Bacterial cross-feeding can promote gene retention by lowering gene expression costs

Reciprocal cross-feeding between bacteria can limit the emergence of metabolic dependencies | bioRxiv /* */ /* */ <!-- <!-- /*! * yepnope1.5.4 * (c) WTFPL, GPLv2 */ (function(a,b,c){function d(a){return"[object Function]"==o.call(a)}function e(a){return"string"==typeof a}function f(){}function g(a){return!a||"loaded"==a||"complete"==a||"uninitialized"==a}function h(){var a=p.shift();q=1,a?a.t?m(function(){("c"==a.t?B.injectCss:B.injectJs)(a.s,0,a.a,a.x,a.e,1)},0):(a(),h()):q=0}function i(a,c,d,e,f,i,j){function k(b){if(!o&&g(l.readyState)&&(u.r=o=1,!q&&h(),l.onload=l.onreadystatechange=null,b)){"img"!=a&&m(function(){t.removeChild(l)},50);for(var d in y[c])y[c].hasOwnProperty(d)&&y[c][d].onload()}}var j=j||B.errorTimeout,l=b.createElement(a),o=0,r=0,u={t:d,s:c,e:f,a:i,x:j};1===y[c]&&(r=1,y[c]=[]),"object"==a?l.data=c:(l.src=c,l.type=a),l.width=l.height="0",l.onerror=l.onload=l.onreadystatechange=function(){k.call(this,r)},p.splice(e,0,u),"img"!=a&&(r||2===y[c]?(t.insertBefore(l,s?null:n),m(k,j)):y[c].push(l))}function j(a,b,c,d,f){return q=0,b=b||"j",e(a)?i("c"==b?v:u,a,b,this.i++,c,d,f):(p.splice(this.i++,0,a),1==p.length&&h()),this}function k(){var a=B;return a.loader={load:j,i:0},a}var l=b.documentElement,m=a.setTimeout,n=b.getElementsByTagName("script")[0],o={}.toString,p=[],q=0,r="MozAppearance"in l.style,s=r&&!!b.createRange().compareNode,t=s?l:n.parentNode,l=a.opera&&"[object Opera]"==o.call(a.opera),l=!!b.attachEvent&&!l,u=r?"object":l?"script":"img",v=l?"script":u,w=Array.isArray||function(a){return"[object Array]"==o.call(a)},x=[],y={},z={timeout:function(a,b){return b.length&&(a.timeout=b[0]),a}},A,B;B=function(a){function b(a){var a=a.split("!"),b=x.length,c=a.pop(),d=a.length,c={url:c,origUrl:c,prefixes:a},e,f,g;for(f=0;f<d;f++)g=a[f].split("="),(e=z[g.shift()])&&(c=e(c,g));for(f=0;f<b;f++)c=x[f](c);return c}function g(a,e,f,g,h){var i=b(a),j=i.autoCallback;i.url.split(".").pop().split("?").shift(),i.bypass||(e&&(e=d(e)?e:e[a]||e[g]||e[a.split("/").pop().split("?")[0]]),i.instead?i.instead(a,e,f,g,h):(y[i.url]?i.noexec=!0:y[i.url]=1,f.load(i.url,i.forceCSS||!i.forceJS&&"css"==i.url.split(".").pop().split("?").shift()?"c":c,i.noexec,i.attrs,i.timeout),(d(e)||d(j))&&f.load(function(){k(),e&&e(i.origUrl,h,g),j&&j(i.origUrl,h,g),y[i.url]=2})))}function h(a,b){function c(a,c){if(a){if(e(a))c||(j=function(){var a=[].slice.call(arguments);k.apply(this,a),l()}),g(a,j,b,0,h);else if(Object(a)===a)for(n in m=function(){var b=0,c;for(c in a)a.hasOwnProperty(c)&&b++;return b}(),a)a.hasOwnProperty(n)&&(!c&&!--m&&(d(j)?j=function(){var a=[].slice.call(arguments);k.apply(this,a),l()}:j[n]=function(a){return function(){var b=[].slice.call(arguments);a&&a.apply(this,b),l()}}(k[n])),g(a[n],j,b,n,h))}else!c&&l()}var h=!!a.test,i=a.load||a.both,j=a.callback||f,k=j,l=a.complete||f,m,n;c(h?a.yep:a.nope,!!i),i&&c(i)}var i,j,l=this.yepnope.loader;if(e(a))g(a,0,l,0);else if(w(a))for(i=0;i (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0];var j=d.createElement(s);var dl=l!='dataLayer'?'&l='+l:'';j.src='//www.googletagmanager.com/gtm.js?id='+i+dl;j.type='text/javascript';j.async=true;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-M677548'); Skip to main content Home About Submit ALERTS / RSS Search for this keyword Advanced Search New Results Reciprocal cross-feeding between bacteria can limit the emergence of metabolic dependencies View ORCID Profile Ying-Chih Chuang , View ORCID Profile Megan G. Behringer , View ORCID Profile Gillian Patton , View ORCID Profile Jordan T. Bird , Jeffrey L. Mazny , Jennifer R. Gliessman , View ORCID Profile Crystal E. Love , View ORCID Profile Ankur Dalia , View ORCID Profile James B. McKinlay doi: https://doi.org/10.1101/2024.08.19.608702 Ying-Chih Chuang 1 Department of Biology, Indiana University , Bloomington, IN, USA 2 Biochemistry Program, Indiana University , Bloomington, IN, USA Find this author on Google Scholar Find this author on PubMed Search for this author on this site ORCID record for Ying-Chih Chuang Megan G. Behringer 3 Department of Biological Sciences, Vanderbilt University , Nashville, TN, USA Find this author on Google Scholar Find this author on PubMed Search for this author on this site ORCID record for Megan G. Behringer Gillian Patton 3 Department of Biological Sciences, Vanderbilt University , Nashville, TN, USA Find this author on Google Scholar Find this author on PubMed Search for this author on this site ORCID record for Gillian Patton Jordan T. Bird 4 Battelle Memorial Institute , Columbus, OH 43201 Find this author on Google Scholar Find this author on PubMed Search for this author on this site ORCID record for Jordan T. Bird Jeffrey L. Mazny 1 Department of Biology, Indiana University , Bloomington, IN, USA Find this author on Google Scholar Find this author on PubMed Search for this author on this site Jennifer R. Gliessman 1 Department of Biology, Indiana University , Bloomington, IN, USA Find this author on Google Scholar Find this author on PubMed Search for this author on this site Crystal E. Love 1 Department of Biology, Indiana University , Bloomington, IN, USA Find this author on Google Scholar Find this author on PubMed Search for this author on this site ORCID record for Crystal E. Love Ankur Dalia 1 Department of Biology, Indiana University , Bloomington, IN, USA Find this author on Google Scholar Find this author on PubMed Search for this author on this site ORCID record for Ankur Dalia James B. McKinlay 1 Department of Biology, Indiana University , Bloomington, IN, USA Find this author on Google Scholar Find this author on PubMed Search for this author on this site ORCID record for James B. McKinlay For correspondence: jmckinla{at}iu.edu Abstract Full Text Info/History Metrics Supplementary material Preview PDF ABSTRACT Cross-feeding is prevalent in microbial communities. Through time, cross-feeding is thought to enrich for loss-of-function mutations, thereby creating or reinforcing dependencies between community members. However, few studies have compared how cross-feeding affects the evolutionary trajectory of partners compared to monoculture conditions. Here we compared mutations that were differentially enriched in bacterial monocultures versus cocultures pairing phototrophic Rhodopseudomonas palustris with fermentative Escherichia coli in an obligate cross-feeding relationship based on the exchange of nitrogen and carbon for 650-800 generations. Opposite trends for the number of differentially enriched mutations were observed for each species; R. palustris accumulated more unique mutations in monoculture whereas E. coli accumulated more unique mutations in coculture. Contrary to expectations, the emergence of additional dependencies was more apparent for both species in monoculture, even though additional layers of cross-feeding involving iron and adenine were present in coculture. We reasoned that iron and adenine cross-feeding occurred at levels sufficient to repress gene expression in the recipient, thereby promoting gene retention by lowering gene cost. Thus, the influence of cross-feeding on evolutionary trajectories can vary with organisms and conditions, and there are situations where cross-feeding can limit, rather than promote, emergent dependencies. IMPORTANCE Bacteria commonly engage in cross-feeding where nutrients are transferred between neighbors. Cross-feeding is thought to alleviate energy expenditures for genes whose role can be met by cross-fed nutrients, leading to eventual gene loss. However, few examples have been documented, especially in comparison to monocultures where bacteria are grown without a cross-feeding partner. Here we grew cocultures pairing phototrophic Rhodopseudomonas palustris with fermentative Escherichia coli alongside corresponding monocultures for 650-800 generations. While coculture conditions required obligate exchange of nitrogen and carbon, additional cross-feeding of adenine and iron likely occurred. Contrary to expectations, iron and uncharacterized dependencies emerged in monocultures but not in cocultures. Low expression for iron scavenging and adenine synthesis genes in cocultures suggested that cross-feeding repressed gene expression, thereby lowering gene cost. Thus, whereas there are likely cases where cross-feeding makes costly genes dispensable, there are also cases where cross-feeding lowers gene cost, thereby promoting gene retention. INTRODUCTION Microbial metabolism shapes local chemical environments in ways that elicit regulated responses from neighboring cells, or even select for adaptive mutations. In some cases, extracellular metabolites can be used as nutrients by neighboring cells, establishing cross-feeding ( 1 ). Cross-feeding can lead to obligate dependencies through adaptive loss-of-function (LOF) mutations, when the benefit of exploiting an extracellular metabolite outweighs the cost of retaining a biosynthetic gene ( 2 ). When the recipient benefits at the expense of the producer, the recipient can be labeled a ‘cheater’ but when the recipient does not detrimentally affect the producer, the recipient can be labeled a ‘beneficiary’ ( 2 , 3 ). The latter is described in the Black Queen Hypothesis (BQH), wherein cross-feeding provides a cost-benefit advantage that promotes adaptive gene-loss ( 2 ). The BQH is commonly used to explain the prevalence of auxotrophs, which cannot synthesize one or more essential nutrients ( 4 , 5 ). Indeed, several groups observed that auxotrophs with biosynthetic gene deletions have a fitness advantage over their prototrophic counterparts when corresponding biosynthetic precursors were provided ( 6 – 10 ). Whereas cross-feeding can promote adaptive gene loss, gene loss is also potentially constrained during cross-feeding by several factors. For example, the energetic costs of N 2 fixation, converting N 2 into 2 NH 4 + via nitrogenase, was predicted to be under pressure for gene loss because NH 4 + can escape and potentially enable higher growth rate for recipients ( 2 ). However, we found that growth rates supported by leaked NH 4 + did not exceed those possible via N 2 -fixation due to low NH 4 + availability relative to N 2 , even when bacteria were engineered to excrete NH 4 + ( 11 ). Moreover, as has been documented with cheaters ( 12 – 14 ), nutrient availability can repress gene expression and thereby lower gene cost ( 4 , 13 , 15 ); most of a gene’s cost comes from its expression ( 16 – 20 ). Thus, repressing gene expression could have similar cost-savings as a LOF mutation and thereby decrease selective pressure for LOF mutants. Previously we engineered cocultures of phototrophic R. palustris and fermentative E. coli to study the mechanisms and evolution of obligate cross-feeding ( 21 – 25 ). Using anoxic conditions, E. coli ferments glucose into organic acids that provide essential carbon to R. palustris while R. palustris converts N 2 into NH 4 + , providing essential nitrogen to E. coli ( 21 ). The two species are not believed to interact in nature, providing an opportunity to observe how the relationship evolves without pre-adaptation ( 22 ). Whereas the coculture was engineered to rely on obligate NH 4 + cross-feeding, other interaction layers exist. For example, we discovered that R. palustris externalizes enough adenine to support an E. coli purine auxotroph population ( 26 , 27 ). How such secondary interactions affect the evolutionary trajectory of each species is unknown. Here we compared differential mutation accumulation in E. coli and R. palustris monocultures versus cocultures for >600 generations. The number of differentially enriched mutations in monocultures versus cocultures was opposite for each organism. For several mutations, we performed a preliminary investigation to gain insights into why they were differentially enriched. In some cases, we correlated a lack of dependency emergence in coculture to inhibitory effects of cross-feeding on gene expression. Our results offer diverse insights into how cross-feeding influences evolutionary trajectories and caution against general expectations for dependencies through LOF mutations. RESULTS AND DISCUSSION R. palustris and E. coli can coexist for hundreds of generations We established long-term, batch, anaerobic (i) monocultures of phototrophic, N 2 -fixing, NH 4 + -excreting R. palustris NifA* (CGA676) ( 28 ), supplied with a mixture of organic acid salts (neutral pH), ethanol, and bicarbonate to mimic E. coli fermentation products, (ii) monocultures of fermentative E. coli PFM2, a MG1655 derivative with repaired rpoS and rph genes ( 29 , 30 ), supplied with 10 mM glucose and 10 mM NH 4 Cl, and (iii) cocultures supplied with 25 mM glucose ( Fig 1A ). All cultures were constantly illuminated, and 2.5% v/v was transferred every seven days. All conditions included an exponential growth phase but only monocultures experienced a nutrient-limited stationary phase. In cocultures, glucose and consumable organic acids remained at the time of transfer and likely supported growth, however slow due to the low pH. Ultimately, all conditions were serially transferred for 650 – 803 generations (∼5 generations per transfer; Fig 1 ). Download figure Open in new tab Fig 1. Long-term anaerobic monoculture and coculture conditions. All cultures were laid flat without shaking and were constantly illuminated. (A) Phototrophic R. palustris NifA* (CGA676) was fed a mixture of organic acids (OAcs), ethanol, and NaHCO 3 to mimic carbon sources available from E. coli fermentation products. The NifA* mutation causes NH 4 + excretion during N 2 fixation. H 2 is also produced as a co-product of N 2 fixation. R. palustris also excretes adenine independent of the NifA* mutation. (B) E. coli PFM2 was fed 10 mM NH 4 Cl and 10 mM glucose, which it fermented to a mixture of OAcs, ethanol, CO 2 , and H 2 . (C) Cocultures were fed 25 mM glucose. Growth of each species is dependent on the exchange of NH 4 + and OAcs. We sequenced gDNA from populations at several timepoints. We then compared mutations that were differentially enriched in monoculture versus coculture by applying following criteria: (i) exclude synonymous mutations, (ii) no exact mutation in the opposing monoculture or coculture condition, (iii) observed in the last two sequenced timepoints, (iv) with a frequency >0.1 in the last timepoint. These criteria resulted in a list of 167 mutations from R. palustris monocultures, 31 from E. coli monocultures, 43 from R. palustris in cocultures, and 132 from E. coli in cocultures ( Fig 2A , Table S1-5). These counts include redundancy, where some mutations were observed in multiple lines. In all conditions, most mutations were nonsynonymous, except for E. coli monocultures, where most mutations were small insertions/deletions ( Fig 2B ). Mutations due to mobile genetic elements were only observed in E. coli ( Fig 2B ). To narrow our focus on positive selection, we added a fifth criteria of parallelism where the same gene or intergenic region must be mutated in at least two lineages. Download figure Open in new tab Fig 2. Differentially enriched mutations in long-term monocultures and cocultures. Mutation counts and fractions are the sum of mutations across all 10 lines in each condition. To be counted, mutations had to meet the following criteria: (i) exclude synonymous mutations, (ii) no exact mutation in the opposing monoculture or coculture condition, (iii) observed in the last two sequenced timepoints, ( Rp monoculture, generations 781 and 803; Ec monoculture, generations 500 and 650; coculture, generations 655 and 732), (iv) with a frequency >0.1 in the last timepoint. Differentially enriched R. palustris mutations mostly occurred in monocultures Most differentially mutated R. palustris genes and intergenic regions occurred in monocultures: 19 in monoculture vs 5 in coculture ( Fig 3 ). We considered that the higher mutation frequency in monoculture could be due to a mutS, encoding a mismatch repair protein; lines 2 and 7 had a mutS mutations that met our criteria and most other lines had a low frequency mutS mutation at some point during the experiment. Download figure Open in new tab Fig 3. Mutations enriched in R. palustris NifA* in long-term monocultures and cocultures. Each point represents a mutation in a single lineage. Mutations are shown if they are not a synonymous mutation and were observed: (i) exclude synonymous mutations, (ii) no exact mutation in the opposing monoculture or coculture condition, (iii) observed in the last two sequenced timepoints, ( Rp monoculture, gen 781 and 803; Rp coculture, gen 655 and 732), (iv) with a frequency >0.1 in the last timepoint, and (v) the same gene or intergenic region was mutated in at least two lines for a given condition. Generation: gen. However, the total number of differentially-enriched mutations was similar across R. palustris monoculture lines (range: 8-23; Table S1) with only 6 of 21 mutations in line 2 and 11 of 23 mutations in line 7 potentially attributed to mutS , since they occurred at a lower frequency than the mutS mutation ( mutS mutation frequency was 13 and 22% in lines 2 and 7, respectively; Table S1). Thus, mutS mutations did not likely drive many of the other mutations. Several differentially mutated genes in coculture also acquired different mutations in monoculture. Among these was ppsR2 , encoding an O 2 -responsive regulator of genes for phototrophic growth. ppsR2 is commonly mutated in R. palustris because they provide an advantage under low-light conditions, which occurs at high cell densities ( 31 ). Other genes that were commonly mutated in monoculture and coculture were cbbM , encoding the Calvin cycle type II Rubisco and cbbR , encoding a Calvin cycle regulator. It is unclear why mutations were enriched in these genes, but it is possible that they are associated with the role of CO 2 fixation in electron balance during photoheterotrophic growth or electron distribution between CO 2 and N 2 fixation ( 28 , 32 ). We did not investigate Calvin cycle gene mutations further herein. R. palustris monocultures evolved to utilize H 2 Both species produce H 2 , E. coli via fermentation and R. palustris as a coproduct of nitrogenase activity ( 33 ). Curiously, H 2 levels decreased through serial transfers in R. palustris monocultures but not in cocultures ( Fig 4A, B ). A commonly mutated gene in R. palustris monocultures was hoxJ, which encodes a sensor kinase that governs H 2 -oxidizing hydrogenase by repressing the transcriptional activator HoxA ( Fig 4C ). Normally, HoxJ repression of HoxA is relieved when HupUV senses H 2 . However, the R. palustris ancestor had a hupV mutation, causing HoxJ to constitutively repress HoxA ( 34 ). Deleting HoxJ was previously shown to lead to hydrogenase activity ( 34 ). Download figure Open in new tab Fig 4. R. palustris monocultures evolve H 2 utilization. H 2 was quantified in 7-day-old cultures at selected time points for R. palustris monoculture lines 1-10 (A) and coculture lines 21-30 (B) . (C) R. palustris NifA* (CGA676) contains a hupV mutation that prevents the H 2 sensor HupUV from repressing HoxJ. HoxJ then constitutively represses the transcriptional activator of hydrogenase (H 2 ase) genes, HoxA. (D) Stationary phase H 2 in the NifA* ancestor without and with a 5-bp hoxJ deletion commonly observed in evolved monocultures. Cultures were grown in the same MDC medium used for monoculture serial transfers plus 1 µM NiSO 4 . (E) Stationary phase H 2 levels in the NifA* ancestor without or with hupV repaired ( hupV +) or hoxJ deleted. Monocultures were grown in MDC with 20 mM acetate. Cocultures were grown in same media used for coculture serial transfers. (D, E) Bars, mean ± SD; n=3. Based on the above, we hypothesized that our hoxJ mutants also permitted H 2 oxidation. To test this hypothesis, we introduced a 5-bp hoxJ deletion, observed in 7 monoculture lines, into the NifA* ancestor. Indeed, the mutant did not accumulate H 2 , ( Fig 4D ). Thus, we reason that hoxJ mutations that were enriched in monoculture allowed R. palustris to use H 2 as an electron source. We questioned why hoxJ mutants were only enriched in monoculture. Previously, transposon disruption of hoxJ in monoculture led to a fitness benefit in stationary phase ( 35 ). The benefit likely stemmed from H 2 -mediated CO 2 fixation after organic substrates were exhausted; strains with WT hoxJ could not grow after this point ( 35 ). We first asked whether there is something about coculture conditions that generally prevents H 2 oxidation. When we repaired the hupV mutation ( hupV +) or deleted hoxJ , no H 2 accumulated in either monoculture or coculture, indicating that R. palustris hydrogenase can be activated in coculture ( Fig 4E ). Therefore, something must have prevented enrichment of metabolically active hoxJ mutants in coculture. We considered two possibilities that could explain the lack of hoxJ mutant enrichment in coculture. First, consumable organic acids, which remained available at the time of transfer in cocultures, could have allowed the general R. palustris population to remain competitive against hoxJ mutants, preventing their enrichment. A second possibility is that E. coli outcompeted R. palustris for nickel, which is required for hydrogenase activity. Since our media used for serial transfers lacked a nickel supplement, nickel must have come as a contaminant in other media components, and its availability was likely low. Although the absence of H 2 in cocultures with Δ hoxJ or hupV + mutants indicates that R. palustris can compete against E. coli for nickel, this might only be true when the H 2 -oxidizing R. palustris population is large. An emergent, low frequency, H 2 -oxidizing R. palustris subpopulation might instead be outcompeted by E. coli for nickel. R. palustris siderophore gene mutations are correlated with expression level A frequently mutated gene cluster in R. palustris monocultures was RPA2382-90, for siderophore synthesis ( Fig 3 , Fig 5A, B ). We speculate that these were LOF mutations, giving rise to ‘cheaters’ that scavenged siderophores from a siderophore-producing subpopulation. However, we questioned why R. palustris siderophore mutants were not observed in cocultures. Download figure Open in new tab Fig 5. Siderophore gene loss is prevalent in R. palustris monocultures where gene expression is high. A, B. R. palustris siderophore gene (RPA2386-90) mutation frequencies in monoculture ( A ) and coculture lines ( B ). No mutations were observed in lines 21, 25, and 27. Repeated lines with the same symbols in a given graph indicate different mutations in the same gene. C, D. RT-qPCR quantification of siderophore synthesis genes in R. palustris ( C ; RPA2390 relative to fixJ ) and in E. coli ( D ; entF relative to hcaT ). Values are the mean ± SD, n = 4. Statistically significant differences from the monoculture (mono) condition for each strain were determined using an unpaired two-tail t-test; *, p < 0.05; **, p < 0.01; co, coculture; Fe, ammonium ferric citrate. We hypothesized that siderophore gene mutations were not enriched in coculture because E. coli facilitated R. palustris iron acquisition, thus decreasing the need for R. palustris to synthesize siderophores. The situation would be analogous to that observed for P. aeruginosa , which produces siderophores only when needed, due to their cost. High P. aeruginosa siderophore production in low-iron conditions led to a higher frequency of siderophore-deficient mutants compared to iron-rich conditions where siderophore production was low ( 12 – 14 ). In support of this hypothesis, we previously saw that R. palustris siderophore gene transcripts were low in coculture with E. coli MG1655, relative to monocultures ( 23 ). We verified that this trend was also true for R. palustris in coculture with E. coli PFM2; RT-qPCR analysis showed that the RPA2390 transcript level in coculture was 20% of that in monoculture ( Fig 5C ). Adding soluble iron to monocultures decreased RPA2390 expression to 3% of that observed without added iron, suggesting that the low expression in coculture was in response to higher iron availability ( Fig 5C ). Thus, iron must be scarce in R. palustris monocultures, inducing siderophore production, whereas in coculture, E. coli somehow increases iron availability, repressing R. palustris siderophore synthesis. The mechanism of increased iron availability for R. palustris in coculture deserves investigation beyond the scope of the current study. However, we suspect that R. palustris can use E. coli siderophores. Others have speculated that R. palustris uses foreign siderophores because its genome encodes multiple siderophore transporters but only one synthesis cluster ( 36 , 37 ). In support of this hypothesis, the E. coli siderophore entF transcript level was 3.3-fold higher in coculture than in monoculture ( Fig. 5D ), perhaps in response to iron loss to R. palustris . Other R. palustris mutations with unknown phenotypes Other genes that were mutated in monoculture include a methylmalonyl-CoA mutase encoded by RPA1837, a possible flagellin encoded by RPA2297 that is outside the flagellar gene cluster, and a hypothetical protein encoded by RPA4084 (6 of 10 lines). One of the few R. palustris genes that was differentially mutated in at least 5 coculture lines was RPA3297, annotated to encode a urea or short-chain amino acid transporter. Each mutation was an insertion or deletion that should cause a frame shift and thus they are all likely LOF mutations. This transporter was previously observed to have elevated protein levels in coculture compared to monoculture ( 23 ). We speculate that this elevated expression was disadvantageous, but the function of the gene product remains unknown. Two R. palustris coculture lines also had a common intergenic mutation between hbdA and etfA . These genes are likely involved in fatty acid degradation and might be commonly regulated by LiuR ( 38 ). The effect of this mutation is unknown. Differentially enriched E. coli mutations mostly occurred in cocultures The number of differentially enriched mutations in E. coli were oppositely skewed than for R. palustris ( Fig 2 ). The same pattern held for differentially mutated E. coli genes and intergenic regions in at least two lineages per condition: 6 were mutated in monoculture vs 19 in coculture ( Fig 6 ). Similar observations of cocultures leading to more mutations were made by another group who paired E. coli auxotrophs in reciprocal cross-feeding cocultures versus monocultures ( 39 ). Download figure Open in new tab Fig 6. Mutations enriched in E. coli PFM2 in long-term monocultures and cocultures. Each point represents a mutation in a single lineage. Mutations are shown if they are not a synonymous mutation and were observed: (i) exclude synonymous mutations, (ii) no exact mutation in the opposing monoculture or coculture condition, (iii) observed in the last two sequenced timepoints, ( Ec monoculture, gen 500 and 650; Ec coculture, gen 655 and 732), (iv) with a frequency >0.1 in the last timepoint, and (v) the same gene or intergenic region was mutated in at least two lines for a given condition. Generation: gen. E. coli mutations in coculture reflect adaptation to nitrogen starvation In coculture, the E. coli growth rate is limited to ∼20% of maximum due to the rate at which it receives NH 4 + from R. palustris ( 21 , 22 ). As in previous long-term cocultures pairing a different R. palustris NifA* strain with E. coli MG1655 ( 22 ), several E. coli mutations in coculture suggested adaptation to this low-nitrogen condition. For example, mutations were commonly observed in glnGL, encoding the two-component system NtrBC that controls nitrogen scavenging. Previously, we demonstrated that NtrBC is critical for E. coli survival in the low-nitrogen coculture environment ( 22 , 23 ). Other mutations in glnK and amtB , which respectively encode a nitrogen-responsive regulatory protein and an NH 4 + transporter, likely also help with nitrogen acquisition in coculture ( Fig 6 ); we also previously demonstrated the importance of E. coli AmtB in coculture ( 24 ). E. coli purine transporter mutants are enriched in coculture We designed our cocultures to be based on the reciprocal exchange of fermentation products and NH 4 + ( 21 ). We later discovered that R. palustris also externalizes adenine that can be consumed by E. coli ( 26 , 27 ). Herein, E. coli exhibited several mutations in coculture that could be in response to adenine availability. For example, mutations were enriched upstream of adeD (yicP) encoding adenine deaminase ( Fig 6 ). Insertion elements upstream of adeD in the adenine permease gene adeQ (yicO) are known to activate adeD expression ( 40 ). Mutations were also enriched in the adenine permease gene, adeP ( purP ) ( Fig 6 ). The adeP mutations were accompanied by a multi-gene amplification that created up to 40 copies of adeP and neighboring genes ( Fig 7A, B ). We revisited sequencing results from our previous long-term cocultures featuring E. coli MG1655 ( 22 ) and found similar mutations upstream of adeD and amplification of the same region ( Fig 7C ). Evolved E. coli monocultures did not show this amplification but had up to 8-fold amplification in another region, with no obvious connection to adenine (Fig S1). Download figure Open in new tab Fig 7. Amplification of adeP alone does not benefit PFM2 in the presence of adenine. A. Region of amplified E. coli genes observed in MG1655 and PFM2 cocultures but not in monocultures. The boundaries of the amplification differed between lineages but always contained these genes. B, C. Level of amplification observed in MG1655 in coculture ( B ) and in PFM2 in coculture ( C ). D. Effect of over-expressing adeP on PFM2 growth in monocultures with and without 35 μM adenine (ade). Over-expression was achieved by expressing adeP from the plasmid pCA24N with 30 μM IPTG. Data points are mean ± 95% C.I.; n=3. EV, empty vector. E. Growth rates from the same conditions as panel D ; Bars, mean ± 95% C.I.; n=6, combining two separate experiments. Different letters indicate statistical differences ( p <0.05) from a one-way ANOVA with Tukey’s multiple comparisons test. We hypothesized that these mutations could lead to increased fitness in the presence of adenine. We attempted to mimic a high adeP copy number via IPTG-induced adeP expression from a plasmid. However, over-expressing adeP in PFM2 did not improve growth in the presence of adenine over an empty vector control ( Fig 7D, E ). It is possible that adeP amplification is only advantageous in the context of the surrounding amplified region or other enriched mutations. To address an adenine-related fitness benefit more generally, we compared evolved E. coli growth with and without adenine. Adenine benefited several coculture-evolved PFM2 isolates, suggesting that some mutations were adaptive to adenine availability ( Fig 8 ). Although some monoculture isolates also responded positively to adenine, including two possible auxotrophs, the trend was less pronounced; when isolates were treated as replicates, the change in cell density with adenine versus without was significantly higher for coculture-evolved PFM2 isolates but not for monoculture evolved isolates (paired one-tail t-test, p = 0.002 versus 0.14, respectively). Download figure Open in new tab Fig 8. E. coli adenine auxotrophs are not prevalent in long-term cocultures. Randomly chosen evolved E. coli isolates (generation 650; n=4) were grown overnight in monoculture conditions with and without 50 μM adenine alongside the ancestral strain (Anc; PFM2) and a ΔpurH ( Δ H) mutant (n=3). Each point represents a single measurement for a single isolate; replicate measurements were not made for any isolate. Each line connects measurements for the same isolate grown with and without adenine to facilitate comparisons. Strain-specific adenine toxicity likely influences E. coli evolution Adenine availability in coculture might also explain differential enrichment of mutations in previous cocultures featuring E. coli MG1655 ( 22 ). MG1655 is inhibited by adenine due to an rph mutation that disrupts pyrE regulation ( 41 ), leading to suboptimal precursor distribution between purine and pyrimidine synthesis in the presence of adenine ( 42 ). The rph mutation is repaired in E. coli PFM2 ( 29 , 30 ). Through serial transfers, PFM2 mutations in rph and purine and pyrimidine synthesis genes were rare in monoculture and coculture ( Fig 9A ). In contrast, MG1655 populations (only evolved in coculture) contained a high frequency of rph mutations and a low frequency of purine synthesis gene ( pur ) mutations ( Fig 9B ). We hypothesized that MG1655 rph and purine biosynthesis mutations alleviated adenine toxicity. Indeed, repairing the ancestral rph mutation or deleting purH , rendering MG1655 a purine auxotroph, alleviated alanine toxicity ( Fig 9C ). Download figure Open in new tab Fig 9. Mutations in genes for nucleotide synthesis are enriched in E. coli MG1655 but not PFM2. (A,B) Each point represents a mutation frequency in a given gene for a given line of E. coli PFM2 (A) or MG1655 (B) . (C). E. coli MG1655 adenine sensitivity can be relieved by repairing rph or deleting purH . Strains were grown in E. coli monoculture conditions without (-) or with (+) 35 μM adenine. * Statistical differences ( p <0.05) compared to the wild-type (WT) conditions without adenine using a one-way ANOVA. NG, no growth. Coculture adenine availability did not enrich for adenine auxotrophs Given that de novo purine synthesis is energetically costly, we expected to observe E. coli purine synthesis LOF mutations in cocultures where adenine is available from R. palustris ; adenine is amply available in long-term cocultures ( 43 ). However, no purine biosynthesis gene mutations were observed in cocultures ( Fig 9A ), nor were adenine auxotrophs observed among 40 coculture isolates screened ( Fig 8 ); only four monoculture isolates in lines 18 and 20 were possibly adenine auxotrophs ( Fig 8 ). The BQH predicts that cross-feeding will lead to LOF mutations only when the benefit of cross-feeding outweighs the cost of maintaining the gene ( 2 ). To directly test whether adenine auxotrophs would have an advantage in coculture, we compared PFM2 growth versus that of a ΔpurH mutant in a minimal medium with and without adenine. The ΔpurH mutant required adenine for growth, but despite the availability of adenine, the ΔpurH mutant did not grow faster than the PFM2 parent ( Fig 10A, B ). We also directly compared the fitness of PFM2 versus a ΔpurH mutant in coculture with R. palustris NifA* (CGA676). We used a range of initial frequencies so that we could simultaneously test for possible coexistence by mutual invasion, where an equilibrium frequency can be extrapolated from the x-intercept by linear regression analysis ( 11 , 44 – 46 ). Although there was poor linear correlation in the invasion-from-rare assay, the ΔpurH mutant tended to decrease in frequency relative to parent in cocultures, where adenine is available ( Fig 10C ). Thus, our results suggest that a ΔpurH mutant does not have a competitive advantage over PFM2 in the presence of adenine, thus explaining why PFM2 adenine auxotrophs were not enriched in cocultures. Download figure Open in new tab Fig 10. Adenine availability does not provide a fitness advantage to an engineered E. coli PFM2 purine auxotroph. A. Growth of WT E. coli PFM2 and its ΔpurH mutant in monoculture with and without 35 μM adenine (ade). Points, mean ± SD as shading; n=3. B. Monoculture growth rates for WT PFM2 vs a corresponding ΔpurH mutant with and without 35 μM adenine. ‘a,’ indicates statistically similar values ( p > 0.01) as determined using a one-way ANOVA with Tukey’s multiple comparisons test. Bars, mean ± SD; n=3. C. Invasion-from-rare assay competing a ΔpurH mutant against its PFM2 parent in coculture with R. palustris CGA676 under N 2 -fixing conditions where R. palustris excretes NH 4 + and adenine. The orange line is the best fit from a linear regression analysis with 95% CI shaded in purple. Change in frequency = ( E. coli Δ purH / ( E. coli WT + E. coli Δ purH )) final – ( E. coli Δ purH / ( E. coli WT + E. coli Δ purH )) initial . B, C. Each data point represents a measurement from a single biological replicate. Adenine represses E. coli purine synthesis gene expression One reason why PFM2 purine auxotrophs were not enriched in coculture could be if adenine availability repressed gene expression, thereby decreasing gene cost. This would be analogous to the situation described above for R. palustris siderophore genes. In support of this hypothesis, we previously saw that E. coli MG1655 had lower pur transcript levels in coculture ( 23 ). We verified that the same is true for PFM2 by RT-qPCR quantification of purH transcript; coculture PFM2 purH levels were 20% of those in monoculture ( Fig 11A ). Adding adenine to monocultures resulted in a purH transcript level that was 12% of that observed without adenine, suggesting that the low expression in coculture was due to adenine availability ( Fig 11A ). Download figure Open in new tab Fig 11. Purine biosynthesis genes are down-regulated in coculture. RT-qPCR quantification of purH transcript levels in E. coli PFM2 ( A ; relative to hcaT ) and R. palustris CGA676 ( B ; relative to fixJ ). 100 μM adenine (+ ade) was used to ensure that it was not used up before harvesting RNA. Bars, mean ± SD, n = 3-4. Statistically significant differences from the monoculture condition for each strain were determined using an unpaired two-tail t-test; ***, p < 0.001; ****, p < 0.0001. R. palustris also had less purH transcript in coculture, 50% of that in monoculture ( Fig 11B ). It is unclear how E. coli would influence R. palustris purH transcript levels. Regardless, it is curious that adenine availability combined with purH expression in monoculture did not lead to R. palustris pur LOF mutations (Table S1) nor any purine auxotrophs ( Fig 12 ). We speculate that R. palustris lacks adenine uptake mechanisms, which would also help explain extracellular adenine accumulation ( 26 , 43 ); poor uptake mechanisms can be an important factor in extracellular metabolite accumulation ( 47 – 49 ). Download figure Open in new tab Fig 12. R. palustris auxotrophs are not prevalent in long-term cocultures. Randomly chosen evolved- R. palustris isolates (generation 650) were grown overnight in monoculture conditions with and without 50 μM adenine alongside the ancestral strain (Anc; CGA676) (n=4). Monoculture isolates were grown for 6 days and coculture isolates were grown for 7 days due to inclement weather, each with a corresponding ancestral control. Each point represents a single measurement for a single isolate; replicate measurements were not made for any isolate. Each line connects measurements for the same isolate grown with and without adenine to facilitate comparisons. E. coli monocultures enrich for auxotrophs Whereas cocultures did not enrich for E. coli auxotrophs, several evolved monoculture PFM2 isolates could not grow in a minimal medium ( Fig 8 ). Enrichment of auxotrophs in monoculture could suggest the emergence of cross-feeding between E. coli subpopulations that were not available in coculture, perhaps due to competition from an R. palustris population that is an order-of-magnitude larger ( 21 , 22 ). Although we did not investigate these possible E. coli monoculture cross-feeding relationships further, we speculate that a source of the auxotrophies could be mutations in rpoC , encoding the RNA polymerase β’-subunit. RpoC mutations are known to result in polyauxotrophs ( 50 ). Curiously, we did not see a similar emergence of R. palustris auxotrophs ( Fig 12 ). E. coli and other diverse bacteria are known to support, or quickly evolve to support, auxotrophic strains via cross-feeding ( 51 – 55 ). It is possible that R. palustris cannot support diverse auxotrophs beyond those that require purines ( 26 ). Other E. coli mutations with unknown phenotypes Other differentially enriched E. coli mutations include those in aceE that evolved in monocultures ( Fig 6 ). aceE encodes a pyruvate dehydrogenase subunit, which is important for respiration but is generally not involved in E. coli fermentative metabolism ( 56 ). It is possible that expression of aceE is sub-optimally high under the high growth rates supported in monoculture conditions ( 23 ). Other mutations enriched in monoculture occurred in genes for post-translational regulation of the TCA cycle enzyme isocitrate dehydrogenase ( aceK ), ribosome modification ( rluC ), a vitamin C transporter ( ulaA/B ), and a deoxyribose salvage enzyme ( deoB ). Several genes mutated in coculture might impact biofilm formation and acid sensitivity. We speculate that these mutations could help E. coli associate with R. palustris and/or E. coli cross-feeding partners and potentially respond to fermentative acid stress. For example, csgD/B is involved in curlin-based biofilm formation ( 57 ), which is regulated in part by cpxRP genes that were also mutated ( 58 ). ycgV encodes an outer membrane protein that might be involved in biofilm formation and fimH/A are involved in fimbriae-mediated adhesion ( 57 ). Mutations in yifE ( maoP )/ hdfR , might also affect biofilm formation. MaoP was noted to affect biofilm formation in addition to its role in chromosome positioning and segregation ( 59 ). HdfR might have a regulatory link to the transition between motile and sessile lifestyles ( 60 ). HdfR might also be linked to acid resistance by regulating amino acid decarboxylation genes encoded by gltBD ( 61 ). Mutations in hdfR were also observed in cocultures with MG1655 ( 22 ). Mutations were also observed in cspC / yobF , which encode for proteins that respond to acid and other stressors ( 62 ). Some of the genes associated with the amplified region in cocultures might also be linked to acid resistance such as mnmE ( 63 ) and yidZ ( 64 ). Concluding remarks Long-term serial transfers of R. palustris and E. coli monocultures and cocultures differentially enriched for mutations. More differentially enriched mutations were observed in R. palustris monocultures compared to cocultures, and vice-versa for E. coli . Our results suggest that the extent to which cross-feeding influences the evolutionary landscape depends on the organisms and conditions. A generalization from our observations is that monocultures enriched for dependencies more than cocultures. Against expectations for cross-feeding to promote LOF mutations in a recipient related to synthesis of a cross-fed metabolite, R. palustris siderophore mutants and unknown E. coli auxotrophs were only enriched in monoculture. We reason that adenine and siderophore LOF mutants were not observed in coculture because cross-feeding generated metabolite levels that were sufficient to repress gene expression, thereby lowering gene cost; when gene cost is minimized, selective pressure for LOF mutations is also minimized, thus promoting gene retention. Our results suggest that the effects of cross-feeding on gene loss versus retention can be more nuanced than is often assumed. MATERIALS AND METHODS Bacterial strains R. palustris CGA676 is derived from CGA0092 ( 65 ) and carries a nifA * mutation that causes NH 4 + excretion under N 2 -fixing conditions ( 21 , 28 ). R. palustris mutants were made by homologous recombination using constructs in the suicide vector pJQ200SK ( 66 ). All primers are in Table S6. pJQhoxJΔ5bp was made by amplifying 1.4 kb from an evolved isolate from line A2 using primers JLM32 and JLM33, digesting with BamHI and XhoI sites, and then ligating into pJQ200SK that had been digested with the same enzymes. The ligation reaction was used to transform NEB10β for plasmid verification by PCR and Sanger sequencing. pJQhoxJΔ5bp was then used to transform E. coli S17-1 by electroporation. In each case, transformed E. coli was recovered on LB agar with gentamycin (30 μg ml -1 ). E. coli S17-1 with pJQ-ΔhoxJ and pJQ-hupV were gifts from C. S. Harwood ( 67 ). R. palustris nifA* hoxJ Δ5bp (CGA4065), nifA* Δ hoxJ (CGA4050) and nifA * hupV + (CGA4062) were then made by introducing pJQhoxJΔ5bp, pJQ-ΔhoxJ, or pJQ-hupV, respectively, into CGA676 by conjugation with an E. coli S17-1 donor as described ( 67 ). Single recombinants were recovered on photosynthetic medium (PM) agar ( 68 , 69 ) with 10 mM succinate and 100 μg ml -1 gentamycin. Double recombinants were recovered by counterselection on PM succinate plates with 10% sucrose, screened for gentamycin sensitivity, and then verified by PCR amplification and/or Sanger sequencing. E. coli PFM2 was a gift from P. Foster ( 29 , 30 ). The Δ purH mutant was made via lambda Red recombination ( 70 ) using constructs amplified from the Δ purH ::km R KEIO mutant JW3970 ( 71 ) using primers YCC29 and YCC30. FLP-mediated excision was used to remove the km resistance cassette ( 70 ). All mutants were verified by PCR amplification and Sanger sequencing. Growth conditions Colonies were recovered from 25% glycerol −80 °C frozen stocks by streaking to lysogeny agar ( E. coli ) or PM agar with 10 mM succinate. Single colonies were then used to inoculate starter cultures. Anoxic media in test tubes were prepared by bubbling N 2 through 10 ml of media in 27-ml anaerobic test tubes, then sealing with rubber stoppers and aluminum crimps prior to autoclaving. Monocultures and cocultures were grown horizontally at 30°C with light from a 45 W halogen bulb (430 lumens) and shaking at 150 rpm in minimal M9-derived coculture medium (MDC) ( 21 ) with either (i) E. coli monocultures: 10 mM glucose, 10 mM NH 4 Cl, and cation solution (100X stock: 100 mM MgSO 4 and 10 mM CaCl 2 ); R. palustris monocultures: 20 mM sodium acetate and 10 mM NH 4 Cl; or cocultures: 25 mM glucose and cation solution. Cultures with plasmid-carrying strains were also supplemented with 100 μg/ml gentamycin or 25 μg/ml chloramphenicol as appropriate. Starter cultures were inoculated with single colonies. R. palustris starter cultures were grown in MDC with 20 mM acetate and 10 mM NH 4 Cl. E. coli starter cultures were grown aerobically in lysogeny broth, with 30 μg/ml kanamycin (km) when appropriate. E. coli starter cultures were washed twice in 1 ml MDC prior to inoculating test cultures or bioassays. Cocultures were inoculated with 0.1 ml each of R. palustris and E. coli to an initial optical density (OD 660 ) of ∼0.003 each. Cultures in 96-well plates were treated similarly except that oxic solutions were used to prepare 0.2 ml volumes in each well. Anoxic conditions were achieved by sealing plates inside a BD GasPak EZ large incubation container with anaerobic sachets. Experimental evolution Founder monocultures of E. coli PFM2 and R. palustris NifA* CGA676 were each grown from a single colony in anoxic MDC with either 10 mM glucose, cation solution, and 3 mM NH 4 Cl for PFM2 or 20 mM sodium acetate for CGA676. A single founder monoculture was then used to inoculate 10 monocultures and 10 cocultures. All cultures were grown horizontally without shaking at 30°C with light in MDC. PFM2 monocultures were supplemented with 10 mM glucose, cation solution, 25 mM NaCl, and 2.3 mM NH 4 Cl. These cultures were started later after finding that 25 mM glucose led to periodic extinction due to prolonged exposure to acidic fermentation products. As a result, E. coli monocultures were transferred less times than for other conditions. No data from monocultures with 25 mM glucose were presented herein. CGA676 monocultures were supplemented with 25 mM glucose, cation solution, 8 mM disodium succinate, 7.3 mM sodium acetate, 0.25 mM sodium formate, 1.4 mM sodium lactate, and 6.3 mM ethanol. Cocultures were supplemented with 25 mM glucose, cation solution, and 25 mM NaCl. Acidification was not a problem despite using 25 mM glucose because R. palustris consumes enough organic acids to dampen the pH drop ( 21 ). Every 7 days, cultures were vortexed and 0.25 ml was transferred to fresh medium. About every 5 transfers, stocks were diluted 1:1 with 50% glycerol and stored at −80°C. Separately, cell pellets from 1 ml samples were frozen for gDNA extraction. Analytical procedures Cell densities were measured via turbidity (OD 660 ) using a Genesys 20 visible spectrophotometer (Thermo-Fisher). H 2 was measured using a Shimadzu GC-2014 gas chromatograph with a thermal conductivity detector as described by sampling 0.1 ml of headspace using a gas-tight syringe ( 72 ). Statistical analyses were performed using Graphpad Prism (v10). Competition assays Competition assays were conducted in an invasion-from-rare format to consider coexistence by the mutual invasion criterion, where each population can increase when rare ( 44 , 46 ). Cocultures were started from various initial frequencies (targeting 0.01 – 0.99) of each E. coli strain (WT vs Δ purH or WT vs Δ purH ::km R ) for a total initial cell density of ∼10 6 colony forming units (CFU) / ml. R. palustris CGA676 was inoculated to an initial density of ∼10 6 CFU / ml. Frequencies were determined upon inoculation and after 5 days. WT and Δ purH mutants were distinguished by counting CFUs on M9 agar with cations and 25 mM glucose and on LB agar and then determining Δ purH populations from the difference. When Δ purH ::km R mutants were used LB agar included km to allow for direct determination of population size. Change in frequency = ( E. coli Δ purH / ( E. coli WT + E. coli Δ purH )) final – ( E. coli Δ purH / ( E. coli WT + E. coli Δ purH )) initial ( 45 ). Genome sequencing and mutation analysis gDNA was purified from cells using a Qiagen DNeasy Blood and Tissue kit following the manufacturer’s instructions. Lysis was facilitated after resuspension by adding proteinase K (50 μg/ml final), and incubating at 56°C for 10 min. RNaseA (4 μl, Promega) was then added and the lysate was incubated for 2 min before proceeding. DNA fragment libraries were made using a NextFlex Bioo Rapid DNA kit and libraries were sequenced using Illumina NextSeq 500 150×150 paired-end runs by the IU Center for Genomics and Bioinformatics. Paired-end reads were pre-processed for quality with cutadapt 3.4 ( 73 ) with the following options: -a AGATCGGAAGAGC -A AGATCGGAAGAGC; -q 15,10; -u 6. Mutations were called using breseq v. 0.32.0 on polymorphism mode ( 74 ). E. coli monoculture population sequences were mapped to the MG1655 genome (accession NC_000913 ); R. palustris monoculture population sequences were mapped to a concatenated reference genome consisting of the CGA009 chromosome (accession BX571963 ) , and its plasmid pRPA (accession BX571964 ). Co-culture sequences were mapped to a concatenation of the E. coli and R. palustris reference genomes. Polymorphisms that co-occurred in both the monoculture and co-culture datasets were filtered and maintained as a subset to enrich for the most informative variants representing treatment differences (Table S1-4). This filtering step also removed sequence differences between the reference sequences and those of the experimental strains used. Variants were prioritized as mutations of interest if they were detectable at the final two sequencing timepoints and co-occurred across multiple populations in the same locus. All mutations can be found in Tables S1-4. Locus tag conversions can be found in Table S5. Reverse transcription quantitative real-time PCR (RT-qPCR) Cultures received 100 μM adenine or ammonium ferric [iron(III)] citrate as indicated. Cultures were harvested in exponential phase at 0.6-0.8 OD 660 except for E. coli monocultures (+/- adenine experiment), which were harvested at 0.3-0.4 OD 660 to avoid adenine depletion. Cultures were chilled on ice and pelleted by centrifugation. Lysis, RNA purification, and cDNA generation were performed exactly as described ( 26 ). Standard curves were generated using gDNA. Transcripts were quantified as described ( 26 ) using the appropriate primers (Table S6) with a Mastercycler ep realplex real-time PCR system (Eppendorf). Data was analyzed by realplex software using Noiseband. Specificities were validated by melting curves and by the presence of a single band on an agarose gel. ACKNOWLEDGEMENTS This work was supported in part by US Army Research Office grants W911NF-14-1-0411 and W911NF-17-1-0159, the National Science Foundation CAREER award MCB-1749489 to JBM, and National Institutes of Health grants R35GM128674 to ABD and R35GM150625 to MB. Supercomputing resources were supported in part by Lilly Endowment, Inc., through its support for the IU Pervasive Technology Institute. We are grateful to A. Cairo, S. Kamaran, and N. Ramli, for maintaining long-term cultures. We thank A. Drummond, J. Drummond, P. Foster, J.P. Gerdt, J. Lennon, M. Lynch, T. Romeo, and the McKinlay lab for helpful advice. Funder Information Declared United States Army Research Office, https://ror.org/05epdh915 , W911NF-14-1-0411 , W911NF-17-1-0159 U.S. National Science Foundation, https://ror.org/021nxhr62 , MCB-1749489 National Institutes of Health , R35GM128674 , R35GM150625 Footnotes Manuscript has been expanded to include a more detailed view of genomic and phenotypic changes that occurred over long-term monocultures and cocultures. REFERENCES 1. ↵ R. K. Fritts , A. L. McCully , J. B. McKinlay , Extracellular metabolism sets the table for microbial cross-feeding . Microbiol. Mol. Biol. Rev . 85 , e00135 – 00120 ( 2021 ). OpenUrl CrossRef PubMed 2. ↵ J. J. Morris , R. E. Lenski , E. R. Zinser , The Black Queen Hypothesis: Evolution of dependencies through adaptive gene loss . mBio 3 , e00036 – 00012 ( 2012 ). OpenUrl CrossRef PubMed 3. ↵ M. Ghoul , A. S. Griffin , S. A. West , Toward an evolutionary definition of cheating . Evolution 68 , 318 – 331 ( 2014 ). OpenUrl CrossRef PubMed 4. ↵ G. D’Souza et al. , Ecology and evolution of metabolic cross-feeding interactions in bacteria . Nat Prod Rep 35 , 455 – 488 ( 2018 ). OpenUrl CrossRef PubMed 5. ↵ S. Pande , C. Kost , Bacterial unculturability and the formation of intercellular metabolic networks . Trends Microbiol . 25 , 349 – 361 ( 2017 ). OpenUrl CrossRef PubMed 6. ↵ S. Zamenhof , H. H. Eichhorn , Study of microbial evolution through loss of biosynthetic functions: establishment of “defective” mutants . Nature 216 , 456 – 458 ( 1967 ). OpenUrl CrossRef PubMed 7. G. D’Souza et al. , Less is More: Selective advantages can explain the prevalent loss of biosynthetic genes in bacteria . Evolution 68 , 2559 – 2570 ( 2014 ). OpenUrl CrossRef PubMed 8. D. Dykhuizen , Selection for tryptophan auxotrophs of Escherichia coli in glucose-limited chemostats as a test of the energy conservation hypothesis of evolution . Evolution 32 , 125 – 150 ( 1978 ). OpenUrl CrossRef PubMed Web of Science 9. S. Pande et al. , Fitness and stability of obligate cross-feeding interactions that emerge upon gene loss in bacteria . ISME J 8 , 953 – 962 ( 2013 ). OpenUrl PubMed 10. ↵ G. D’Souza , C. Kost , Experimental evolution of metabolic dependency in bacteria . PLoS Genet . 12 , e1006364 ( 2016 ). OpenUrl CrossRef PubMed 11. ↵ O. F. Schakel , R. K. Fritts , A. J. Zmuda , S. Setayeshgar , J. B. McKinlay , Microbial cross-feeding stabilized by segregation of a dependent mutant from its independent ancestor . ISME J 19 , wraf131 ( 2025 ). OpenUrl PubMed 12. ↵ F. Harrison , J. Paul , R. C. Massey , A. Buckling , Interspecific competition and siderophore-mediated cooperation in Pseudomonas aeruginosa . ISME J 2 , 49 – 55 ( 2008 ). OpenUrl CrossRef PubMed Web of Science 13. ↵ Z. Dumas , R. Kümmerli , Cost of cooperation rules selection for cheats in bacterial metapopulations . J. Evol. Biol . 25 , 473 – 484 ( 2012 ). OpenUrl CrossRef PubMed 14. ↵ A. R. T. Figueiredo , A. Wagner , R. Kümmerli , Ecology drives the evolution of diverse social strategies in Pseudomonas aeruginosa . Mol. Ecol . 30 , 5214 – 5228 ( 2021 ). OpenUrl CrossRef 15. ↵ T. N. Nguyen , Q. G. Phan , L. P. Duong , K. P. Bertrand , R. E. Lenski , Effects of carriage and expression of the Tn10 tetracycline-resistance operon on the fitness of Escherichia coli K12 . Mol. Biol. Evol . 6 , 213 – 225 ( 1989 ). OpenUrl CrossRef PubMed Web of Science 16. ↵ F. C. Neidhardt , J. L. Ingraham , M. Schaechter , Physiology of the bacterial cell: a molecular approach ( Sinauer Associates Inc , Sunderland, MA , 1990 ). 17. M. Lynch , G. K. Marinov , The bioenergetic costs of a gene . Proc Natl Acad Sci U S A 112 , 15690 – 15695 ( 2015 ). OpenUrl Abstract / FREE Full Text 18. D. M. Stoebel , A. M. Dean , D. E. Dykhuizen , The cost of expression of Escherichia coli lac operon proteins is in the process, not in the products . Genetics 178 , 1653 – 1660 ( 2008 ). OpenUrl Abstract / FREE Full Text 19. E. Dekel , U. Alon , Optimality and evolutionary tuning of the expression level of a protein . Nature 436 , 588 – 592 ( 2005 ). OpenUrl CrossRef PubMed Web of Science 20. ↵ Z. Wu et al. , Expression level is a major modifier of the fitness landscape of a protein coding gene . Nat Ecol Evol 6 , 103 – 115 ( 2022 ). OpenUrl PubMed 21. ↵ B. LaSarre , A. L. McCully , J. T. Lennon , J. B. McKinlay , Microbial mutualism dynamics governed by dose-dependent toxicity of cross-fed nutrients . ISME J 11 , 337 – 348 ( 2017 ). OpenUrl CrossRef PubMed 22. ↵ R. K. Fritts et al. , Enhanced nutrient uptake is sufficient to drive emergent cross-feeding between bacteria in a synthetic community . ISME J 14 , 2816 – 2828 ( 2020 ). OpenUrl CrossRef PubMed 23. ↵ A. L. McCully et al. , An Escherichia coli nitrogen starvation response Is important for mutualistic coexistence with Rhodopseudomonas palustris . Appl and Environ Microbiol 84 , e00404 – 00418 ( 2018 ). OpenUrl PubMed 24. ↵ A. L. McCully , B. LaSarre , J. B. McKinlay , Recipient-biased competition for an intracellularly generated cross-fed nutrient is required for coexistence of microbial mutualists . mBio 8 , e01620 – 01617 ( 2017 ). OpenUrl CrossRef PubMed 25. ↵ A. L. McCully , B. LaSarre , J. B. McKinlay , Growth-independent cross-feeding modifies boundaries for coexistence in a bacterial mutualism . Environ. Microbiol . 19 , 3538 – 3550 ( 2017 ). OpenUrl CrossRef 26. ↵ Y. C. Chuang et al. , Bacterial adenine cross-feeding stems from a purine salvage bottleneck . ISME J 18 , wrae034 ( 2024 ). OpenUrl PubMed 27. ↵ B. LaSarre , A. M. Deutschbauer , C. E. Love , J. B. McKinlay , Covert Cross-Feeding Revealed by Genome-Wide Analysis of Fitness Determinants in a Synthetic Bacterial Mutualism . Applied and Environmental Microbiology 86 , e00543 – 00520 ( 2020 ). OpenUrl PubMed 28. ↵ J. B. McKinlay , C. S. Harwood , Carbon dioxide fixation as a central redox cofactor recycling mechanism in bacteria . Proc Natl Acad Sci U S A 107 , 11669 – 11675 ( 2010 ). OpenUrl Abstract / FREE Full Text 29. ↵ P. L. Foster , H. Lee , E. Popodi , J. P. Townes , H. Tang , Determinants of spontaneous mutation in the bacterium Escherichia coli as revealed by whole-genome sequencing . Proc Natl Acad Sci U S A 112 , E5990 – 5999 ( 2015 ). OpenUrl Abstract / FREE Full Text 30. ↵ H. Lee , E. Popodi , H. Tang , P. L. Foster , Rate and molecular spectrum of spontaneous mutations in the bacterium Escherichia coli as determined by whole-genome sequencing . Proc Natl Acad Sci U S A 109 , E2774 – 2783 ( 2012 ). OpenUrl Abstract / FREE Full Text 31. ↵ K. R. Fixen , C. S. Harwood , A polymorphism in the oxygen-responsive repressor PpsR2 confers a growth advantage to Rhodopseudomonas palustris under low light . Photosynth Res 129 , 199 – 204 ( 2016 ). OpenUrl CrossRef PubMed 32. ↵ J. B. McKinlay , C. S. Harwood , Calvin cycle flux, pathway constraints, and substrate oxidation state together determine the H2 biofuel yield in photoheterotrophic bacteria . mBio 2 , e00323 – 00310 ( 2011 ). OpenUrl CrossRef PubMed 33. ↵ A. A. Sangani , A. L. McCully , B. LaSarre , J. B. McKinlay , Fermentative Escherichia coli makes a substantial contribution to H 2 production in coculture with phototrophic Rhodopseudomonas palustris . FEMS Microbiol. Lett . 366 , fnz162 ( 2019 ). OpenUrl PubMed 34. ↵ F. E. Rey , Y. Oda , C. S. Harwood , Regulation of uptake hydrogenase and effects of hydrogen utilization on gene expression in Rhodopseudomonas palustris . J. Bacteriol . 188 , 6143 – 6152 ( 2006 ). OpenUrl Abstract / FREE Full Text 35. ↵ K. B. Pechter et al. , Molecular basis of bacterial longevity . mBio 8 , e01726 – 01717 ( 2017 ). OpenUrl PubMed 36. ↵ F. W. Larimer et al. , Complete genome sequence of the metabolically versatile photosynthetic bacterium Rhodopseudomonas palustris . Nat. Biotechnol . 22 , 55 – 61 ( 2004 ). OpenUrl CrossRef PubMed Web of Science 37. ↵ O. Baars , F. M. M. Morel , X. Zhang , The purple non-sulfur bacterium Rhodopseudomonas palustris produces novel petrobactin-related siderophores under aerobic and anaerobic conditions . Environ. Microbiol . 20 , 1667 – 1676 ( 2018 ). OpenUrl CrossRef PubMed 38. ↵ A. E. Kazakov et al. , Comparative genomics of regulation of fatty acid and branched-chain amino acid utilization in proteobacteria . J. Bacteriol . 191 , 52 – 64 ( 2009 ). OpenUrl Abstract / FREE Full Text 39. ↵ D. Preussger , A. Herbig , C. Kost , Synergistic coevolution accelerates genome evolution . bioRxiv , doi: 10.1101/2021.1105.1119.444833 ( 2021 ). OpenUrl CrossRef 40. ↵ C. Petersen , L. B. Møller , P. Valentin-Hansen , The cryptic adenine deaminase gene of Escherichia coli . Silencing by the nucleoid-associated DNA-binding protein, H-NS, and activation by insertion elements . J. Biol. Chem . 277 , 31373 – 31380 ( 2002 ). OpenUrl Abstract / FREE Full Text 41. ↵ K. F. Jensen , The Escherichia coli K-12 “wild types” W3110 and MG1655 have an rph frameshift mutation that leads to pyrimidine starvation due to low pyrE expression levels . J. Bacteriol . 175 , 3401 – 3407 ( 1993 ). OpenUrl Abstract / FREE Full Text 42. ↵ A. S. Bagnara , L. R. Finch , The effects of bases and nucleosides on the intracellular contents of nucleotides and 5-phosphoribosyl 1-pyrophosphate in Escherichia coli . Eur. J. Biochem . 41 , 421 – 430 ( 1974 ). OpenUrl PubMed 43. ↵ Y. C. Chuang , J. B. McKinlay , Metabolite toxicity as a driver of bacterial metabolite externalization . bioRxiv , doi: 10.64898/62026.64801.64817.700106 ( 2026 ). OpenUrl CrossRef 44. ↵ T. N. Grainger , J. M. Levine , B. Gilbert , The invasion criterion: a common currency for ecological research . Trends Ecol. Evol . 34 , 925 – 935 ( 2019 ). OpenUrl PubMed 45. ↵ S. P. Hammarlund , J. M. Chacón , W. R. Harcombe , A shared limiting resource leads to competitive exclusion in a cross-feeding system . Environ. Microbiol . 21 , 759 – 771 ( 2019 ). OpenUrl 46. ↵ P. Chesson , Mechanisms of maintenance of species diversity . Annu Rev Ecol Syst 31 , 343 – 366 ( 2000 ). OpenUrl CrossRef Web of Science 47. ↵ J. B. McKinlay , Are Bacteria Leaky? Mechanisms of Metabolite Externalization in Bacterial Cross-Feeding . Annu. Rev. Microbiol . 77 , 277 – 297 ( 2023 ). OpenUrl CrossRef PubMed 48. S. Sulheim et al. , Microbes use transporters to regulate the release of metabolites based on value . bioRxiv , doi: 10.1101/2025.1108.1119.671024 ( 2025 ). OpenUrl CrossRef 49. ↵ C. M. Jones , N. J. Hernandez Lozada , B. F. Pfleger , Efflux systems in bacteria and their metabolic engineering applications . Appl Microbiol Biotechnol 99 , 9381 – 9393 ( 2015 ). OpenUrl CrossRef 50. ↵ D. Satory , J. A. Halliday , P. Sivaramakrishnan , R. C. Lua , C. Herman , Characterization of a novel RNA polymerase mutant that alters DksA activity . J Bacteriol 195 , 4187 – 4194 ( 2013 ). OpenUrl Abstract / FREE Full Text 51. ↵ G. Yousif et al. , Obligate cross-feeding of metabolites is common in soil microbial communities . bioRxiv , doi: 10.1101/2025.1101.1129.635426 ( 2025 ). OpenUrl CrossRef 52. M. T. Mee , J. J. Collins , G. M. Church , H. H. Wang , Syntrophic exchange in synthetic microbial communities . Proc Natl Acad Sci U S A 111 , E2149 – 2156 ( 2014 ). OpenUrl Abstract / FREE Full Text 53. S. Giri , G. Yousif , S. Shitut , L. Ona , C. Kost , Prevalent emergence of reciprocity among cross-feeding bacteria . ISME Commun 2 , 71 ( 2022 ). 54. S. Giri et al. , Metabolic dissimilarity determines the establishment of cross-feeding interactions in bacteria . Curr. Biol . 31 , 5547 – 5557 e5546 ( 2021 ). OpenUrl CrossRef PubMed 55. ↵ E. H. Wintermute , P. A. Silver , Emergent cooperation in microbial metabolism . Mol. Syst. Biol . 6 , 407 ( 2010 ). 56. ↵ J. L. Snoep et al. , Differences in sensitivity to NADH of purified pyruvate dehydrogenase complexes of Enterococcus faecalis , Lactococcus lactis , Azotobacter vinelandii and Escherichia coli : implications for their activity in vivo . FEMS Microbiol. Lett . 114 , 279 – 283 ( 1993 ). OpenUrl CrossRef PubMed 57. ↵ C. Beloin , A. Roux , J. M. Ghigo , Escherichia coli biofilms . Curr Top Microbiol Immunol 322 , 249 – 289 ( 2008 ). OpenUrl CrossRef PubMed 58. ↵ C. Prigent-Combaret et al. , Complex regulatory network controls initial adhesion and biofilm formation in Escherichia coli via regulation of the csgD gene . J. Bacteriol . 183 , 7213 – 7223 ( 2001 ). OpenUrl Abstract / FREE Full Text 59. ↵ E. R. Holden et al. , Massively parallel transposon mutagenesis identifies temporally essential genes for biofilm formation in Escherichia coli . Microb Genom 7 , 000673 ( 2021 ). 60. ↵ C. S. Barker , B. M. Pruss , P. Matsumura , Increased motility of Escherichia coli by insertion sequence element integration into the regulatory region of the flhD operon . J. Bacteriol . 186 , 7529 – 7537 ( 2004 ). OpenUrl Abstract / FREE Full Text 61. ↵ E. Krin , A. Danchin , O. Soutourina , Decrypting the H-NS-dependent regulatory cascade of acid stress resistance in Escherichia coli . BMC Microbiol . 10 , 273 ( 2010 ). 62. ↵ E. C. Hobbs , J. L. Astarita , G. Storz , Small RNAs and small proteins involved in resistance to cell envelope stress and acid shock in Escherichia coli : analysis of a bar-coded mutant collection . J. Bacteriol . 192 , 59 – 67 ( 2010 ). OpenUrl Abstract / FREE Full Text 63. ↵ A. K. Sayed , J. W. Foster , A 750 bp sensory integration region directs global control of the Escherichia coli GadE acid resistance regulator . Mol. Microbiol . 71 , 1435 – 1450 ( 2009 ). OpenUrl CrossRef PubMed Web of Science 64. ↵ Y. Gao et al. , Unraveling the functions of uncharacterized transcription factors in Escherichia coli using ChIP-exo . Nucleic Acids Res . 49 , 9696 – 9710 ( 2021 ). OpenUrl CrossRef PubMed 65. ↵ B. E. Mazny , O. F. Sheff , B. LaSarre , A. McKinlay , J. B. McKinlay , Complete genome sequence of Rhodopseudomonas palustris CGA0092 and corrections to the R. palustris CGA009 genome sequence . Microbiol Resour Announc 12 , e0128522 ( 2023 ). OpenUrl CrossRef PubMed 66. ↵ J. Quandt , M. F. Hynes , Versatile suicide vectors which allow direct selection for gene replacement in Gram-negative bacteria . Gene 127 , 15 – 21 ( 1993 ). OpenUrl CrossRef PubMed Web of Science 67. ↵ F. E. Rey , E. K. Heiniger , C. S. Harwood , Redirection of metabolism for biological hydrogen production . Appl Environ Microbiol 73 , 1665 – 1671 ( 2007 ). OpenUrl Abstract / FREE Full Text 68. ↵ M.-K. Kim , C. S. Harwood , Regulation of benzoate-CoA ligase in Rhodopseudomonas palustris . FEMS Microbiol. Lett . 83 , 199 – 203 ( 1991 ). OpenUrl CrossRef Web of Science 69. ↵ B. LaSarre , R. Morlen , G. C. Neumann , C. S. Harwood , J. B. McKinlay , Nitrous oxide reduction by two partial denitrifying bacteria requires denitrification intermediates that cannot be respired . Appl Environ Microbiol 90 , e0174123 ( 2024 ). OpenUrl CrossRef PubMed 70. ↵ K. A. Datsenko , B. L. Wanner , One-step inactivation of chromosomal genes in Escherichia coli K-12 using PCR products . Proc. Natl. Acad. Sci. USA 97 , 6640 – 6645 ( 2000 ). OpenUrl Abstract / FREE Full Text 71. ↵ T. Baba et al. , Construction of Escherichia coli K-12 in-frame, single-gene knockout mutants: the Keio collection . Mol. Syst. Biol . 2 , 2006.0008 ( 2006 ). 72. ↵ J. J. Huang , E. K. Heiniger , J. B. McKinlay , C. S. Harwood , Production of hydrogen gas from light and the inorganic electron donor thiosulfate by Rhodopseudomonas palustris . Appl Environ Microbiol 76 , 7717 – 7722 ( 2010 ). OpenUrl Abstract / FREE Full Text 73. ↵ M. Martin , Cutadapt removes adapter sequences from high-throughput sequencing reads . EMBnet j 17 ( 2011 ). 74. ↵ D. E. Deatherage , J. E. Barrick , “ Identification of mutations in laboratory-evolved microbes from next-generation sequencing data using breseq ” in Engineering and Analyzing Multicellular Systems: Methods and Protocols , L. Sun , W. Shou , Eds. ( Springer New York , New York, NY , 2014 ) , doi: 10.1007/978-1-4939-0554-6_12 , pp. 165 – 188 . OpenUrl CrossRef PubMed Web of Science View the discussion thread. Back to top Previous Next Posted February 17, 2026. Download PDF Supplementary Material Email Thank you for your interest in spreading the word about bioRxiv. NOTE: Your email address is requested solely to identify you as the sender of this article. Your Email * Your Name * Send To * Enter multiple addresses on separate lines or separate them with commas. You are going to email the following Reciprocal cross-feeding between bacteria can limit the emergence of metabolic dependencies Message Subject (Your Name) has forwarded a page to you from bioRxiv Message Body (Your Name) thought you would like to see this page from the bioRxiv website. Your Personal Message CAPTCHA This question is for testing whether or not you are a human visitor and to prevent automated spam submissions. Share Reciprocal cross-feeding between bacteria can limit the emergence of metabolic dependencies Ying-Chih Chuang , Megan G. Behringer , Gillian Patton , Jordan T. Bird , Jeffrey L. Mazny , Jennifer R. Gliessman , Crystal E. Love , Ankur Dalia , James B. McKinlay bioRxiv 2024.08.19.608702; doi: https://doi.org/10.1101/2024.08.19.608702 Share This Article: Copy Citation Tools Reciprocal cross-feeding between bacteria can limit the emergence of metabolic dependencies Ying-Chih Chuang , Megan G. Behringer , Gillian Patton , Jordan T. Bird , Jeffrey L. Mazny , Jennifer R. Gliessman , Crystal E. Love , Ankur Dalia , James B. McKinlay bioRxiv 2024.08.19.608702; doi: https://doi.org/10.1101/2024.08.19.608702 Citation Manager Formats BibTeX Bookends EasyBib EndNote (tagged) EndNote 8 (xml) Medlars Mendeley Papers RefWorks Tagged Ref Manager RIS Zotero Tweet Widget Facebook Like Google Plus One Subject Area Microbiology Subject Areas All Articles Animal Behavior and Cognition (7644) Biochemistry (17728) Bioengineering (13916) Bioinformatics (42037) Biophysics (21489) Cancer Biology (18637) Cell Biology (25553) Clinical Trials (138) Developmental Biology (13401) Ecology (19941) Epidemiology (2067) Evolutionary Biology (24367) Genetics (15622) Genomics (22547) Immunology (17764) Microbiology (40475) Molecular Biology (17208) Neuroscience (88747) Paleontology (667) Pathology (2842) Pharmacology and Toxicology (4834) Physiology (7659) Plant Biology (15175) Scientific Communication and Education (2047) Synthetic Biology (4304) Systems Biology (9835) Zoology (2272)