Abstract
Adaptation to an environment is enabled by the accumulation of beneficial mutations. How do adaptive trajectories and pleiotropic effects of adaptation change in response to “subtle” changes in the environment? Since there exists no molecular framework to quantify “subtle” environmental change, designing experiments to answer this question has been challenging. In this work, we address this question by studying the effects of evolution in environments which differ solely in the way sugars are presented to a bacterial population. Specifically, we focus on glucose and galactose, which can be supplied to an E. coli population as a mixture of glucose and galactose, lactose, or melibiose. We evolve six replicate populations of E coli for 300 generations in these three chemically correlated or “synonymous” environments, and show that the adaptive responses of these populations are not similar. When tested for pleiotropic effects of fitness in a range of non-synonymous environments, our results show that despite uncorrelated adaptive changes, the nature of pleiotropic effects is largely predictable based on the fitness of the ancestor in the non-home environments. Overall, our results highlight how subtle changes in the environment can alter adaptation, but despite sequence-level variations, pleiotropy is qualitatively predictable.
Similar content being viewed by others
Background
An evolving population accumulates mutations, and natural selection increases the frequency of beneficial ones1. The environment in which evolution occurs can influence both the mutation rate and the types of mutations that become fixed2. The effects of mutations depend not only on the environment but also on the genetic background in which they arise3,4. While similar environments often lead to correlated adaptive responses, there are notable exceptions5. In fact, several studies suggest that adaptation in similar environments is, to some extent, predictable6,7,8. However, designing experimental systems where environmental similarity can be precisely quantified to understand adaptive responses remains challenging.
One approach to measuring environmental similarity is by examining differences in the adaptive responses they induce, under the assumption that similar environments exert similar selection pressures9. Phenotypic parallelism may reflect such environmental similarity, though it is important to recognize that similar phenotypic outcomes can arise from different genetic changes. This makes it difficult to predict pleiotropic effects in other environments10.
Another approach is to compare the pleiotropic outcomes of adaptation itself. For example, if populations evolved in environments A and B show similar performance in a third environment C, this may suggest that environments A and B are similar9,11,12. A number of laboratory studies have explored the dynamics and variability of pleiotropic effects of adaptation9,11,13,14,15,16,17,18,19,20,21,22. While generalist evolution is common, the emergence of specialists has also been observed, highlighting that fitness effects in non-home environments can vary widely.
Both methods for inferring environmental similarity—via adaptive response or pleiotropic performance—have limitations. They often rely on fitness data from laboratory experiments that suffer from a form of circularity: similarity in adaptive effects does not always imply similar pleiotropic outcomes, and vice versa.
To address these challenges, we developed an experimental system in which environmental similarity is defined chemically and metabolically23, rather than inferred from adaptive or pleiotropic outcomes. We focused on three carbon environments that differ only in how glucose and galactose are presented to E. coli populations: (a) a mixture of glucose and galactose, (b) lactose, and (c) melibiose. The disaccharides lactose and melibiose are formed by linking glucose and galactose via β-1,4-glycosidic and α-1,6-glucosidic bonds, respectively. As shown in Fig. 1, E. coli processes both sugars through similar pathways—import via permeases followed by hydrolysis into their monosaccharide components24,25. Due to their shared chemical structure and metabolic processing, we refer to these environments as “synonymous.” This chemically grounded classification allows us to examine how subtle environmental differences impact evolutionary outcomes without relying on circular, outcome-based definitions of similarity.
a The lac operon includes lacZ (β-galactosidase) and lacY (lactose permease), whose expression is repressed by the LacI repressor protein in the absence of lactose. When lactose (white square linked with white circle) is present, it binds LacI, relieving repression and enabling transcription from the lacZY promoter. lacY-encoded permease facilitates lactose import, creating a positive feedback loop. Hydrolysis of lactose by LacZ releases glucose (white square) and galactose (white circle). LacI is also an autorepressor. b The mel operon includes melA (α-galactosidase) and melB (melibiose transporter), regulated by the activator MelR. In the presence of melibiose (rotated white square linked with white circle), MelR activates transcription of melA and melB, with MelB-mediated import enhancing inducer availability, forming a similar positive feedback loop as in the case of lactose. MelR is an auto-repressor. Hydrolysis of melibiose by MelB releases glucose (white square) and galactose (white circle).
Using this system, we evolved six replicate populations in each of the three synonymous (or, chemically correlated) environments for 300 generations. We then measured their fitness across both the synonymous environments and a panel of non-synonymous sugar environments using short-term growth assays. Despite the biochemical similarity of the environments, the adaptive responses of the populations diverged significantly, indicating that even minimal environmental differences can lead to distinct evolutionary trajectories. However, when we examined pleiotropic effects in non-home environments, a different pattern emerged: pleiotropic responses were more consistent and, in most cases, predictable. Moreover, we found a correlation between the number of distinct mutational targets and the proportion of unique pleiotropic outcomes, suggesting a mechanistic link between genetic divergence and pleiotropic variability.
While our study, like many others, relies on temporally limited fitness assays, it offers a framework to chemically define environmental similarity independent of evolutionary outcomes. By decoupling the definition of environment from the consequences of adaptation, our work provides a more rigorous foundation for studying how environmental differences shape evolutionary dynamics, and highlights the complex, sometimes decoupled relationship between adaptation and pleiotropy.
Results
Adaptation in synonymous environments yields distinct outcomes
Replicate populations of E. coli were evolved in M9 minimal media containing either 0.2% of glucose-galactose (0.1% of each), lactose, or melibiose. After 300 generations, we measured two fitness parameters – growth rate in the exponential phase and yield when bacterial cells enter stationary phase.
As shown in Fig. 2, either the growth rate (r), yield (K), or both, increased when compared to the ancestor (p < 0.00833, one-tailed t-test, Bonferroni correction applied for six comparisons) in the eighteen populations after evolution in one of the three environments for 300 generations (Supplementary Table 1). Upon comparing the average r and K of the three sets of evolved populations with the ancestor, we see that evolution in melibiose results in a significant increase in yield (p < 0.001, one-tailed t-test), but not in growth rate (p > 0.05, one-tailed t-test). On the other hand, evolution in lactose and glucose-galactose resulted in an increase in both average growth rate (p < 0.001, one-tailed t-test), and yield (p < 0.001, one-tailed t-test). These findings show that the selection pressure imposed by chemically correlated environments in identical evolution experiments is dissimilar, and as a result, adaptive responses are divergent.
Six replicate populations of E. coli, founded from a single clone, were evolved in three different, yet synonymous sugar environments – glucose-galactose mixture, lactose, and melibiose. After 300 generations of evolution, growth assays were performed in triplicate for each of the evolved populations and the ancestor to measure growth rate(r) and yield (K). Triangular data points show r and K for each of the replicate populations, and the error bars correspond to the standard deviations obtained from technical replicates. The square blocks show the average r and K for the evolved populations in each of the environments (obtained as the average of the mean of the six replicates), and the error bars correspond to the standard deviation (obtained as the standard deviation of the mean of the six replicates). The circular blocks show the ancestor’s average r and K in each of the evolution environments, and the error bars correspond to the standard deviation obtained from technical replicated. For all the eighteen evolved populations, changes in r and K depended on the exact composition on the evolution environment. Clearly, selection acting on a population changes even with a minor change in the environment.
Populations with an increased growth rate (r) are fast growers, while populations with an increased yield (K) are considered better competitors26. While intuition suggests that experimentally evolved populations should exhibit r-K trade-offs, there exists very little empirical evidence in support. Marshall et. al. showed theoretically how and why r and K may covary positively27, and Wei et. al. show that reduction in environment quality leads to such “trade-ups” in r and K28. In our study, although trade-ups in r and K are common in the evolved populations, the average growth rate of melibiose-evolved populations does not increase compared to the ancestor, indicating that the likelihood of occurrences of trade-ups is dependent on the exact chemical nature of the evolution environment.
Pleiotropic effects of adaptation in synonymous environments
To further understand the pleiotropic fitness effects, we shift these three sets of evolved populations to non-home environments, that contain synonymous sugars as the source of carbon. We calculate the growth r, and K at the end of 16 h, assuming exponential growth. Relative fitness changes [\(\frac{{f}_{{evolved}}-{f}_{{ancestor}}\,}{{f}_{{ancestor}}}\), where f is r, or K] of twelve populations, relative to the ancestor, are calculated in home and away environments, and these values are distributed as shown in Fig. 3a, b.
We tested the fitness (r and K) of the three sets of evolved populations in away environments, in which the source of carbon was a synonymous sugar, and checked if fitness gains in home and away environments were correlated. For each set of evolved population, there were two synonymous away environments in our experiment. (a, b) show the relative gains in home and away environments, in r and K, respectively. The error bars indicate the propagated uncertainties. We count the number of distinct types of pleiotropic effects. As shown in (c) (growth rate) and (d) (yield), the relative frequencies of the three types of pleiotropic responses are different for different sets of evolved populations.
Next, the fitness gains for each population relative to the ancestor, in home and non-home environments were compared, using one-tailed t-tests (p values are shown in Supplementary Table 2). Each such comparison between relative fitness change in non-home and home environments can be classified into one of the following three categories: (a) identical – the magnitude and sign of fitness effect are statistically identical, (b) magnitude change – the magnitudes of fitness effects are statistically non-identical, and (c) sign change – the magnitude and sign of fitness effects are different in home and non-home environments. Figure 3c, d shows the proportion of these cases (out of 12), for r and K, for the three sets of evolved populations, when grown in non-home synonymous environments.
In the lactose-evolved populations, fitness changes in the away environments were identical to the changes in the home environment. Correlated fitness gains are the most common in the lactose-evolved populations, followed by the glucose-galactose-evolved populations. The melibiose-evolved populations, showed the most diverse range of pleiotropic responses.
By and large, our results agree with the observations made by Ostrowski et al. that adaptation in a sugar environment confers positive pleiotropic effects in other sugars18. However, despite being shifted to synonymous environments, our evolved populations show considerable variability in their pleiotropic responses. What happens to variability in pleiotropic responses in non-synonymous sugars?
Pleiotropic effects of adaptation in non-synonymous environments
The three sets of populations which evolved independently in melibiose, lactose, and glucose-galactose were shifted to non-synonymous sugar environments and tested for fitness. These non-synonymous environments consisted of pentoses (arabinose, xylose), a methyl pentose (rhamnose), sugar alcohols (glycerol, sorbitol), or hexoses (fructose, mannose) as the source of carbon. Again, we measured r in exponential phase and K as cells enter stationary phase. Fitness changes of the eighteen evolved populations, relative to that of the ancestor, were calculated in home and non-home environments, and are as shown in Fig. 4a, b (See Supplementary Fig. 1 for exact fitness changes in each environment). A one-tailed t-test was used to compare the fitness changes between home and non-home environments (p-values are shown in Supplementary Table 2).
We tested the fitness (r and K) of the three sets of evolved populations in away environments, in which the source of carbon was a non-synonymous sugar, and checked if fitness gains in home and away environments were correlated. For each set of evolved population, there were eight non-synonymous away environments in our experiment. (a, b) show the relative gains in home and away environments, in r and K, respectively. We count the number of distinct types of pleiotropic effects, and the results are as shown in (c) (growth rate) and (d) (yield).
We found no environment in which the fitness of all the evolved populations dropped compared to the ancestor. Arabinose and sorbitol were two environments in which both r and K of all the evolved populations increased relative to the ancestor. Therefore, all the populations that evolved in one of the synonymous environments were fit in arabinose and sorbitol. However, we did not find any set of evolved populations in which all the replicates performed identically in home and any non-home environments. As a result, given a non-home environment, we could only say if an evolved population is fitter than the ancestor or not; we were unable to comment on the exact difference in fitness based on the fitness of the evolved population in the home environment.
In the seven non-synonymous sources of carbon, lactose evolved populations showed the maximum number of similar fitness changes, followed by the glucose-galactose evolved populations. The melibiose-evolved populations continued to show diverse responses in non-synonymous environments as well, as shown in Fig. 4c, d.
Figure 5 shows the overall proportion of cases (considering both r and K, and synonymous and non-synonymous environments) of pleiotropy for the three sets of evolved populations. It is evident that the variability in the type of pleiotropic responses depends on their evolution environment.
To test whether the pleiotropic fitness changes (both r and K) of glucose-galactose, lactose, and melibiose evolved populations in away environments (both synonymous and non-synonymous) follow any patterns, we compared the relative frequencies of the three types of pleiotropic responses exhibited by the three sets of evolved populations. Melibiose evolved populations exhibited the greatest number of pleiotropic responses, followed by the lactose evolved populations. Overall, we observe that even minute changes in the evolution environment play a significant role in dictating the variability in pleiotropic responses.
Fitness of the ancestor as a predictor of pleiotropic effects
Global epistasis has enabled the prediction of the fitness effect of a beneficial mutation as a function of the fitness of the background on which it occurs29. We tried to identify if ancestral fitness can help predict pleiotropic effects, as reported in the past12. Specifically, we checked if the fitness of the ancestor can be used as a predictor of the pleiotropic fitness of an evolved population. Therefore, we compared the fitness changes in the evolved populations with that of the ancestor, in both synonymous and non-synonymous environments.
In synonymous environments, there existed no consistent correlation between ancestral fitness and the nature of pleiotropic responses, as shown in Fig. 6a, b. The three sets of evolved populations showed qualitatively similar pleiotropic responses in the non-synonymous environments (refer to Supplementary Fig. 2 and Supplementary Fig. 3). As shown in Fig. 6c, d, in non-synonymous environments, average gains in r and K correlated negatively with ancestral fitness.
We test if the fitness of the ancestral population in an environment can act as a predictor of the fitness of an evolved population. We compare the fitness gain of an evolved population with ancestral fitness in away environments. (a, b) show these comparisons for r and K in synonymous environments, respectively. It must be noted that for each set of evolved populations, there were two synonymous away environments. We also plot relative fitness gains against ancestral fitness in eight non-synonymous environments. As shown in (c) (for growth rate) and (d) (yield), in most cases, there exists a trend – in non-synonymous environments, average pleiotropic fitness gains decrease with an increase in the fitness of the ancestor. However, there exists no such quantitative trend in the case of synonymous environments.
We next check if the variability in pleiotropic responses of these evolved populations is predictable based on ancestor’s fitness in the non-home environment. To do so, we consider the variance of fitness gains (in r and K), as a proxy for variability in pleiotropic responses. As shown by the Spearman’s rank correlations in Supplementary Table 3, as the fitness of the ancestor increases in the non-home environment, the variability in pleiotropic responses decreases. Therefore, given the fitness of the ancestor in a non-home environment, we show that a qualitative prediction of pleiotropic effects of an evolved population is possible.
Melibiose-evolved populations are the most genetically diverse
We genome sequenced the eighteen evolved populations to identify the genetic changes accumulated in the three synonymous evolution environments. Mutation analyses reveal that mutations in the rpo genes were the most common among all sets of evolved populations. Rpo genes, that code for proteins that make up RNA polymerase, are known to facilitate adaptation in not only thermal, osmotic, and antibiotic stresses but also to M9 media30. Other mutational targets included yafW which is a part of a toxin-antitoxin pair, nanR and acrR, which are transcription regulators31,32, and the methionine transporter metN33. Overall, the melibiose-evolved populations had the highest number of distinct mutational targets, as shown in Fig. 7a (Supplementary Table 4 lists the mutations in each population). Figure 7b shows the correlation between the proportion of pleiotropic responses and the number of mutational targets.
a The mutational targets in the three sets of evolved populations are shown. b The proportion of pleiotropic responses is correlated with the number of distinct mutational targets in the evolved populations. The number of mutational targets is high in the melibiose-evolved populations, and they show the highest number of pleiotropic responses.
We counted the number of instances where the three sets of evolved populations showed any type of pleiotropy (change in magnitude and/or sign of fitness effect), across the ten non-home environments (two synonymous, eight non-synonymous) their fitness was measured in. Considering changes in both r and K, we see that pleiotropic effects are the maximum for the melibiose-evolved populations, as shown in Fig. 7b. This proportion of pleiotropic responses, along with that of lactose-evolved and glucose-galactose evolved populations, correlated predictably with the number of distinct genes in which adaptive mutations occurred in replicate populations.
Discussion
Adaptation in similar environments has often been correlated, offering insights into the predictability of evolution34,35. However, the pleiotropic consequences of adaptation can vary depending on evolutionary timescales11,13,15,19,20. Long-term evolution experiments have shown that prolonged selection in a particular environment can lead to specialization11. Despite this, the role of environment in shaping adaptation and determining whether generalists or specialists evolve remains poorly understood. A central challenge has been the lack of a method to quantify the similarity of environments, making predictions of adaptation and pleiotropy difficult.
In this work, we address that limitation using a novel framework. Instead of relying on phenotypic similarity to infer environmental relatedness—a method confounded by the unpredictability of pleiotropy—we designed environments that are chemically correlated. Specifically, we presented E. coli populations with two monosaccharides (glucose and galactose) packaged in three different ways: (a) a mixture of the two, (b) lactose, and (c) melibiose. Lactose and melibiose are disaccharides formed by chemically linking glucose and galactose via β-1,4 and α-1,6 glycosidic bonds, respectively. These “synonymous” sugar environments provide a controlled, chemically quantifiable perturbation.
To probe how such minute chemical differences shape adaptation and pleiotropy, we evolved six replicate E. coli populations in each of the three environments for 300 generations—capturing the early onset of adaptation. Notably, all evolution environments supported ancestral growth, avoiding confounding effects of harsh selection. This setup enabled us to isolate the impact of subtle environmental differences on evolutionary outcomes.
Our results show that evolution in chemically synonymous environments does not lead to correlated adaptive outcomes. Despite similar uptake and hydrolysis mechanisms for lactose and melibiose36,37, the selective pressures they impose differ substantially. In melibiose, selection strongly favored biomass accumulation, sometimes even at a small growth rate cost. In two melibiose-evolved populations, growth rate declined (albeit with low statistical significance)—one of the few empirical demonstrations of an r/K trade-off20,28,38,39,40,41,42. These findings challenge the assumption that cells always prioritize minimizing division time, showing instead that the trait under selection depends on environmental context.
Fluctuating environments are known to drive genetic diversity43,44,45, and act via global regulators46,47,48. We show that even stable, chemically similar environments can lead to divergent genotypic and phenotypic outcomes, positioning subtle environmental differences as a driver of microbial diversity.
We hypothesized that evolution in synonymous environments would yield positive pleiotropic effects in each other’s contexts21. Fitness assays supported this broadly: positive pleiotropic gains were common across both synonymous and non-synonymous environments. Surprisingly, in some cases, evolved populations performed better in a non-home environment than in their own. For instance, two melibiose-evolved populations had reduced home fitness but increased growth in nearly all other environments. In fact, populations adapted in environment B sometimes outperformed all A-evolved populations in environment A. This suggests that foreign environments may unlock fitness peaks inaccessible through local adaptation, echoing ideas about epistasis and adaptation roadblocks49. While nearly all melibiose-evolved populations gained fitness in non-home environments, the mutations they acquired did not fix in populations evolved in other synonymous environments. This supports the idea that subtle environmental cues can reshape the adaptive landscape, either enabling or blocking specific evolutionary paths.
Interestingly, generalists were more common than specialists even after just 300 generations. To investigate whether ancestral fitness predicts pleiotropic effects, we analyzed correlations. In synonymous environments, no such correlation was found—suggesting early pleiotropic outcomes are not constrained by ancestral fitness. But in non-synonymous environments, a clear pattern emerged: ancestral fitness did predict the qualitative pleiotropic outcome, even amid genetic stochasticity. Moreover, the frequency of pleiotropic responses correlated with the number of distinct mutational targets9, reinforcing the need for sequence-level mapping of adaptation using high-throughput experiments10. For instance, a deep-mutational scan of the rpo genes, which are often implicated in nutrient-limiting evolution experiments and shown to have diverse pleiotropic responses10, may reveal how correlated (or not) the landscape of adaptation and pleiotropy is.
Most prior studies of adaptation and pleiotropy focused on different types of stressful conditions. In contrast, our study shows how related environments with minimal stress can generate both divergent adaptive responses and predictable pleiotropic effects. While our current analysis is limited to short-term fitness measurements and chemically defined environments, this work sets the stage for exploring more complex environmental differences—including temperature or osmotic stress—in future studies.
We also continue to propagate these populations to investigate long-term outcomes, such as whether early generalists transition to specialists over time. Ultimately, our findings highlight the importance of environmental subtlety in shaping evolutionary trajectories and motivate the development of high-throughput evolution systems to systematically map pleiotropic effects across diverse conditions.
Material and methods
Strains used and media composition
E. coli K12 MG1655 (ATCC 47076) was used as the founder in this study.
Evolution experiments
The evolution experiment was conducted in M9 minimal media (composition: Na2HPO4.7H2O (12.8 g/L), KH2PO4 (3 g/L), NaCl (0.5 g/L), NH4Cl (1 g/L), 1 M MgSO4 and 0.1 M CaCl2), containing either a mixture of glucose and galactose (0.1% each) or lactose (0.1%) or melibiose (0.1%).
Six independent replicate populations of E. coli were evolved at 37 °C and at 250 rpm in three environments - glucose-galactose mixture, lactose, or melibiose. Evolution experiment was performed by serially transferring 50 μL of culture into 4.95 mL of fresh M9 media containing appropriate sugar, every 12 h.
Fitness measurements
To measure fitness, cells were grown in triplicate in LB (1% tryptone, 0.5% yeast extract, 0.05% NaCl) and incubated at 37 °C for 12 h with shaking at 250 rpm. Thereafter, cells were sub-cultured 1:100 in fresh M9 media containing 0.2% glycerol for 18 h. To measure fitness, cells were cultured in M9 media containing the appropriate carbon environment (glucose-galactose, lactose, melibiose, L-arabinose, D-xylose, D-mannose, D-fructose, D-sorbitol) to an initial OD600 of 0.01.
Cells were then transferred to a 96-well plate (Costar) and the plate was covered with a breathable membrane (Breathe-Easy) to prevent evaporation. The 96-well plate was incubated at 37 °C with shaking in a microplate reader (Tecan Infinite Pro 200). OD600 was measured every 30 minutes until each culture reached stationary phase (~20 h).
Growth rate was measured as described in the past50. Maximum OD attained in each line was used as a measure for biomass accumulation. There was no observable lag phase in the glucose-galactose and lactose-evolved populations, as observed in another study50. Therefore, we did not consider it as a measure of fitness change.
Statistical tests
The error bars in the plots shown correspond to standard deviation, unless specified otherwise. A comparison of two means was done using a two-tailed t test. A one-tailed t test was used to ascertain if one mean is greater than or lesser than another. In all cases, the significance level was set to 0.05.
Whole-genome sequencing
Single clones of ancestor and evolved E. coli populations were used to isolate genomic DNA using the standard phenol/chloroform DNA isolation protocol51. DNA quality and concentration were measured using Nanodrop Spectrophotometer (Eppendorf), and band size was confirmed through gel electrophoresis. Genome sequencing was carried out using Illumina NovaSeq 6000 platform with paired-end reads of 150 base pairs, and an average read depth of 100×.
Raw sequencing reads were assessed for quality using FastQC (v- 0.11.9)52, and fastq mcf (v-1.04.803) was used to trim adapters and remove low-quality bases, retaining reads with a minimum Phred score ≥20. The filtered high-quality reads were aligned to E. coli K-12 MG1655 reference genome (ATCC 47076) using Burrows Wheeler Aligner (BWA) tool (v- 0.7.17)53 with default parameters (bwa mem -t 8). SAM files were converted to BAM, sorted, and indexed using SAMtools (v- 1.21)54.
Variant calling was performed using GATK HaplotypeCaller (v- 4.6.1.0)55, with default settings optimized for haploid bacterial genomes (--ploidy 1). For downstream analysis, only variants with a minimum read depth of 30 were considered to filter out variants. Variants identified in the ancestral population were filtered out manually using custom R scripts. The remaining unique variants were annotated using SnpEff (v- 5.1)56 with the E. coli K-12 MG1655 genome database. Raw sequencing data is available at ncbi.nlm.nih.gov/sra/PRJNA1022868.
Data availability
Genome sequencing is available at ncbi.nlm.nih.gov/sra/PRJNA1022868.
References
Gregory, T. R. Understanding Natural Selection: Essential Concepts and Common Misconceptions. Evol. Educ. Outreach 2, 156–175 (2009).
Carja, O., Liberman, U. & Feldman, M. W. Evolution in changing environments: Modifiers of mutation, recombination, and migration. Proc. Natl Acad. Sci. 111, 17935–17940 (2014).
Bateson, W., Mendel, G. & Leighton, A. G. Mendel’s Principles of Heredity, by W. Bateson. (University Press, Cambridge [Eng.], https://doi.org/10.5962/bhl.title.1057 1909).
Plate, L. Genetics and Evolution, Pp. 536–610 in Festschrift Zum Sechzigsten Geburtstag Richard Hertwigs. Fischer, Jena, Germany (in German). (1910).
Blount, Z. D., Borland, C. Z. & Lenski, R. E. Historical contingency and the evolution of a key innovation in an experimental population of Escherichia coli. Proc. Natl Acad. Sci. Usa. 105, 7899–7906 (2008).
Langerhans, R. B. & DeWitt, T. J. Shared and unique features of evolutionary diversification. Am. Nat. 164, 335–349 (2004).
Turner, C. B., Marshall, C. W. & Cooper, V. S. Parallel genetic adaptation across environments differing in mode of growth or resource availability. Evol. Lett. 2, 355–367 (2018).
Bailey, S. F., Rodrigue, N. & Kassen, R. The Effect of Selection Environment on the Probability of Parallel Evolution. Mol. Biol. Evol. 32, 1436–1448 (2015).
Kinsler, G., Geiler-Samerotte, K. & Petrov, D. A. Fitness variation across subtle environmental perturbations reveals local modularity and global pleiotropy of adaptation. eLife 9, e61271 (2020).
Venkataraman, P., Nagendra, P., Ahlawat, N., Brajesh, R. G. & Saini, S. Convergent genetic adaptation of Escherichia coli in minimal media leads to pleiotropic divergence. Front. Mol. Biosci. 11, 1286824 (2024).
Bakerlee, C. W., Phillips, A. M., Nguyen Ba, A. N. & Desai, M. M. Dynamics and variability in the pleiotropic effects of adaptation in laboratory budding yeast populations. eLife 10, e70918 (2021).
Jerison, E. R. et al. Genetic variation in adaptability and pleiotropy in budding yeast. eLife 6, e27167 (2017).
Cooper, V. S. & Lenski, R. E. The population genetics of ecological specialization in evolving Escherichia coli populations. Nature 407, 736–739 (2000).
Dillon, M. M., Rouillard, N. P., Van Dam, B., Gallet, R. & Cooper, V. S. Diverse phenotypic and genetic responses to short-term selection in evolving Escherichia coli populations: DIVERSE RESPONSES TO SELECTION OF E. coli POPULATIONS. Evolution 70, 586–599 (2016).
Bailey, S. F. & Kassen, R. Spatial Structure of Ecological Opportunity Drives Adaptation in a Bacterium. Am. Nat. 180, 270–283 (2012).
Leiby, N. & Marx, C. J. Metabolic Erosion Primarily Through Mutation Accumulation, and Not Tradeoffs, Drives Limited Evolution of Substrate Specificity in Escherichia coli. PLoS Biol. 12, e1001789 (2014).
Jasmin, J.-N., Dillon, M. M. & Zeyl, C. The yield of experimental yeast populations declines during selection. Proc. R. Soc. B Biol. Sci. 279, 4382–4388 (2012).
Ostrowski, E. A., Rozen, D. E. & Lenski, R. E. Pleiotropic effects of beneficial mutations in escherichia coli. Evolution 59, 2343–2352 (2005).
Meyer, J. R. et al. Parallel changes in host resistance to viral infection during 45,000 generations of relaxed selection: evolution under relaxed selection for resistance. Evolution no-no https://doi.org/10.1111/j.1558-5646.2010.01049.x (2010).
Novak, M., Pfeiffer, T., Lenski, R. E., Sauer, U. & Bonhoeffer, S. Experimental Tests for an Evolutionary Trade-Off between Growth Rate and Yield in E. coli. Am. Nat. 168, 242–251 (2006).
Jerison, E. R., Nguyen Ba, A. N., Desai, M. M. & Kryazhimskiy, S. Chance and necessity in the pleiotropic consequences of adaptation for budding yeast. Nat. Ecol. Evol. 4, 601–611 (2020).
Travisano, M. & Lenski, R. E. Long-term experimental evolution in Escherichia coli. IV. Targets of selection and the specificity of adaptation. Genetics 143, 15–26 (1996).
Ahlawat, N., Mahilkar, A. & Saini, S. Resource presentation dictates genetic and phenotypic adaptation in yeast. BMC Ecol. Evol. 25, 33 (2025).
Webster, C., Kempsell, K., Booth, L. & Busby, S. Organisation of the regulatory region of the Escherichia coli melibiose operon. Gene 59, 253–263 (1987).
Sanders, R. E. The Lactose Operon of Escherichia coli. in Bacteria, Bacteriophages, and Fungi (ed. King, R. C.) 163–168 (Springer US, Boston, MA, https://doi.org/10.1007/978-1-4899-1710-2_8 1974).
Pianka, E. R. On r- and K-Selection. Am. Nat. 104, 592–597 (1970).
Marshall, D. J., Cameron, H. E. & Loreau, M. Relationships between intrinsic population growth rate, carrying capacity and metabolism in microbial populations. ISME J. 17, 2140–2143 (2023).
Wei, X. & Zhang, J. Environment-dependent pleiotropic effects of mutations on the maximum growth rate r and carrying capacity K of population growth. PLOS Biol. 17, e3000121 (2019).
Kryazhimskiy, S., Rice, D. P., Jerison, E. R. & Desai, M. M. Microbial evolution. Global epistasis makes adaptation predictable despite sequence-level stochasticity. Science 344, 1519–1522 (2014).
Hews, C. L., Cho, T., Rowley, G. & Raivio, T. L. Maintaining Integrity Under Stress: Envelope Stress Response Regulation of Pathogenesis in Gram-Negative Bacteria. Front. Cell. Infect. Microbiol. 9, 313 (2019).
Chu, D., Roobol, J. & Blomfield, I. C. A theoretical interpretation of the transient sialic acid toxicity of a nanR mutant of Escherichia coli. J. Mol. Biol. 375, 875–889 (2008).
Amores, G. R., de Las Heras, A., Sanches-Medeiros, A., Elfick, A. & Silva-Rocha, R. Systematic identification of novel regulatory interactions controlling biofilm formation in the bacterium Escherichia coli. Sci. Rep. 7, 16768 (2017).
Kadaba, N. S., Kaiser, J. T., Johnson, E., Lee, A. & Rees, D. C. The high-affinity E. coli methionine ABC transporter: structure and allosteric regulation. Science 321, 250–253 (2008).
Anderson, S. A. S. & Weir, J. T. The role of divergent ecological adaptation during allopatric speciation in vertebrates. Science 378, 1214–1218 (2022).
Fong, S. S., Joyce, A. R. & Palsson, B. Ø Parallel adaptive evolution cultures of Escherichia coli lead to convergent growth phenotypes with different gene expression states. Genome Res. 15, 1365–1372 (2005).
Jacob, F. & Monod, J. Genetic regulatory mechanisms in the synthesis of proteins. J. Mol. Biol. 3, 318–356 (1961).
Pardee, A. B. An inducible mechanism for accumulation of melibiose in Escherichia coli. J. Bacteriol. 73, 376–385 (1957).
Reding-Roman, C. et al. The unconstrained evolution of fast and efficient antibiotic-resistant bacterial genomes. Nat. Ecol. Evol. 1, 0050 (2017).
Beardmore, R. E., Gudelj, I., Lipson, D. A. & Hurst, L. D. Metabolic trade-offs and the maintenance of the fittest and the flattest. Nature 472, 342–346 (2011).
Warringer, J., et al. Trait variation in yeast is defined by population history. PLoS Genet 7, e1002111 (2011).
Meyer, J. R., Gudelj, I. & Beardmore, R. Biophysical mechanisms that maintain biodiversity through trade-offs. Nat. Commun. 6, 6278 (2015).
Fitzsimmons, J. M., Schoustra, S. E., Kerr, J. T. & Kassen, R. Population consequences of mutational events: effects of antibiotic resistance on the r/K trade-off. Evol. Ecol. 24, 227–236 (2010).
Yamamichi, M., Letten, A. D. & Schreiber, S. J. Eco-evolutionary maintenance of diversity in fluctuating environments. Ecol. Lett. 26, S152-S167 (2023).
Abdul-Rahman, F., Tranchina, D. & Gresham, D. Fluctuating environments maintain genetic diversity through neutral fitness effects and balancing selection. Mol. Biol. Evol. 38, 4362–4375 (2021).
Ellegren, H. & Galtier, N. Determinants of genetic diversity. Nat. Rev. Genet. 17, 422–433 (2016).
Mahilkar, A., Venkataraman, P., Mall, A. & Saini, S. Experimental Evolution of Anticipatory Regulation in Escherichia coli. Front. Microbiol. 12, 796228 (2021).
Podrabsky, J. E. & Somero, G. N. Changes in gene expression associated with acclimation to constant temperatures and fluctuating daily temperatures in an annual killifish Austrofundulus limnaeus. J. Exp. Biol. 207, 2237–2254 (2004).
López-Maury, L., Marguerat, S. & Bähler, J. Tuning gene expression to changing environments: from rapid responses to evolutionary adaptation. Nat. Rev. Genet. 9, 583–593 (2008).
Kosterlitz, O. et al. Evolutionary “Crowdsourcing”: Alignment of Fitness Landscapes Allows for Cross-species Adaptation of a Horizontally Transferred Gene. Mol. Biol. Evol. 40, msad237 (2023).
Choudhury, D. & Saini, S. Evolution of Escherichia coli in different carbon environments for 2,000 generations. J. Evol. Biol. 32, 1331–1341 (2019).
He, F. E. coli Genomic DNA Extraction. BIO-Protoc 1, 1 (2011).
Cock, P. J. A., Fields, C. J., Goto, N., Heuer, M. L. & Rice, P. M. The Sanger FASTQ file format for sequences with quality scores, and the Solexa/Illumina FASTQ variants. Nucleic Acids Res. 38, 1767–1771 (2010).
Li, H. & Durbin, R. Fast and accurate short read alignment with Burrows-Wheeler transform. Bioinforma. Oxf. Engl. 25, 1754–1760 (2009).
Li, H. et al. The Sequence Alignment/Map format and SAMtools. Bioinforma. Oxf. Engl. 25, 2078–2079 (2009).
McKenna, A. et al. The Genome Analysis Toolkit: a MapReduce framework for analyzing next-generation DNA sequencing data. Genome Res. 20, 1297–1303 (2010).
Cingolani, P. et al. A program for annotating and predicting the effects of single nucleotide polymorphisms, SnpEff: SNPs in the genome of Drosophila melanogaster strain w1118; iso-2; iso-3. Fly. (Austin) 6, 80–92 (2012).
Acknowledgements
This work was funded by a DBT/Wellcome Trust (India Alliance) grant (Award No. IA/S/19/2/504632) to SS. PV is supported by the Prime Minister’s Research Fellowship (PMRF ID 1302050). NA is supported by the Institute Post-Doctoral Fellowship by IIT Bombay.
Author information
Authors and Affiliations
Contributions
N.A.: Performed Characterization Experiments, Wrote manuscript. R.G.B: Performed Evolution Experiment. P.V.: Performed data analysis, Wrote manuscript, Performed Experiments. S.S.: Conceived & supervised the study, wrote manuscript. All authors reviewed the manuscript.
Corresponding authors
Ethics declarations
Competing interests
N.A., P.V., R.G.B. declare no financial or non-financial competing interests. S.S. serves as Guest Editor for the collection on Evolutionary Systems Biology of this journal and had no role in the peer-review or decision to publish this manuscript. S.S. declares no financial competing interests.
Additional information
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Supplementary information
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.
About this article
Cite this article
Ahlawat, N., Venkataraman, P., Brajesh, R.G. et al. Effects of resource packaging on the adaptative and pleiotropic consequences of evolution. npj Syst Biol Appl 11, 78 (2025). https://doi.org/10.1038/s41540-025-00558-2
Received:
Accepted:
Published:
DOI: https://doi.org/10.1038/s41540-025-00558-2
