.Stachowicz JJ. Mutualism, facilitation, and the structure of ecological communities: positive interactions play a critical, but underappreciated, role in ecological communities by reducing physical or biotic stresses in existing habitats and by creating new habitats on which many species depend. Bioscience. 2001;51(3):235–46. [Google Scholar]
4.Silknetter S, Creed RP, Brown BL, Frimpong EA, Skelton J, Peoples BK. Positive biotic interactions in freshwaters: a review and research directive. Freshwater Biology. 2020;65(4):811–32. doi: 10.1111/fwb.13476 [DOI] [Google Scholar]
5.Kost C, Patil KR, Friedman J, Garcia SL, Ralser M. Metabolic exchanges are ubiquitous in natural microbial communities. Nat Microbiol. 2023;8(12):2244–52. doi: 10.1038/s41564-023-01511-x [DOI] [PubMed] [Google Scholar]
6.Bascompte J, Jordano P. Plant-animal mutualistic networks: the architecture of biodiversity. Annu Rev Ecol Evol Syst. 2007;38(1):567–93. [Google Scholar]
7.Magkourilou E, Bell CA, Daniell TJ, Field KJ. The functionality of arbuscular mycorrhizal networks across scales of experimental complexity and ecological relevance. Function Ecol. 2024;39(6):1384–99. doi: 10.1111/1365-2435.14618 [DOI] [Google Scholar]
8.Herre E, Knowlton N, Mueller U, Rehner S. The evolution of mutualisms: exploring the paths between conflict and cooperation. Trends Ecol Evol. 1999;14(2):49–53. doi: 10.1016/s0169-5347(98)01529-8 [DOI] [PubMed] [Google Scholar]
9.Chacón JM, Hammarlund SP, Martinson JN, Smith Jr LB, Harcombe WR. The ecology and evolution of model microbial mutualisms. Ann Rev Ecol Evol System. 2021;52(1):363–84. [Google Scholar]
10.Pauli B, Oña L, Hermann M, Kost C. Obligate mutualistic cooperation limits evolvability. Nat Commun. 2022;13(1):337. doi: 10.1038/s41467-021-27630-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
11.Apprill A, Marlow HQ, Martindale MQ, Rappé MS. The onset of microbial associations in the coral Pocillopora meandrina. ISME J. 2009;3(6):685–99. doi: 10.1038/ismej.2009.3 [DOI] [PubMed] [Google Scholar]
12.Russell JA, Moreau CS, Goldman-Huertas B, Fujiwara M, Lohman DJ, Pierce NE. Bacterial gut symbionts are tightly linked with the evolution of herbivory in ants. Proc Natl Acad Sci U S A. 2009;106(50):21236–41. doi: 10.1073/pnas.0907926106 [DOI] [PMC free article] [PubMed] [Google Scholar]
13.van der Heijden MGA, Klironomos JN, Ursic M, Moutoglis P, Streitwolf-Engel R, Boller T, et al. Mycorrhizal fungal diversity determines plant biodiversity, ecosystem variability and productivity. Nature. 1998;396(6706):69–72. doi: 10.1038/23932 [DOI] [Google Scholar]
14.Afkhami ME, Rudgers JA, Stachowicz JJ. Multiple mutualist effects: conflict and synergy in multispecies mutualisms. Ecology. 2014;95(4):833–44. doi: 10.1890/13-1010.1 [DOI] [PubMed] [Google Scholar]
15.Chomicki G, Weber M, Antonelli A, Bascompte J, Kiers ET. The impact of mutualisms on species richness. Trends Ecol Evol. 2019;34(8):698–711. doi: 10.1016/j.tree.2019.03.003 [DOI] [PubMed] [Google Scholar]
16.Chomicki G, Kiers ET, Renner SS. The evolution of mutualistic dependence. Ann Rev Ecol Evol System. 2020;51(1):409–32. [Google Scholar]
17.Wang RW, Shi AN, Zhang XW, Liu M, Jandér KC, Dunn DW. Asymmetric and uncertain interactions within mutualisms. J Plant Ecol. 2024;17(1):rtad042. [Google Scholar]
18.Vidal MC, Liu C, Wang S, Segraves KA. Partner dependency alters patterns of coevolutionary selection in mutualisms. Proc Natl Acad Sci U S A. 2025;122(21):e2424983122. doi: 10.1073/pnas.2424983122 [DOI] [PMC free article] [PubMed] [Google Scholar]
19.Dawkins R, Krebs JR. Arms races between and within species. Proc R Soc Lond B Biol Sci. 1979;205(1161):489–511. doi: 10.1098/rspb.1979.0081 [DOI] [PubMed] [Google Scholar]
20.Agarwal R, Althoff DM. Extreme specificity in obligate mutualism-A role for competition?. Ecol Evol. 2024;14(6):e11628. doi: 10.1002/ece3.11628 [DOI] [PMC free article] [PubMed] [Google Scholar]
21.Toby Kiers E, Palmer TM, Ives AR, Bruno JF, Bronstein JL. Mutualisms in a changing world: an evolutionary perspective. Ecol Lett. 2010;13(12):1459–74. doi: 10.1111/j.1461-0248.2010.01538.x [DOI] [PubMed] [Google Scholar]
22.Keeler AM. Tritrophic mutualisms in a changing climate. Riverside: University of California; 2022. [Google Scholar]
23.Magnoli SM, Keller KR, Lau JA. Mutualisms in a warming world: How increased temperatures affect the outcomes of multi-mutualist interactions. Ecology. 2023;104(3):e3955. doi: 10.1002/ecy.3955 [DOI] [PubMed] [Google Scholar]
24.Shou W, Ram S, Vilar JMG. Synthetic cooperation in engineered yeast populations. Proc Natl Acad Sci U S A. 2007;104(6):1877–82. doi: 10.1073/pnas.0610575104 [DOI] [PMC free article] [PubMed] [Google Scholar]
25.Vidal MC, Wang SP, Rivers DM, Althoff DM, Segraves KA. Species richness and redundancy promote persistence of exploited mutualisms in yeast. Science. 2020;370(6514):346–50. doi: 10.1126/science.abc1234 [DOI] [PubMed] [Google Scholar]
26.Hart SFM, Skelding D, Waite AJ, Burton JC, Shou W. High-throughput quantification of microbial birth and death dynamics using fluorescence microscopy. Quant Biol. 2019;7(1):69–81. doi: 10.1007/s40484-018-0160-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
27.VanDusen WJ, Fu J, Bailey FJ, Burke CJ, Herber WK, George HA. Adenine quantitation in yeast extracts and fermentation media and its relationship to protein expression and cell growth in adenine auxotrophs of Saccharomyces cerevisiae. Biotechnol Prog. 1997;13(1):1–7. doi: 10.1021/bp9600896 [DOI] [PubMed] [Google Scholar]
28.Kokina A, Kibilds J, Liepins J. Adenine auxotrophy--be aware: some effects of adenine auxotrophy in Saccharomyces cerevisiae strain W303-1A. FEMS Yeast Res. 2014;14(5):697–707. doi: 10.1111/1567-1364.12154 [DOI] [PubMed] [Google Scholar]
29.Kokina A, Tanilas K, Ozolina Z, Pleiko K, Shvirksts K, Vamza I, et al. Purine Auxotrophic Starvation Evokes Phenotype Similar to Stationary Phase Cells in Budding Yeast. J Fungi (Basel). 2021;8(1):29. doi: 10.3390/jof8010029 [DOI] [PMC free article] [PubMed] [Google Scholar]
30.Boer VM, Crutchfield CA, Bradley PH, Botstein D, Rabinowitz JD. Growth-limiting intracellular metabolites in yeast growing under diverse nutrient limitations. Mol Biol Cell. 2010;21(1):198–211. doi: 10.1091/mbc.e09-07-0597 [DOI] [PMC free article] [PubMed] [Google Scholar]
31.Eisler H, Fröhlich K-U, Heidenreich E. Starvation for an essential amino acid induces apoptosis and oxidative stress in yeast. Exp Cell Res. 2004;300(2):345–53. doi: 10.1016/j.yexcr.2004.07.025 [DOI] [PubMed] [Google Scholar]
32.Lewis AG, Carmichael L, Wang RY, Gibney PA. Characterizing a panel of amino acid auxotrophs under auxotrophic starvation conditions. Yeast. 2024;41(1–2):5–18. doi: 10.1002/yea.3910 [DOI] [PubMed] [Google Scholar]
33.Bates D, Mächler M, Bolker B, Walker S. Fitting linear mixed-effects models Usinglme4. J Stat Soft. 2015;67(1). doi: 10.18637/jss.v067.i01 [DOI] [Google Scholar]
34.Johnson NC. Can Fertilization of Soil Select Less Mutualistic Mycorrhizae?. Ecol Appl. 1993;3(4):749–57. doi: 10.2307/1942106 [DOI] [PubMed] [Google Scholar]
35.Hoeksema JD, Chaudhary VB, Gehring CA, Johnson NC, Karst J, Koide RT, et al. A meta-analysis of context-dependency in plant response to inoculation with mycorrhizal fungi. Ecol Lett. 2010;13(3):394–407. doi: 10.1111/j.1461-0248.2009.01430.x [DOI] [PubMed] [Google Scholar]
36.Regus JU, Wendlandt CE, Bantay RM, Gano-Cohen KA, Gleason NJ, Hollowell AC, et al. Nitrogen deposition decreases the benefits of symbiosis in a native legume. Plant Soil. 2016;414(1–2):159–70. doi: 10.1007/s11104-016-3114-8 [DOI] [Google Scholar]
37.Foster KR, Wenseleers T. A general model for the evolution of mutualisms. J Evol Biol. 2006;19(4):1283–93. doi: 10.1111/j.1420-9101.2005.01073.x [DOI] [PubMed] [Google Scholar]
38.Harcombe W. Novel cooperation experimentally evolved between species. Evolution. 2010;64(7):2166–72. [DOI] [PubMed] [Google Scholar]
39.Bachmann H, Bruggeman FJ, Molenaar D, Branco Dos Santos F, Teusink B. Public goods and metabolic strategies. Curr Opin Microbiol. 2016;31:109–15. doi: 10.1016/j.mib.2016.03.007 [DOI] [PubMed] [Google Scholar]
40.Frank SA. The trade-off between rate and yield in the design of microbial metabolism. J Evol Biol. 2010;23(3):609–13. doi: 10.1111/j.1420-9101.2010.01930.x [DOI] [PubMed] [Google Scholar]
41.Wang S, Agarwal R, Segraves KA, Althoff DM. Trait and plasticity evolution under competition and mutualism in evolving pairwise yeast communities. PLoS One. 2025;20(1):e0311674. doi: 10.1371/journal.pone.0311674 [DOI] [PMC free article] [PubMed] [Google Scholar]
42.Lenski RE, Rose MR, Simpson SC, Tadler SC. Long-term experimental evolution in Escherichia coli. I. Adaptation and divergence during 2,000 generations. The American Naturalist. 1991;138(6):1315–41. [Google Scholar]
43.Dragosits M, Mattanovich D. Adaptive laboratory evolution -- principles and applications for biotechnology. Microb Cell Fact. 2013;12:64. doi: 10.1186/1475-2859-12-64 [DOI] [PMC free article] [PubMed] [Google Scholar]
44.Gerrish PJ, Lenski RE. The fate of competing beneficial mutations in an asexual population. Genetica. 1998;102–103(1–6):127–44. doi: 10.1023/a:1017067816551 [DOI] [PubMed] [Google Scholar]
45.Sepp S-K, Davison J, Jairus T, Vasar M, Moora M, Zobel M, et al. Non-random association patterns in a plant-mycorrhizal fungal network reveal host-symbiont specificity. Mol Ecol. 2019;28(2):365–78. doi: 10.1111/mec.14924 [DOI] [PubMed] [Google Scholar]
46.Tedersoo L, Bahram M, Zobel M. How mycorrhizal associations drive plant population and community biology. Science. 2020;367(6480):eaba1223. doi: 10.1126/science.aba1223 [DOI] [PubMed] [Google Scholar]

PLoS One. doi: 10.1371/journal.pone.0340707.r001

Decision Letter 0

Karthik Raman

21 Oct 2025

Dear Dr. Agarwal,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

Please submit your revised manuscript by Dec 05 2025 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org . When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.
A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.
An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: https://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols . Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols .

We look forward to receiving your revised manuscript.

Kind regards,

Karthik Raman, Ph.D.

Academic Editor

PLOS ONE

Journal Requirements:

When submitting your revision, we need you to address these additional requirements.

1.Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and

https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2. When completing the data availability statement of the submission form, you indicated that you will make your data available on acceptance. We strongly recommend all authors decide on a data sharing plan before acceptance, as the process can be lengthy and hold up publication timelines. Please note that, though access restrictions are acceptable now, your entire data will need to be made freely accessible if your manuscript is accepted for publication. This policy applies to all data except where public deposition would breach compliance with the protocol approved by your research ethics board. If you are unable to adhere to our open data policy, please kindly revise your statement to explain your reasoning and we will seek the editor's input on an exemption. Please be assured that, once you have provided your new statement, the assessment of your exemption will not hold up the peer review process.

3. If the reviewer comments include a recommendation to cite specific previously published works, please review and evaluate these publications to determine whether they are relevant and should be cited. There is no requirement to cite these works unless the editor has indicated otherwise.

4. Please review your reference list to ensure that it is complete and correct. If you have cited papers that have been retracted, please include the rationale for doing so in the manuscript text, or remove these references and replace them with relevant current references. Any changes to the reference list should be mentioned in the rebuttal letter that accompanies your revised manuscript. If you need to cite a retracted article, indicate the article’s retracted status in the References list and also include a citation and full reference for the retraction notice.

Additional Editor Comments:

The reviews for your manuscript are now in. While the reviewers appreciate the contributions, they have also raised several major concerns, which must be addressed for further consideration of the manuscript for publication. Please take into full consideration the comments by the reviewers and submit a revised version of your manuscript.

[Note: HTML markup is below. Please do not edit.]

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

**********

2. Has the statistical analysis been performed appropriately and rigorously? -->?>

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

**********

3. Have the authors made all data underlying the findings in their manuscript fully available??>

The PLOS Data policy

Reviewer #1: No

Reviewer #2: No

Reviewer #3: Yes

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English??>

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

**********

Reviewer #1: The article investigates how fitness asymmetries in exchanged resources shape the evolutionary dynamics and persistence of species using synthetic multi-mutualist yeast communities. The authors have used 8 genetically distinct strains of Saccharomyces cerevisiae that either overproduce adenine or lysine but cannot produce the other, the authors simulate mutualism under environments with and without external nutrient supplementation. The study integrates the ecological persistence with trait evolution under different resource contexts, providing direct experimental evidence for how asymmetric fitness consequences can shape mutualism stability. The study is well executed and methodologically rigorous.

Below are my specific comments to improve clarity and presentation. With these minor changes the manuscript is acceptable for publication.

I would recommend the authors to better represent the model being studied. The current figure is confusing and hard to comprehend. My suggestion is to have a better schematic or flow diagram in the Methods section illustrating how the strains were mixed together, why 96 distinct communities were developed, and why this resulted in 288 evolving cultures under various conditions. This would make the design much more understandable to readers.Maintain clear color codes for genetically different strains.

Further, instead of a section named as “strains used in this study” it could be more like “ synthetic model of multi-mutualist communities” . This would emphasize that the work is not only about the strains themselves but about the synthetic mutualism model being tested.

While the use of a growth-based bioassay to estimate adenine and lysine overproduction is valid and functionally relevant, direct quantification (e.g., HPLC, LC-MS, or kit-based metabolite assays) would provide complementary information on exact metabolite levels. I encourage the authors to comment on the choice of indirect quantification based method.

In the "Overproduction measurements" part, Authors have stated that the LysOPs were heat-shocked to release lysine, which is released primarily upon cell death. For the AdeOPs, they just harvested the filtrate. A minor, complementary experiment would be to subject the AdeOPs to the same heat shock and filtration procedure to determine whether it produces more or a greater amount of adenine that could then be utilized to ascertain that adenine is actually released more quickly and more easily than lysine. This could be a minor addition to the procedures. In the case that this is not done, please clarify it in the methodology or discussion

Minor edits:

Line no : 317 (stains to strains )

Ensure consistency in figure references (e.g., “Fig 5a” vs. “Figure 5a”).

A general proofreading pass to correct minor typographical and grammatical errors will improve readability.

Standardize terminology (sometimes “mutualistic goods,” sometimes “traded resources”).

Reviewer #2: Agarwal et al. studied the asymmetric fitness effects of traded resources in a multi-mutualist community. The overall study is excellent, has many strengths, and is written clearly. I only have a few comments (all well within the scope of this paper and without the need for any experiments) that could further improve the study.

I liked the overall findings. Asymmetric fitness effects led to asymmetric trait evolution. Notably, overproduction by lysine producers increased, potentially in a graded response to the expected selection pressure, while adenine overproduction showed very little change.

How does a guild change trait evolution in an asymmetric setting? In my opinion, it is important to discuss how selection might change when moving from a single strain to a combination of strains performing the same function. This is the main idea of this paper. In fact, since you have the data, it would be worthwhile to test whether the results on average trait evolution presented here would differ if considered at the genotype-specific level. I further explain these points below.

Is there any baseline within-guild genotypic variability? Do all genotypes in a guild have the same fitness, with no measurable differences among them, or is it only the case that they all overproduce adenine and underproduce lysine in similar ways, while other differences have not been measured?

Is there a persistent genotype? I would not expect any particular genotype to consistently persist across communities. This can be easily tested.

Do within-guild cheaters evolve? Is overproduction costly? If it is, could some strains within a guild reduce overproduction and act as cheaters, forcing other strains of the same guild to compensate by producing more? Again, this should be straightforward to test by comparing overproduction in specific genotypes, as opposed to comparing only averages from genotype-mixed populations before and after evolution.

Expectation from pairwise combination. Since some results seem to diverge from your previous pairwise study (only one genotype per guild; Wang et al. [39]), could you compare these results against expectations from the pairwise system beyond what is presented in the Discussion?

Ideally, if there is no within-guild variability, then, on average, the species in a multi-mutualist community would feel similar competition and cooperation, and thus their average persistence would match the expectation from the pairwise system. You can easily test this by comparing your pairwise system and multi-mutualist community results side by side. If this comparison is not readily feasible, I understand.

In addition, because of likely stochastic extinction in each transfer window, the selective pressure on a strain with low abundance would be to focus more on survival and could contribute to the reduced resource-use efficiency in AdeOP (a 13% decrease in obligate and 3% in supplemented), while the same did not happen to LysOP (a 5% decrease in obligate and 3% in supplemented). This can also be tested from the available data.

Modeling opportunity for the future. This experimentally grounded system, together with the evolution-specific results from this study, provides an ideal foundation for a computational model that can quantitatively dissect the remaining ecological dynamics and illuminate complex interaction rules beyond net outcomes.

Minor suggestions

Line 308: Increased by 10%??

On average, lysine overproduction increased by 4% when external adenine and lysine were present in the media and by 16% in obligate media. Strains evolved in media without external resources overproduced lysine more than the ancestral strains, but less than the strains which evolved in media with external resources present (F2,88= 4.66, p= 0.012).

It seems inconsistent to me. Should it not instead be, “Strains evolved in media with external resources overproduced lysine more than the ancestral strains, but less than the strains that evolved in media without external resources present”?

In Fig. 3, AdeOP1 represents the arbitrarily assigned first member of the guild, right? If so, the figure must either come from a specific community or represent an average across combinations. Why does AdeOP1 appear significantly different from the others? Why are there no error bars if this represents an average?

In Fig. 4, could you make the x-axis range the same in both?

Figures 5, 6, and 7 could be combined into a single figure. This would help readers view all trait-evolution measurements in one place. Units are missing on all three y-axes. It would be helpful to use the same y-axis range, at least in the growth-rate figure.

Optional: Could you add a panel in Fig. 1 that visually demonstrates the asymmetric effects of adenine and lysine starvation on their dynamics compared to the wild type, coupled with the corresponding adenine and lysine concentrations in the medium? This would make the conceptual asymmetry more intuitive.

Also, please ensure that all figures appear at high resolution in the published version.

Overall, this is an excellent study. I believe the revisions will make the paper more insightful and complete.

Reviewer #3: 1. In methods section, a bit more discussion on the genetic backgrounds of the strains will be helpful.

2. Effective population size during the evolution experiment should be mentioned so that the readers can have an idea about the strength of selection.

3. Why single colonies of individual strains were used for phenotyping after evolution experiment? This would remove

genetic and phenotypic diversity. Only the fixed mutations will be seen and their phenotypes will be measured.

4. 1% lysine - v/v?

5. Is it possible to directly measure lysine and adenine conc. in the filtrates?

6. The statements in lines 303 and 307 seem to be contradicting each other.

7. Line 317 - strains

8. Decreased resource use efficiency in AdeOP strains - I presume that this is for lysine use.

However, we see lysine overproduction in lysOP strain.

Is it possible that the reduction in RUE in AdeOP is caused by lysine overproduction?

9. It is quite puzzling to see reduction in resource use efficiency in LysOP and AdeOP communities in obligate condition.

Why would this happen? It would be good if the authors can discuss this point along with some plausible explanations.

10. Sub-headings in the results could be changed to statements that summarize the results discussed in that section

**********

what does this mean? ). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy

Reviewer #1: No

Reviewer #2: Yes: Aamir Ansari

Reviewer #3: No

**********

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

To ensure your figures meet our technical requirements, please review our figure guidelines: https://journals.plos.org/plosone/s/figures

You may also use PLOS’s free figure tool, NAAS, to help you prepare publication quality figures: https://journals.plos.org/plosone/s/figures#loc-tools-for-figure-preparation.

NAAS will assess whether your figures meet our technical requirements by comparing each figure against our figure specifications.

Attachment

Submitted filename: Review_PONE-D-25-49040.docx

pone.0340707.s004.docx^{(14.1KB, docx)}

PLoS One. 2026 Feb 3;21(2):e0340707. doi: 10.1371/journal.pone.0340707.r002

Author response to Decision Letter 1

4 Dec 2025

Comments to Reviewers

We sincerely thank the reviewers for their thoughtful and constructive comments, which have helped us strengthen the manuscript. Below, we provide point-by-point responses and describe the corresponding revisions.

Reviewer #1:

The article investigates how fitness asymmetries in exchanged resources shape the evolutionary dynamics and persistence of species using synthetic multi-mutualist yeast communities. The authors have used 8 genetically distinct strains of Saccharomyces cerevisiae that either overproduce adenine or lysine but cannot produce the other, the authors simulate mutualism under environments with and without external nutrient supplementation. The study integrates the ecological persistence with trait evolution under different resource contexts, providing direct experimental evidence for how asymmetric fitness consequences can shape mutualism stability. The study is well executed and methodologically rigorous.

Below are my specific comments to improve clarity and presentation. With these minor changes the manuscript is acceptable for publication.

Comment 1: I would recommend the authors to better represent the model being studied. The current figure is confusing and hard to comprehend. My suggestion is to have a better schematic or flow diagram in the Methods section illustrating how the strains were mixed together, why 96 distinct communities were developed, and why this resulted in 288 evolving cultures under various conditions. This would make the design much more understandable to readers. Maintain clear color codes for genetically different strains.

Response 1: We understand reviewer’s concern. We have now updated figure 1 for better representation.

Comment 2: Further, instead of a section named as “strains used in this study” it could be more like “synthetic model of multi-mutualist communities”. This would emphasize that the work is not only about the strains themselves but about the synthetic mutualism model being tested.

Response 2: We have changed the “Strains used in this study” to “Synthetic model of multi-mutualist communities”

Comment 3: While the use of a growth-based bioassay to estimate adenine and lysine overproduction is valid and functionally relevant, direct quantification (e.g., HPLC, LC-MS, or kit-based metabolite assays) would provide complementary information on exact metabolite levels. I encourage the authors to comment on the choice of indirect quantification-based method.

Response 3: We thank the reviewer for this insightful comment. We agree that direct quantification methods such as HPLC or LC-MS would yield absolute concentrations of adenine and lysine and provide a more direct measure of metabolite production. However, we used a growth-based bioassay to estimate adenine and lysine overproduction because it provides a functional measure of the metabolites available to mutualistic partners. This indirect approach integrates both metabolite release and biological accessibility, offering an ecologically relevant estimate of the exchangeable resource. Our study focuses on the relative functional differences among treatments rather than absolute metabolite levels. Moreover, this assay is well-validated for Saccharomyces cerevisiae auxotrophs (e.g., Vidal et al. 2020; Shou et al. 2007), allowing us to maintain consistency with prior studies. Additionally, we explored this possibility in the early stages of working with this system, but realized the number of samples we would have to process would make it monetarily prohibitive.

Comment 4: In the "Overproduction measurements" part, Authors have stated that the LysOPs were heat-shocked to release lysine, which is released primarily upon cell death. For the AdeOPs, they just harvested the filtrate. A minor, complementary experiment would be to subject the AdeOPs to the same heat shock and filtration procedure to determine whether it produces more or a greater amount of adenine that could then be utilized to ascertain that adenine is actually released more quickly and more easily than lysine. This could be a minor addition to the procedures. In the case that this is not done, please clarify it in the methodology or discussion

Response 4: We thank the reviewer for this suggestion. We agree that symmetric extraction protocols could improve comparability between overproducer types. However, adenine and lysine differ fundamentally in their cellular release mechanisms. Adenine is secreted continuously into the medium by AdeOP strains during active growth, while lysine is largely retained intracellularly and released upon cell death and lysis. For this reason, we directly harvested the AdeOP filtrate without heat shock, which accurately represents the biologically available adenine under growth conditions. We have now clarified this rationale in the Methods section (line 243-246):

“We did not heat shock the adenine overproducers so that we would have a more accurate estimate of the adenine excreted without the addition of adenine from disrupted cells. Thus, separate extraction protocols were used to accurately represent biologically available metabolite levels for each overproducer type.”

Minor edits:

Comment 5: Line no: 317 (stains to strains )

Response 5: Changed “stains” to “strains” (now in line 337).

Comment 6: Ensure consistency in figure references (e.g., “Fig 5a” vs. “Figure 5a”).

Response 6: We have used “Fig” across entire manuscript.

Comment 7: A general proofreading pass to correct minor typographical and grammatical errors will improve readability.

Response 7: We thank the reviewer for this suggestion. The entire manuscript has been carefully proofread to correct typographical and grammatical errors, ensuring improved clarity and readability.

Comment 8: Standardize terminology (sometimes “mutualistic goods,” sometimes “traded resources”).

Response 8: We have now changed “mutualistic goods” to “traded resources”.

Reviewer #2

Agarwal et al. studied the asymmetric fitness effects of traded resources in a multi-mutualist community. The overall study is excellent, has many strengths, and is written clearly. I only have a few comments (all well within the scope of this paper and without the need for any experiments) that could further improve the study.

Comment 1: I liked the overall findings. Asymmetric fitness effects led to asymmetric trait evolution. Notably, overproduction by lysine producers increased, potentially in a graded response to the expected selection pressure, while adenine overproduction showed very little change.

Response 1: We appreciate the reviewer’s insightful point regarding potential within-guild genotypic variation. Each strain contributing to a guild was initiated from a single colony, minimizing within-strain genetic variation at the start of the experiment. However, to explicitly evaluate baseline differences among genotypes within the AdeOP guild, we compared ancestral strains grouped by family. Growth rate did not differ significantly among families (F₍3,28₎ = 2.66, p = 0.068), and no pairwise contrasts were significant (p > 0.06). These results indicate that baseline genotypic variability for this trait was limited within the AdeOP guild. Therefore, the guild-level patterns reported in the main text are not driven by pre-existing fitness differences among genotypes, but instead reflect evolutionary responses at the functional-group level. In addition, we examined baseline variation in resource overproduction among AdeOP families. Overproduction differed significantly among families (ANOVA: F₍3,28₎ = 6.57, p = 0.0017), with post-hoc tests indicating that genotype of 1057-1058 exhibited slightly lower overproduction levels compared to the other families (p < 0.012). Although this effect indicates some baseline variation within the AdeOP guild, the magnitude of these differences was modest, and does not alter the overall guild-level conclusions.

We performed the same analysis for the LysOP guild. Ancestral LysOP families did not differ in growth rate (ANOVA: F₍3,26₎ = 0.11, p = 0.95), and no significant differences were detected in overproduction (ANOVA: F₍3,26₎ = 2.07, p = 0.13). We observed a moderate difference in resource use efficiency (ANOVA: F₍3,26₎ = 5.11, p = 0.0065), driven by slightly higher values in strain 1085–1086. Together with the AdeOP analyses, these results show that baseline variation in growth-related performance among genotypes within each guild is limited, while modest differences are present. Therefore, the guild-level trait evolution patterns we report are not the result of strong pre-existing fitness asymmetries among genotypes but instead reflect evolutionary responses occurring at the functional-group level.

Comment 2: Is there a persistent genotype? I would not expect any particular genotype to consistently persist across communities. This can be easily tested.

Response 2: We assume that ‘persistent genotype” means “persistent strain”. We examined persistence at the strain level using the strain-specific data now presented in Fig. 3. While there were differences in the likelihood of persistence among strains within each guild, no strains consistently persisted across all communities. Thus, although persistence varied among strains, these differences did not drive the evolutionary patterns we report, and we do not find evidence for a single “persistent strain” or strain-specific evolutionary advantage. It is possible that genotypes within a strain could evolve and be persistent but testing that was beyond the scope of our study.

Comment 3: Do within-guild cheaters evolve? Is overproduction costly? If it is, could some strains within a guild reduce overproduction and act as cheaters, forcing other strains of the same guild to compensate by producing more? Again, this should be straightforward to test by comparing overproduction in specific genotypes, as opposed to comparing only averages from genotype-mixed populations before and after evolution.

Response 3: We examined overproduction at the level of individual genotypes within each guild to test whether any strain reduced its contribution disproportionately, which would indicate the emergence of within-guild cheaters. The strain-wise overproduction measurements (now shown in Fig. S1) revealed consistent evolutionary shifts across all genotypes within each guild. AdeOP strains showed a similar reduction in overproduction relative to their ancestral values, and LysOP strains showed a similar increase, with no strain deviating strongly from the guild-wide pattern. Therefore, we do not find evidence that any strain evolved a reduced contribution while relying on others to compensate. The observed changes instead reflect a coordinated, guild-wide evolutionary response rather than the evolution of within-guild cheaters (line 411-418).

Comment 4: Expectation from pairwise combination. Since some results seem to diverge from your previous pairwise study (only one genotype per guild; Wang et al. [39]), could you compare these results against expectations from the pairwise system beyond what is presented in the Discussion?

Response 4: We agree that comparing to expectations from our previous pairwise system is informative. However, a direct numerical comparison is not appropriate because (i) the pairwise study used a single genotype per guild while our communities contain four per guild, introducing within-guild competition and frequency-dependent effects; (ii) our outcome is persistence across replicate communities (binomial), not continuous growth/efficiency, so the mapping from pairwise dynamics to multi-member persistence is non-linear (iii) the studies were done at different times and with different media, researchers, and other factors that could influence the results. The reviewer’s question is a great one and as part of another study we are going to directly compare the pairwise and 4x4 communities since we are evolving them at the same time.

Comment 5: In addition, because of likely stochastic extinction in each transfer window, the selective pressure on a strain with low abundance would be to focus more on survival and could contribute to the reduced resource-use efficiency in AdeOP (a 13% decrease in obligate and 3% in supplemented), while the same did not happen to LysOP (a 5% decrease in obligate and 3% in supplemented). This can also be tested from the available data.

Response 5: We thank the reviewer for this suggestion. We agree that repeated population bottlenecks may have imposed stronger selection on AdeOP strains to prioritize survival under resource limitation, potentially contributing to the greater reduction in resource-use efficiency observed in AdeOP relative to LysOP strains. We have now added this interpretation to the Discussion, and added a note that this system provides opportunities for future modeling to examine the role of extinction risk and survival-based selection in shaping mutualist evolution.

Line 414-418: Because AdeOP strains often occurred at lower abundances and were more prone to stochastic extinction during serial transfers, selection may have favored survival-oriented traits over efficient resource production. In this context, maintaining viability under severe resource limitation may have had greater fitness benefits than maximizing the mutualistic exchange.

Line 464-466: This system also provides a strong foundation for future modeling work aimed at quantifying how extinction risk and survival-based selection pressures shape evolutionary trajectories in multi-partner mutualisms.

Minor suggestions:

Comment 6: Line 308: Increased by 10%??

Response 6: We agree with the reviewer. We have now made the corrections in line 327-329:

“On average, overproduction by adenine mutualists reduced by 1 % in the obligate environment, while it increased by 10% in the environment with external resources.”

Comment 7: On average, lysine overproduction increased by 4% when external adenine and lysine were present in the media and by 16% in obligate media. Strains evolved in media without external resources overproduced lysine more than the ancestral strains, but less than the strains which evolved in media with external resources present (F2,88= 4.66, p= 0.012).

Response 7: We agree with the comment. We have now corrected the sentence as per suggestion (line 331-334)

Comment 8: In Fig. 3, AdeOP1 represents the arbitrarily assigned first member of the guild, right? If so, the figure must either come from a specific commu

Attachment

Submitted filename: Response_to_Reviewers.docx

pone.0340707.s006.docx^{(31.7KB, docx)}

PLoS One. doi: 10.1371/journal.pone.0340707.r003

Decision Letter 1

Karthik Raman

21 Dec 2025

Dear Dr. Agarwal,

Please submit your revised manuscript by Feb 04 2026 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org . When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

A letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.
A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.
An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

We look forward to receiving your revised manuscript.

Kind regards,

Karthik Raman, Ph.D.

Academic Editor

PLOS One

Journal Requirements:

If the reviewer comments include a recommendation to cite specific previously published works, please review and evaluate these publications to determine whether they are relevant and should be cited. There is no requirement to cite these works unless the editor has indicated otherwise.

Please review your reference list to ensure that it is complete and correct. If you have cited papers that have been retracted, please include the rationale for doing so in the manuscript text, or remove these references and replace them with relevant current references. Any changes to the reference list should be mentioned in the rebuttal letter that accompanies your revised manuscript. If you need to cite a retracted article, indicate the article’s retracted status in the References list and also include a citation and full reference for the retraction notice.

Additional Editor Comments:

The manuscript is nearly ready. Pls. Make the minor fixes required by the reviewers and I will accept the manuscript right away. Congratulations on a solid piece of work.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

Reviewer #1: All comments have been addressed

Reviewer #2: All comments have been addressed

Reviewer #3: All comments have been addressed

**********

2. Is the manuscript technically sound, and do the data support the conclusions??>

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

**********

3. Has the statistical analysis been performed appropriately and rigorously? -->?>

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

**********

4. Have the authors made all data underlying the findings in their manuscript fully available??>

The PLOS Data policy

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English??>

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

**********

Reviewer #1: (No Response)

Reviewer #2: The authors have adequately addressed all of my major and minor concerns, and I recommend the paper for publication. I congratulate the authors on a well-executed study. I have one minor change request that does not require further review.

Figure 4 (Comment 9): I understand and agree with your concern, but my suggestion was purely graphical. Please keep the same x-axis limits to aid visual comparison, even if one curve ends earlier (the remaining region can be left blank). No additional measurements required.

Minor check: Please verify the reported F values around lines 301–303.

Reviewer #3: The authors have addressed all my previous comments satisfactorily.

However, in the caption of figure 5, the authors should mention the name of the statistical test that is being performed.

**********

what does this mean? ). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy

Reviewer #1: No

Reviewer #2: No

Reviewer #3: No

**********

To ensure your figures meet our technical requirements, please review our figure guidelines: https://journals.plos.org/plosone/s/figures

You may also use PLOS’s free figure tool, NAAS, to help you prepare publication quality figures: https://journals.plos.org/plosone/s/figures#loc-tools-for-figure-preparation.

NAAS will assess whether your figures meet our technical requirements by comparing each figure against our figure specifications.

PLoS One. 2026 Feb 3;21(2):e0340707. doi: 10.1371/journal.pone.0340707.r004

Author response to Decision Letter 2

22 Dec 2025

Comments to Reviewers

We thank the reviewers for their careful re-evaluation of our manuscript and for accepting most of our responses from the previous revision. We appreciate the constructive feedback provided in the second round of review. Below, we address the remaining comments and have revised the manuscript accordingly. We believe that these additional clarifications and modifications have further strengthened the clarity and rigor of the study.

Reviewer #1

We thank reviewer for accepting our responses and changes.

Reviewer #2

The authors have adequately addressed all of my major and minor concerns, and I recommend the paper for publication. I congratulate the authors on a well-executed study. I have one minor change request that does not require further review.

Comment 1: Figure 4 (Comment 9): I understand and agree with your concern, but my suggestion was purely graphical. Please keep the same x-axis limits to aid visual comparison, even if one curve ends earlier (the remaining region can be left blank). No additional measurements required.

Response 1: We have revised Figure 4 in accordance with the reviewer’s suggestion by modifying the x-axis to improve visual comparison. Specifically, rather than leaving the terminal time points blank, we have set the proportion to zero, as the proportion of live colonies reaches zero after 72 hours. This provides a clearer and more accurate representation of the observed dynamics across time.

Comment 2: Minor check: Please verify the reported F values around lines 301–303.

Response 2: We thank the reviewer for the comment. After verifying, we have made changes to the lines (now line 295-298 in file with track changes on):

“A two-way ANOVA revealed a significant main effect of media type (F₁,₅₄ = 42.54, p < 0.001). In contrast, the main effect of overproducer type was not significant (F₁,₁₆ = 0.00, p = 1.00). Importantly, a significant interaction between media type and overproducer type was detected (F₁,₅₄ = 15.78, p < 0.001), indicating that the effect of media on persistence depends on the overproduction classification (mutualist type).”

Reviewer #3

Comment 1: The authors have addressed all my previous comments satisfactorily.

However, in the caption of figure 5, the authors should mention the name of the statistical test that is being performed.

Response 1: We have now mentioned the name of statistical test that was performed to make figure 5. We have added the following line in the caption of figure 5 (Line 633-634 in file with track changes on):

“Statistical analyses were performed using linear mixed-effects models, with significance of fixed effects assessed using F tests (lmerTest package in R).”

Attachment

Submitted filename: Response_to_Reviewers_auresp_2.docx

pone.0340707.s007.docx^{(18KB, docx)}

PLoS One. doi: 10.1371/journal.pone.0340707.r005

Decision Letter 2

Karthik Raman

25 Dec 2025

<p>Differences in the fitness effects of traded resources shape traits and persistence in multi-mutualist communities.

PONE-D-25-49040R2

Dear Dr. Agarwal,

We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements.

Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication.

An invoice will be generated when your article is formally accepted. Please note, if your institution has a publishing partnership with PLOS and your article meets the relevant criteria, all or part of your publication costs will be covered. Please make sure your user information is up-to-date by logging into Editorial Manager at Editorial Manager® and clicking the ‘Update My Information' link at the top of the page. For questions related to billing, please contact billing support .

If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

Kind regards,

Karthik Raman, Ph.D.

Academic Editor

PLOS One

Additional Editor Comments (optional):

All corrections have been carried out.

Reviewers' comments:

PLoS One. doi: 10.1371/journal.pone.0340707.r006

Acceptance letter

Karthik Raman

PONE-D-25-49040R2

PLOS One

Dear Dr. Agarwal,

I'm pleased to inform you that your manuscript has been deemed suitable for publication in PLOS One. Congratulations! Your manuscript is now being handed over to our production team.

At this stage, our production department will prepare your paper for publication. This includes ensuring the following:

* All references, tables, and figures are properly cited

* All relevant supporting information is included in the manuscript submission,

* There are no issues that prevent the paper from being properly typeset

You will receive further instructions from the production team, including instructions on how to review your proof when it is ready. Please keep in mind that we are working through a large volume of accepted articles, so please give us a few days to review your paper and let you know the next and final steps.

Lastly, if your institution or institutions have a press office, please let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

You will receive an invoice from PLOS for your publication fee after your manuscript has reached the completed accept phase. If you receive an email requesting payment before acceptance or for any other service, this may be a phishing scheme. Learn how to identify phishing emails and protect your accounts at https://explore.plos.org/phishing.

If we can help with anything else, please email us at customercare@plos.org.

Thank you for submitting your work to PLOS ONE and supporting open access.

Kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Dr. Karthik Raman

Academic Editor

PLOS One

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

S1 Table. Yeast strains used in this study.

(DOCX)

pone.0340707.s001.docx^{(16.1KB, docx)}

S2 Table. Community of yeast mutualists used for phenotyping.

(DOCX)

pone.0340707.s002.docx^{(17.8KB, docx)}

(TIF)

pone.0340707.s003.tif^{(15.9MB, tif)}

Attachment

Submitted filename: Review_PONE-D-25-49040.docx

pone.0340707.s004.docx^{(14.1KB, docx)}

Attachment

Submitted filename: Response_to_Reviewers.docx

pone.0340707.s006.docx^{(31.7KB, docx)}

Attachment

Submitted filename: Response_to_Reviewers_auresp_2.docx

pone.0340707.s007.docx^{(18KB, docx)}

Data Availability Statement

The data underlying the results presented in this study are publicly available through Dryad at the following DOI: https://doi.org/10.5061/dryad.kwh70rzjq.

[pone.0340707.ref001] 1.Price PW. Species interactions and the. Plant animal interactions: an evolutionary approach. 2002. [Google Scholar]

[pone.0340707.ref002] 2.Bronstein JL. Mutualism. Oxford University Press; 2015. [Google Scholar]

[pone.0340707.ref003] 3.Stachowicz JJ. Mutualism, facilitation, and the structure of ecological communities: positive interactions play a critical, but underappreciated, role in ecological communities by reducing physical or biotic stresses in existing habitats and by creating new habitats on which many species depend. Bioscience. 2001;51(3):235–46. [Google Scholar]

[pone.0340707.ref004] 4.Silknetter S, Creed RP, Brown BL, Frimpong EA, Skelton J, Peoples BK. Positive biotic interactions in freshwaters: a review and research directive. Freshwater Biology. 2020;65(4):811–32. doi: 10.1111/fwb.13476 [DOI] [Google Scholar]

[pone.0340707.ref005] 5.Kost C, Patil KR, Friedman J, Garcia SL, Ralser M. Metabolic exchanges are ubiquitous in natural microbial communities. Nat Microbiol. 2023;8(12):2244–52. doi: 10.1038/s41564-023-01511-x [DOI] [PubMed] [Google Scholar]

[pone.0340707.ref006] 6.Bascompte J, Jordano P. Plant-animal mutualistic networks: the architecture of biodiversity. Annu Rev Ecol Evol Syst. 2007;38(1):567–93. [Google Scholar]

[pone.0340707.ref007] 7.Magkourilou E, Bell CA, Daniell TJ, Field KJ. The functionality of arbuscular mycorrhizal networks across scales of experimental complexity and ecological relevance. Function Ecol. 2024;39(6):1384–99. doi: 10.1111/1365-2435.14618 [DOI] [Google Scholar]

[pone.0340707.ref008] 8.Herre E, Knowlton N, Mueller U, Rehner S. The evolution of mutualisms: exploring the paths between conflict and cooperation. Trends Ecol Evol. 1999;14(2):49–53. doi: 10.1016/s0169-5347(98)01529-8 [DOI] [PubMed] [Google Scholar]

[pone.0340707.ref009] 9.Chacón JM, Hammarlund SP, Martinson JN, Smith Jr LB, Harcombe WR. The ecology and evolution of model microbial mutualisms. Ann Rev Ecol Evol System. 2021;52(1):363–84. [Google Scholar]

[pone.0340707.ref010] 10.Pauli B, Oña L, Hermann M, Kost C. Obligate mutualistic cooperation limits evolvability. Nat Commun. 2022;13(1):337. doi: 10.1038/s41467-021-27630-9 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0340707.ref011] 11.Apprill A, Marlow HQ, Martindale MQ, Rappé MS. The onset of microbial associations in the coral Pocillopora meandrina. ISME J. 2009;3(6):685–99. doi: 10.1038/ismej.2009.3 [DOI] [PubMed] [Google Scholar]

[pone.0340707.ref012] 12.Russell JA, Moreau CS, Goldman-Huertas B, Fujiwara M, Lohman DJ, Pierce NE. Bacterial gut symbionts are tightly linked with the evolution of herbivory in ants. Proc Natl Acad Sci U S A. 2009;106(50):21236–41. doi: 10.1073/pnas.0907926106 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0340707.ref013] 13.van der Heijden MGA, Klironomos JN, Ursic M, Moutoglis P, Streitwolf-Engel R, Boller T, et al. Mycorrhizal fungal diversity determines plant biodiversity, ecosystem variability and productivity. Nature. 1998;396(6706):69–72. doi: 10.1038/23932 [DOI] [Google Scholar]

[pone.0340707.ref014] 14.Afkhami ME, Rudgers JA, Stachowicz JJ. Multiple mutualist effects: conflict and synergy in multispecies mutualisms. Ecology. 2014;95(4):833–44. doi: 10.1890/13-1010.1 [DOI] [PubMed] [Google Scholar]

[pone.0340707.ref015] 15.Chomicki G, Weber M, Antonelli A, Bascompte J, Kiers ET. The impact of mutualisms on species richness. Trends Ecol Evol. 2019;34(8):698–711. doi: 10.1016/j.tree.2019.03.003 [DOI] [PubMed] [Google Scholar]

[pone.0340707.ref016] 16.Chomicki G, Kiers ET, Renner SS. The evolution of mutualistic dependence. Ann Rev Ecol Evol System. 2020;51(1):409–32. [Google Scholar]

[pone.0340707.ref017] 17.Wang RW, Shi AN, Zhang XW, Liu M, Jandér KC, Dunn DW. Asymmetric and uncertain interactions within mutualisms. J Plant Ecol. 2024;17(1):rtad042. [Google Scholar]

[pone.0340707.ref018] 18.Vidal MC, Liu C, Wang S, Segraves KA. Partner dependency alters patterns of coevolutionary selection in mutualisms. Proc Natl Acad Sci U S A. 2025;122(21):e2424983122. doi: 10.1073/pnas.2424983122 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0340707.ref019] 19.Dawkins R, Krebs JR. Arms races between and within species. Proc R Soc Lond B Biol Sci. 1979;205(1161):489–511. doi: 10.1098/rspb.1979.0081 [DOI] [PubMed] [Google Scholar]

[pone.0340707.ref020] 20.Agarwal R, Althoff DM. Extreme specificity in obligate mutualism-A role for competition?. Ecol Evol. 2024;14(6):e11628. doi: 10.1002/ece3.11628 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0340707.ref021] 21.Toby Kiers E, Palmer TM, Ives AR, Bruno JF, Bronstein JL. Mutualisms in a changing world: an evolutionary perspective. Ecol Lett. 2010;13(12):1459–74. doi: 10.1111/j.1461-0248.2010.01538.x [DOI] [PubMed] [Google Scholar]

[pone.0340707.ref022] 22.Keeler AM. Tritrophic mutualisms in a changing climate. Riverside: University of California; 2022. [Google Scholar]

[pone.0340707.ref023] 23.Magnoli SM, Keller KR, Lau JA. Mutualisms in a warming world: How increased temperatures affect the outcomes of multi-mutualist interactions. Ecology. 2023;104(3):e3955. doi: 10.1002/ecy.3955 [DOI] [PubMed] [Google Scholar]

[pone.0340707.ref024] 24.Shou W, Ram S, Vilar JMG. Synthetic cooperation in engineered yeast populations. Proc Natl Acad Sci U S A. 2007;104(6):1877–82. doi: 10.1073/pnas.0610575104 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0340707.ref025] 25.Vidal MC, Wang SP, Rivers DM, Althoff DM, Segraves KA. Species richness and redundancy promote persistence of exploited mutualisms in yeast. Science. 2020;370(6514):346–50. doi: 10.1126/science.abc1234 [DOI] [PubMed] [Google Scholar]

[pone.0340707.ref026] 26.Hart SFM, Skelding D, Waite AJ, Burton JC, Shou W. High-throughput quantification of microbial birth and death dynamics using fluorescence microscopy. Quant Biol. 2019;7(1):69–81. doi: 10.1007/s40484-018-0160-7 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0340707.ref027] 27.VanDusen WJ, Fu J, Bailey FJ, Burke CJ, Herber WK, George HA. Adenine quantitation in yeast extracts and fermentation media and its relationship to protein expression and cell growth in adenine auxotrophs of Saccharomyces cerevisiae. Biotechnol Prog. 1997;13(1):1–7. doi: 10.1021/bp9600896 [DOI] [PubMed] [Google Scholar]

[pone.0340707.ref028] 28.Kokina A, Kibilds J, Liepins J. Adenine auxotrophy--be aware: some effects of adenine auxotrophy in Saccharomyces cerevisiae strain W303-1A. FEMS Yeast Res. 2014;14(5):697–707. doi: 10.1111/1567-1364.12154 [DOI] [PubMed] [Google Scholar]

[pone.0340707.ref029] 29.Kokina A, Tanilas K, Ozolina Z, Pleiko K, Shvirksts K, Vamza I, et al. Purine Auxotrophic Starvation Evokes Phenotype Similar to Stationary Phase Cells in Budding Yeast. J Fungi (Basel). 2021;8(1):29. doi: 10.3390/jof8010029 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0340707.ref030] 30.Boer VM, Crutchfield CA, Bradley PH, Botstein D, Rabinowitz JD. Growth-limiting intracellular metabolites in yeast growing under diverse nutrient limitations. Mol Biol Cell. 2010;21(1):198–211. doi: 10.1091/mbc.e09-07-0597 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0340707.ref031] 31.Eisler H, Fröhlich K-U, Heidenreich E. Starvation for an essential amino acid induces apoptosis and oxidative stress in yeast. Exp Cell Res. 2004;300(2):345–53. doi: 10.1016/j.yexcr.2004.07.025 [DOI] [PubMed] [Google Scholar]

[pone.0340707.ref032] 32.Lewis AG, Carmichael L, Wang RY, Gibney PA. Characterizing a panel of amino acid auxotrophs under auxotrophic starvation conditions. Yeast. 2024;41(1–2):5–18. doi: 10.1002/yea.3910 [DOI] [PubMed] [Google Scholar]

[pone.0340707.ref033] 33.Bates D, Mächler M, Bolker B, Walker S. Fitting linear mixed-effects models Usinglme4. J Stat Soft. 2015;67(1). doi: 10.18637/jss.v067.i01 [DOI] [Google Scholar]

[pone.0340707.ref034] 34.Johnson NC. Can Fertilization of Soil Select Less Mutualistic Mycorrhizae?. Ecol Appl. 1993;3(4):749–57. doi: 10.2307/1942106 [DOI] [PubMed] [Google Scholar]

[pone.0340707.ref035] 35.Hoeksema JD, Chaudhary VB, Gehring CA, Johnson NC, Karst J, Koide RT, et al. A meta-analysis of context-dependency in plant response to inoculation with mycorrhizal fungi. Ecol Lett. 2010;13(3):394–407. doi: 10.1111/j.1461-0248.2009.01430.x [DOI] [PubMed] [Google Scholar]

[pone.0340707.ref036] 36.Regus JU, Wendlandt CE, Bantay RM, Gano-Cohen KA, Gleason NJ, Hollowell AC, et al. Nitrogen deposition decreases the benefits of symbiosis in a native legume. Plant Soil. 2016;414(1–2):159–70. doi: 10.1007/s11104-016-3114-8 [DOI] [Google Scholar]

[pone.0340707.ref037] 37.Foster KR, Wenseleers T. A general model for the evolution of mutualisms. J Evol Biol. 2006;19(4):1283–93. doi: 10.1111/j.1420-9101.2005.01073.x [DOI] [PubMed] [Google Scholar]

[pone.0340707.ref038] 38.Harcombe W. Novel cooperation experimentally evolved between species. Evolution. 2010;64(7):2166–72. [DOI] [PubMed] [Google Scholar]

[pone.0340707.ref039] 39.Bachmann H, Bruggeman FJ, Molenaar D, Branco Dos Santos F, Teusink B. Public goods and metabolic strategies. Curr Opin Microbiol. 2016;31:109–15. doi: 10.1016/j.mib.2016.03.007 [DOI] [PubMed] [Google Scholar]

[pone.0340707.ref040] 40.Frank SA. The trade-off between rate and yield in the design of microbial metabolism. J Evol Biol. 2010;23(3):609–13. doi: 10.1111/j.1420-9101.2010.01930.x [DOI] [PubMed] [Google Scholar]

[pone.0340707.ref041] 41.Wang S, Agarwal R, Segraves KA, Althoff DM. Trait and plasticity evolution under competition and mutualism in evolving pairwise yeast communities. PLoS One. 2025;20(1):e0311674. doi: 10.1371/journal.pone.0311674 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0340707.ref042] 42.Lenski RE, Rose MR, Simpson SC, Tadler SC. Long-term experimental evolution in Escherichia coli. I. Adaptation and divergence during 2,000 generations. The American Naturalist. 1991;138(6):1315–41. [Google Scholar]

[pone.0340707.ref043] 43.Dragosits M, Mattanovich D. Adaptive laboratory evolution -- principles and applications for biotechnology. Microb Cell Fact. 2013;12:64. doi: 10.1186/1475-2859-12-64 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0340707.ref044] 44.Gerrish PJ, Lenski RE. The fate of competing beneficial mutations in an asexual population. Genetica. 1998;102–103(1–6):127–44. doi: 10.1023/a:1017067816551 [DOI] [PubMed] [Google Scholar]

[pone.0340707.ref045] 45.Sepp S-K, Davison J, Jairus T, Vasar M, Moora M, Zobel M, et al. Non-random association patterns in a plant-mycorrhizal fungal network reveal host-symbiont specificity. Mol Ecol. 2019;28(2):365–78. doi: 10.1111/mec.14924 [DOI] [PubMed] [Google Scholar]

[pone.0340707.ref046] 46.Tedersoo L, Bahram M, Zobel M. How mycorrhizal associations drive plant population and community biology. Science. 2020;367(6480):eaba1223. doi: 10.1126/science.aba1223 [DOI] [PubMed] [Google Scholar]

PERMALINK

Differences in the fitness effects of traded resources shape traits and persistence in multi-mutualist communities

Renuka Agarwal

Anne E Curé

Kari A Segraves

David M Althoff

Roles

Abstract

Introduction

Materials and methods

Synthetic model of multi-mutualist communities

Setting up the communities and evolving the strains

Fig 1. Overview of the Saccharomyces cerevisiae multi-mutualist system, strain types, and experimental environments.

Assessing trait evolution

Fig 3. Individual strain persistence over time of adenine overproducers (AdeOP, left panel) and lysine overproducers (LysOP, right panel) of Saccharomyces cerevisiae.

Measuring and analyzing strain persistence

Starvation resistance measurement

Overproduction measurements

Resource use efficiency and growth rate measurements

Results

Media and mutualist type jointly determine strain persistence

Fig 2. Plot of proportion of evolving multi-mutualist communities that retained strains of adenine overproducers (AdeOP) and lysine overproducers (LysOP) of Saccharomyces cerevisiae.

Starvation resistance remained unchanged across evolutionary environments

Fig 4. Starvation resistance of ancestral and evolved Saccharomyces cerevisiae mutualist types over time.

Resource availability increased lysine, but not adenine, overproduction

Fig 5. Measurement of trait evolution.

Reduced efficiency evolved without growth rate compensation

Discussion

Supporting information

Acknowledgments

Data Availability

Funding Statement

References

Decision Letter 0

Karthik Raman

Roles

Author response to Decision Letter 1

Decision Letter 1

Karthik Raman

Roles

Author response to Decision Letter 2

Decision Letter 2

Karthik Raman

Roles

Acceptance letter

Karthik Raman

Roles

Associated Data

Supplementary Materials

Data Availability Statement

ACTIONS

PERMALINK

RESOURCES

Similar articles

Cited by other articles

Links to NCBI Databases