Metabolic imbalance limits fermentation in microbes engineered for high-titer ethanol production

Metabolic imbalance limits fermentation in microbes engineered for high-titer ethanol production | 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 Metabolic imbalance limits fermentation in microbes engineered for high-titer ethanol production Bishal Dev Sharma , Eashant Thusoo , David M. Stevenson , View ORCID Profile Daniel Amador-Noguez , Lee R. Lynd , View ORCID Profile Daniel G. Olson doi: https://doi.org/10.1101/2025.11.21.689677 Bishal Dev Sharma 1 Thayer School of Engineering, Dartmouth College , Hanover, New Hampshire, USA 2 Center for Bioenergy Innovation, Oak Ridge National Laboratory , Oak Ridge, Tennessee, USA Find this author on Google Scholar Find this author on PubMed Search for this author on this site Eashant Thusoo 2 Center for Bioenergy Innovation, Oak Ridge National Laboratory , Oak Ridge, Tennessee, USA 3 Department of Bacteriology, University of Wisconsin-Madison , Madison, Wisconsin, USA 4 Great Lakes Bioenergy Research Center, University of Wisconsin-Madison , Madison, WI, USA Find this author on Google Scholar Find this author on PubMed Search for this author on this site David M. Stevenson 2 Center for Bioenergy Innovation, Oak Ridge National Laboratory , Oak Ridge, Tennessee, USA 3 Department of Bacteriology, University of Wisconsin-Madison , Madison, Wisconsin, USA 4 Great Lakes Bioenergy Research Center, University of Wisconsin-Madison , Madison, WI, USA Find this author on Google Scholar Find this author on PubMed Search for this author on this site Daniel Amador-Noguez 2 Center for Bioenergy Innovation, Oak Ridge National Laboratory , Oak Ridge, Tennessee, USA 3 Department of Bacteriology, University of Wisconsin-Madison , Madison, Wisconsin, USA 4 Great Lakes Bioenergy Research Center, University of Wisconsin-Madison , Madison, WI, USA Find this author on Google Scholar Find this author on PubMed Search for this author on this site ORCID record for Daniel Amador-Noguez Lee R. Lynd 1 Thayer School of Engineering, Dartmouth College , Hanover, New Hampshire, USA 2 Center for Bioenergy Innovation, Oak Ridge National Laboratory , Oak Ridge, Tennessee, USA 5 Terragia Biofuels Inc. , Hanover, NH, USA Find this author on Google Scholar Find this author on PubMed Search for this author on this site Daniel G. Olson 1 Thayer School of Engineering, Dartmouth College , Hanover, New Hampshire, USA 2 Center for Bioenergy Innovation, Oak Ridge National Laboratory , Oak Ridge, Tennessee, USA Find this author on Google Scholar Find this author on PubMed Search for this author on this site ORCID record for Daniel G. Olson For correspondence: daniel.g.olson{at}dartmouth.edu Abstract Full Text Info/History Metrics Preview PDF ABSTRACT Microbial strains engineered for high-titer ethanol production achieve lower maximum titers compared to native producers such as Zymomonas mobilis . A central unresolved question is why fermentation ceases before substrate has been exhausted by these strains. Here, we integrate metabolite profiling with thermodynamic analysis to examine this phenomenon in engineered strains of Escherichia coli and Thermoanaerobacterium saccharolyticum and compare them to Z. mobilis , a native ethanol producer. In the engineered strains, fermentation cessation coincided with marked pyruvate accumulation, due to a lack of ability to convert pyruvate to ethanol. This resulted in a local thermodynamic equilibrium at the pyruvate kinase reaction, as determined by Max-Min Driving Force (MDF) analysis. Relaxing constraints on pyruvate and related metabolites restored positive MDF values, implicating thermodynamic limitations as the underlying constraint. By contrast, Z. mobilis maintained a positive thermodynamic driving force throughout fermentation, suggesting that product titer is limited by a different mechanism in this organism. These findings establish a systems-level framework linking metabolite concentrations to pathway thermodynamics and highlight opportunities for improving microbial performance in ethanol and other bioproduction contexts. 1. INTRODUCTION Fossil fuels have been a central energy source for global industrialization and transportation for centuries. However, growing concerns about their limited supply and their significant contribution to climate change have highlighted the need for a transition to sustainable fuel sources. While electric vehicles are making progress, dense liquid fuels remain essential, particularly for long-haul trucking, ocean shipping, and aviation ( Fulton et al., 2015 ; Lynd et al., 2022 ). Among the options for renewable liquid fuels that can be produced from cellulose, ethanol remains one of the most promising, due to its relatively low toxicity to microorganisms an, ability to be produced at high yield and titer, and ease of separation by distillation ( Lynd et al., 2022 ). For a standalone production facility, ethanol titers need to be in the range of 40-50 g/L to reduce costs associated with fermenter size and distillation ( Dien et al., 2003 ; Lynd et al., 2022 ). Several efforts have been made to engineer microbes capable of producing ethanol at industrially relevant titers. To date, only three microbes — E. coli , T. saccharolyticum, and Corynebacterium glutamicum — which do not naturally produce ethanol as a major product, have been successfully engineered to achieve ethanol titers higher than 50 g/L ( Herring et al., 2016 ; Jojima et al., 2015 ; Yomano et al., 1998 ). The engineering of E. coli involved deletion of genes responsible for the production of other fermentation products like acetate and lactate, along with the deletion of the native alcohol dehydrogenase gene and the introduction of genes for pyruvate decarboxylase and alcohol dehydrogenase from Zymomonas mobilis ( Yomano et al., 1998 ). The same pathway was introduced into C. glutamicum ( Inui et al., 2004 ). Subsequently, overexpression of glycolytic genes and optimization of fermentation conditions resulted in ethanol production at a titer of 106 g/L ( Jojima et al., 2015 ). Similarly, for T. saccharolyticum , engineering involved the deletion of genes for competing products like lactate and acetate ( Herring et al., 2016 ). Unlike these engineered strains, Z. mobilis is a natural ethanologen and the most productive bacterial ethanol producer, capable of achieving ethanol titers up to 127 g/L in batch fermentation ( Rogers et al., 1982 , 2007 ), while retaining high yield and productivity, making it an important benchmark for microbial ethanologen design ( Kalnenieks, 2006 ; Rogers et al., 2007 ; Sprenger, 1996 ). However, despite its efficiency, Z. mobilis has limited substrate range (i.e. it only ferments glucose, fructose, and sucrose) and low thermotolerance, which constrains its use for lignocellulosic biofuel production. Nevertheless, its physiology and metabolic characteristics continue to provide valuable insights for the study and engineering of ethanologenic microbes. The ability to transfer phenotypes from one organism to another organism is a cornerstone of modern biotechnology. Ethanol production is one of the first phenotypes for which this was demonstrated ( Ingram et al., 1987 ). Although it is relatively trivial to engineer an organism to produce small amounts of ethanol, it is much more difficult to transfer the industrially-relevant phenotype of ethanol production at the high yield, titer, and rate that is present in native producers such as Z. mobilis . In this work, we identified fermentation conditions where ethanol production stops despite the presence of residual substrate for E. coli , T. saccharolyticum , and Z. mobilis . Using these fermentation conditions, we then quantified changes in glycolytic metabolite concentrations to observe the dynamic changes to metabolism as growth and fermentation stop. Finally, we analyzed the extent to which thermodynamic equilibrium limits product titer using the Max-Min Driving Force (MDF) framework ( Noor et al., 2014 ). By studying why ethanol production stops in engineered ethanol producers such as E. coli and T. saccharolyticum , and comparing with native producers such as Z. mobilis , we characterize the extent to which engineered strains recapitulate the industrially relevant ethanol production phenotype, as well as the mechanistic basis for existing limitations to product titer. We expect that the improved understanding of high-titer ethanol production developed in this work will build a strong foundation for efforts to transfer this phenotype to other organisms (e.g., Clostridium thermocellum ) or even whole microbial communities, and will also inform strategies for producing other bioproducts at high yield and titer. 2. MATERIALS AND METHODS 2.1 Strains used in this work Strains used in this work are shown in Table 1 . View this table: View inline View popup Download powerpoint Table 1: Strains used in this study . 2.2 Media and growth conditions All reagents used in this study were of molecular grade and obtained from Sigma Aldrich or Fisher Scientific, unless otherwise noted. E. coli cultures were grown anaerobically at 37°C in either Luria Broth medium (BP1427-500, Fisher Scientific) or M9 minimal medium modified to contain calcium and iron ( Sharma et al., 2023 ) with addition of betaine. M9 minimal medium contained a final concentration of 120 g/L glucose, 6.7 g/L Na 2 HPO 4 , 3 g/L KH 2 PO 4 , 0.5 g/L NaCl, 1 g/L NH 4 Cl, 0.246 g/L MgSO 4 , 0.0011 g/L CaCl 2 , 0.0002 g/L FeSO 4 ·7H 2 O, 0.1536 g/L betaine hydrochloride and 50 µg/mL spectinomycin. The medium components were prepared as separate solutions. M9 salts, 5-fold concentrated contained Na 2 HPO 4 ·7H 2 O, KH 2 PO 4 , NaCl and NH 4 Cl (M6030-1KG, Sigma-Aldrich). MgSO 4 , CaCl 2 , FeSO 4 ·7H 2 O and betaine hydrochloride solutions were prepared 500-fold, 100-fold, 100-fold and 1000-fold concentrated respectively. All the solutions were filter-sterilized and stored under nitrogen atmosphere. 2000-fold concentrated spectinomycin was obtained from Sigma-Aldrich (catalog no. S0692-1ML). T. saccharolyticum cultures were grown anaerobically at 51°C in either TSC6 medium or MTC-7 medium with or without yeast extract. The TSC6 rich medium contained a final concentration of 90 g/L cellobiose and 60 g/L maltodextrin, 10 g/L calcium carbonate, 8.5 g/L yeast extract, 0.5 g/L trisodium citrate·2H 2 O, 2.0 g/L KH 2 PO 4 , 2.0 g/L MgSO 4 ·7H 2 O, 5 g/L urea, 0.2 g/L CaCl 2 ·2H2O, 0.2 g/L FeSO 4 ·7H 2 O, 0.12 g/L Methionine, 0.5 g/L L-Cysteine HCl as described previously ( Cui et al., 2019 ; Herring et al., 2016 ). The MTC-7 chemically defined medium contained a final concentration of 140 g/L cellobiose, 9.3 g/L MOPS (morpholinepropanesulfonic acid) sodium salt, 2 g/L potassium citrate monohydrate, 1.3 g/L citric acid monohydrate, 1 g/L Na 2 SO 4 , 1 g/L KH 2 PO 4 , 2.5 g/L NaHCO 3 , 2 g/L urea, 1 g/L MgCl 2 ·6H 2 O, 0.2 g/L CaCl 2 ·2H 2 O, 0.1 g/L FeCl 2 ·4H 2 O, 1 g/L L-cysteine HCl monohydrate, 0.02 g/L pyridoxamine HCl, 0.004 g/L p-aminobenzoic acid, 0.002 g/L D-biotin, 0.002 g/L Vitamin B12, 0.004 g/L thiamine, 0.0005 g/L MnCl 2 ·4H 2 O, 0.0005 g/L CoCl 2 ·6H 2 O, 0.0002 g/L ZnCl 2 , 0.0001 g/L CuCl 2 ·2H 2 O, 0.0001 g/L H 3 BO 3 , 0.0001 g/L Na 2 MoO 4 ·2H 2 O, 0.0001 g/L NiCl 2 ·6H 2 O as described previously ( Sharma et al., 2023 , 2024 ). The medium components were prepared as separate solutions, A-F. Solution A, concentrated 1.2-fold, contained cellobiose and MOPS sodium salt. Solution B, concentrated 25-fold, contained potassium citrate monohydrate, citric acid monohydrate, Na 2 SO 4 , KH 2 PO 4 and NaHCO 3 . Solution C, concentrated 25-fold, contained urea. Solution D, concentrated 50-fold, contained MgCl 2 ·6H 2 O, CaCl 2 ·2H 2 O, FeCl 2 ·4H 2 O, L-cysteine HCl monohydrate and trace minerals (solution F). Solutions E, concentrated 50-fold, contained pyridoxamine HCl, p-aminobenzoic acid, D-biotin, vitamin B12 and thiamine. Solution F, concentrated 1000-fold, contained MnCl 2 ·4H 2 O, CoCl 2 ·6H 2 O, ZnCl 2 , CuCl 2 ·2H 2 O, H 3 BO 3 , Na 2 MoO 4 ·2H 2 O and NiCl 2 ·6H 2 O. For medium without active pH control, the initial pH of the medium was maintained at 6.5 using 10% H 2 SO 4 . For medium with yeast extract, 8.5 g/L yeast extract was added to MTC-7. Note: 140 g/L is near the upper limit of solubility for cellobiose but can solubilize at concentrations as high as 200 g/L when autoclaved. Z. mobilis cultures were grown anaerobically at 37°C in Zymomonas mobilis minimal (ZMM-2) medium with addition of methionine and lysine to ZMM medium ( Jacobson et al., 2019 ). ZMM medium contained a final concentration of 20-160 g/L glucose, 1 g/L KH 2 PO 4 , 1 g/L K 2 HPO 4 , 0.5 g/L NaCl, 1 g/L (NH 4 ) 2 SO 4 , 0.2 g/L MgSO 4 ·7H 2 O, 0.025 g/L NaMoO 4 ·2H 2 O, 0.0025 g/L FeSO 4 ·7H 2 O and 0.02 g/L calcium pantothenate. ZMM-2 had additional 2 g/L methionine and 2 g/L lysine. The medium components were prepared as separate solutions. ZMM salts, concentrated 10-fold, contained KH 2 PO 4 , K 2 HPO 4 , NaCl and (NH 4 ) 2 SO 4 . All other components except glucose were concentrated 1000-fold. Concentrated glucose solution of 400 g/L was prepared by filter sterilization. For the growth medium without active pH control, the initial pH of the medium was maintained at 6.0 using 10% H 2 SO 4 . The medium remained turbid until adjusted to pH 6.0, due to the limited solubility of lysine and methionine at higher pH. All media were filter-sterilized after combining the component solutions. 2.3 pH-controlled bioreactor fermentation conditions T. saccharolyticum cultures were grown anaerobically at 51°C in 400 mL MTC-7 medium with 140 g/L cellobiose as substrate with pH controlled at 6.0 ± 0.05. For a 400-mL working volume with pH control in a bioreactor, 300 mL of solution A was prepared excluding MOPS sodium salt and autoclaved (Note: Although the solubility limit of cellobiose is reported to be ∼120 g/L, autoclaving increases solubility and allows preparation of cellobiose solutions at concentrations up to 200 g/L). The bioreactor was placed in an anaerobic chamber for 12 to 16 hours, with the gas inlet/outlet tube left open for purging. Then, 16 mL of solution B, 8 mL of solution C, 8 mL of solution D, and 16 mL of solution E were added through a 0.22-μm filter (catalog no. 430517; Corning). The final volume was adjusted to 400 mL using autoclaved distilled water stored in a nitrogen atmosphere. E. coli cultures were grown anaerobically at 37°C in 400 mL M9 minimal medium with 120 g/L glucose as substrate with pH controlled at 6.5 ± 0.05. For a 400-mL working volume with pH control in a bioreactor, the bioreactor was filled with 160 mL water and autoclaved. The bioreactor was placed in an anaerobic chamber for 12 to 16 hours, with the gas inlet/outlet tube left open for purging. Then, M9 minimal medium was prepared in 2-fold concentration and stored in the anaerobic chamber. After purging, 200 mL of 2-fold concentrated M9 minimal medium was added through a 0.22-μm syringe filter. The final volume was adjusted to 400 mL using autoclaved distilled water stored in a nitrogen atmosphere. Z. mobilis cultures were grown anaerobically at 37°C in 400 mL ZMM-2 medium with 160 g/L glucose as substrate and 10 g/L initial ethanol. The pH was controlled at 6.0 ± 0.05. For a 400-mL working volume with pH control in a bioreactor, the bioreactor was filled with 160 mL water and autoclaved. The bioreactor was placed in an anaerobic chamber for 12 to 16 hours, with the gas inlet/outlet tube left open for purging. Then, ZMM-2 medium was prepared in 2-fold concentration and stored in the anaerobic chamber. After purging, 200 mL of 2-fold concentrated ZMM-2 medium was added through a 0.22-μm syringe filter. The final volume was adjusted to 400 mL using autoclaved distilled water stored in a nitrogen atmosphere. All the bioreactor fermentations were performed in a Coy (Ann Arbor, MI) anaerobic chamber with a gas phase of 85% N 2 , 10% CO 2 , and 5% H 2 . The pH was maintained using a Mettler-Toledo (Columbus, OH, USA) pH probe (Catalog No. 405-DPAS-SC-K8S), and 4 N KOH was added as needed to control the pH. The bioreactor setup is shown in Figure S1. 2.4 Metabolites extraction and quantification During fermentation, the samples were taken at different time points. To quantify the fermentation products or extracellular metabolites (acetate, ethanol, formate, lactate, pyruvate), the cell culture was centrifuged and the supernatant was processed for HPLC analysis. The metabolites were quantified using HPLC (LC-2030, Shimadzu) with refractive index (RI) and UV detection using an Aminex HPX-87H column (Bio-Rad, Hercules, CA) . The column was maintained at 60°C. The mobile phase (5-mM sulfuric acid) flow rate was 0.6 mL/minute ( Holwerda et al., 2014 ; Sharma et al., 2024 ). To quantify glycolysis or intracellular metabolites, the cell samples were prepared using a previously published protocol optimized for high substrate fermentations ( Sharma et al., 2023 ). Metabolite extraction was performed in the anaerobic chamber. The metabolite extract was prepared by filtering the cell culture using vacuum filtration followed by placing the filter cell-side down for quenching in a cold metabolite extraction buffer (40% acetonitrile, 40% methanol, and 20% water) as previously described (Bennett et al 2008, Sharma et al 2023 , 2024 ). For each sample 0.5-8 mL of cells were filtered, quenched and dried under nitrogen gas using a sample concentrator (catalog no. EW-36620-40, Cole-Parmer). After quenching and extraction, the samples were removed from the anaerobic chamber. The samples were then resuspended in molecular-biology grade water and analyzed by LC-MS. Metabolite quantification was done using calibration curves made using known external standards. Total intracellular volume was calculated as: volume of a single cell (estimated from cell dimensions determined by microscopy) × optical density of the culture (OD 600 ) × cells per mL at OD 600 = 1 × volume of culture filtered, as previously described ( Sharma et al., 2023 ). The volume of a single cell was estimated from cell dimensions using the formula π × (radius) 2 × length. The calculated values were approximately 0.664 µm³ (6.6 × 10 -13 mL) for E. coli (BioNumber identification number [BNID] 106614 ( Milo et al., 2009 )), 0.503 µm³ (5.0 × 10 -13 mL) for T. saccharolyticum ( Lee et al., 1993 ), and 4.52 µm³ (4.5 × 10 -12 mL) for Z. mobilis ( Panesar et al., 2006 ). Cell counts at OD 600 = 1 were approximately 8.9 × 10 8 cells/mL for E. coli , 4.4 × 10 8 cells/mL for Thermoanaerobacterium saccharolyticum and 2.2 × 10 8 cells/mL for Zymomonas mobilis , as determined by cell counting. For metabolites without external standards, metabolites were identified by M/Z (mass/charge) ratio and retention time. For each metabolite, peak area values were normalized by calculating z-scores across all time points. The detailed protocol for intracellular metabolite extraction, peak detection and quantification of metabolites is available at protocols.io ( 10.17504/protocols.io.kqdg3wjbev25/v2 ) ( Sharma & Olson, 2025 ). 2.5 Liquid chromatography-mass spectrometry (LC-MS) analysis Metabolomics analysis was performed using a Vanquish ultra-high-performance liquid chromatography (UHPLC) system (Thermo Scientific), coupled to a hybrid quadrupole-Orbitrap™ mass spectrometer (Q Exactive™; Thermo Scientific) with electrospray ionization operating in negative-ion mode, as previously outlined ( Callaghan et al., 2023 ). The chromatography was conducted at 25°C using a 2.1 × 100 mm reverse-phase C18 column with a 1.7 μm particle size (Water™; Acquity UHPLC BEH). Two distinct chromatography gradients were employed. The first gradient used Solvent A (97:3 H2O: methanol + 10 mM tributylamine) and Solvent B (100% methanol) with the following time schedule: 0–2.5 min, 5% B; 2.5–17 min, a linear gradient from 5% B to 95% B; 17–19.5 min, 95% B; 19.5–20 min, a linear gradient from 95% B to 5% B; 20–25 min, 5% B. The second gradient also used Solvent A and B (100% methanol), with the following: 0–2.5 min, 5% B; 2.5–7.5 min, linear gradient from 5% B to 20% B; 7.5–13 min, 20% B to 55% B; 13–18.5 min, 55% B to 95% B; 18.5–19 min, linear gradient from 95% B to 5% B; 19–25 min, 5% B. The flow rate was kept at 0.2 mL/min for both gradients. Metabolites were identified based on their retention times, determined with pure standards, and their monoisotopic mass using MAVEN ( Clasquin et al., 2012 ) and El-MAVEN ( Agrawal et al., 2019 ) software. 2.6 Thermodynamic analysis Thermodynamic Max-Min Driving Force (MDF) analysis of the fermentation pathways in this study was performed using eQuilibrator ( Flamholz et al., 2012 ), implemented in python programming language ( https://gitlab.com/equilibrator ). For the thermodynamics analysis, the following assumptions were applied: For measured metabolites, the concentration bounds were fixed to the measured values. For non-quantified metabolites, concentrations were allowed to range from 1 µM to 100 mM. Although the standard MDF framework sets the default range from 1 µM to 10 mM ( Noor et al., 2014 ), we relaxed the upper bound because several intracellular metabolites in this study were measured at concentrations above 10 mM. In the model, quantified metabolites were assigned identical upper and lower bounds equal to their measured concentrations, while non-quantified metabolites were allowed to vary between 1 µM and 100 mM. For metabolites measured at a concentration of zero, the lower bound was set to 1 µM. All metabolite identifiers are from the KEGG database ( Kanehisa, 2000 , 2019 ; Kanehisa et al., 2025 ). 3. RESULTS 3.1 Yeast extract allows high titer ethanol production in microbes One of our primary goals in this work was to study factors that limit fermentation while substrate is still present. To do this, we first needed to identify suitable initial substrate concentrations. Furthermore, because yeast extract interferes with quantification of intracellular metabolites, we aimed to identify conditions where high ethanol titers could be achieved in its absence. Previous studies on engineered E. coli strains have reported ethanol titers ranging from 34 to 65 g/L ( Ingram et al., 1987 ; Jilani et al., 2017 ; Martinez et al., 2007 ; Ohta et al., 1991 ; Yomano et al., 1998 , 2008 ), primarily using E. coli B derivatives. Most of these studies used rich medium (Luria Broth) with only a few ( Ingram et al., 1987 ; Jilani et al., 2017 ; Martinez et al., 2007 ; Ohta et al., 1991 ; Yomano et al., 1998 , 2008 ) using chemically defined media like NBS or AM1. While E. coli B strains are widely used in ethanol studies, they are not as extensively characterized as the commonly used E. coli K-12 strain MG1655, which has a well-documented genetic and metabolic background ( Edwards & Palsson, 2000 ). Ethanol titers up to ∼40 g/L have been reported from E. coli MG1655 derivatives when growth in LB medium ( Trinh et al., 2008 ) whereas the highest reported titer for MG1655 derivatives in chemically defined medium is only ∼18 g/L ( Trinh & Srienc, 2009 ). In this study, we used an MG1655 derivative ( E. coli strain RL3019) and found that it could produce up to 43 g/L ethanol in LB medium ( Table 2 ). In contrast, when grown in a defined M9 minimal medium, the maximum ethanol titer decreased to ∼20.5 g/L ( Table 2 ). View this table: View inline View popup Table 2: Ethanol production by engineered strains of E. coli and T. saccharolyticum . ‘-’ represents no pH control . Similarly, engineered strains of T. saccharolyticum have been reported to produce 50-70 g/L ethanol in rich media such as TSC6 ( Cui et al., 2019 ; Herring et al., 2016 ). Using strain M1442, we observed ethanol titers up to 61 g/L when yeast extract was added to the defined MTC7 medium ( Table 2 ). In the absence of yeast extract, the maximum titer dropped to 44 g/L ( Table 2 ) in chemically defined growth medium ( Table 2 ). For Z. mobilis , included as a wild-type benchmark, we sought defined medium conditions that could support high ethanol titers without substrate limitation. Most prior studies of Z. mobilis in ZMM media have used relatively low glucose concentrations (<20 g/L) ( Carreón-Rodríguez et al., 2019 ; Jacobson et al., 2020 ). After confirming that Z. mobilis is auxotrophic for lysine and methionine ( Seo et al., 2005 ), we grew it in ZMM-2 medium supplemented with these amino acids which improved ethanol production from 8 g/L to 45 g/L ( Table 2 ). With active pH control, Z. mobilis was able to completely consume glucose up to ∼160 g/L (Additional File 1), but it was difficult to get the organism to reliably initiate growth at glucose concentrations above 160 g/L in ZMM-2 media. Ultimately, we identified a condition where Z. mobilis achieved a final ethanol titer of 72 g/L when grown on 160 g/L glucose with 10 g/L initial ethanol added. The addition of a small initial amount of ethanol was necessary to observe cessation of fermentation with substrate still remaining. It is interesting to note that yeast extract allows for high titer ethanol production in all of these microbes, even without pH control ( Table 2 ). Yeast extract is also thought to help recover damaged cell membranes due to ethanol toxicity, allowing microbes to tolerate higher ethanol titers ( Osman & Ingram, 1985 ). Yeast extract may also affect redox balance and biosynthesis of cofactors. However, how yeast extract helps microbes for better growth and fermentation is beyond the scope of the current investigation. Additionally, yeast extract can exhibit variability between batches due to differences in raw material sources, production methods, and processing conditions ( Tao et al., 2023 ). This lack of batch-to-batch consistency can lead to variations in its nutrient composition, including amino acids, peptides, and other bioactive compounds, which can cause batch effects that are difficult to control for. Furthermore, yeast extract interferes with LC-MS measurements of intracellular metabolites. Therefore, for our study, we proceeded using fermentation conditions with chemically defined media and with pH control. Although the final ethanol titers were lower than the maximum observed with complex media, these conditions allowed us to both observe the cessation of fermentation and to accurately quantify intracellular metabolites. The replicates of data shown in Table 2 are provided in Additional File-1 (fermentation_all_ext). The order of rows within the table reflects the approximate order in which experiments were performed, as we worked to identify suitable fermentation conditions. 3.2 Fermentation behavior of E. coli RL3019, T. saccharolyticum M1442 and Z. mobilis ZM4 To study fermentation behavior under conditions where substrate availability does not limit ethanol production, we performed high-substrate batch fermentations: 120 g/L glucose for E. coli , 140 g/L cellobiose for T. saccharolyticum , and 160 g/L glucose with 10 g/L initial ethanol for Z. mobilis . For the purposes of analysis, we divided the fermentations into phases: growth-coupled fermentation, growth-uncoupled fermentation, and no ethanol production (panel A, Figures 1 – 3 ), similar to what we have done before for C. thermocellum ( Sharma et al., 2024 ). Download figure Open in new tab Figure 1: Fermentation behavior of E. coli RL3019. A) Concentration of extracellular metabolites B) Concentration of intracellular metabolites C) f6p to fbp ratio D) dhap to 3pg ratio E) Adenylate energy charge ratio. The fermentation was performed with 120 g/L glucose in M9 minimal medium at 37°C with pH-maintained at 6.5 +\- 0.05 by addition of 4 M potassium hydroxide. The shaded background represents different phases of fermentation: dark gray, growth coupled fermentation; medium gray, growth uncoupled fermentation; and light gray, no ethanol production phases. One representative fermentation profile is shown (n=2). Figure S2 shows a biological duplicate of this experiment. Additional File 1 (eco1_ext) shows additional extracellular metabolites data for this experiment. Additional File 1 (eco1_int) shows intracellular metabolites concentration. In subplots B–E, each circle represents an individual measurement, and the line plot represents a trendline, which is smoothed using a rolling average with a window size of 3. The abbreviations used in this figure are defined in the list of abbreviations. Download figure Open in new tab Figure 2: Fermentation profile of T. saccharolyticum M1442. A) Concentration of extracellular metabolites B) Concentration of intracellular metabolites C) g6p to fbp ratio D) dhap to 3pg ratio E) pyr to accoa ratio F) accoa to coa ratio G) Nicotinamide cofactor ratios H) Adenylate energy charge ratio Fermentation was performed with 140 g/L cellobiose in MTC-7 medium at 51°C with pH-maintained at 6.0 +\- 0.05 by addition of 4 M potassium hydroxide. The shaded background represents different phases of fermentation: dark gray, growth coupled fermentation; medium gray, growth uncoupled fermentation; and light gray, no ethanol production phases. One representative fermentation profile is shown (n=2). Figure S3 shows a biological duplicate of this experiment. Additional File 1 (tsac1_ext) shows additional extracellular metabolites data for this experiment. Additional File 1 (tsac1_int) shows intracellular metabolites concentration. In subplots B–H, each circle represents an individual measurement, and the line plot represents a trendline, which is smoothed using a rolling average with a window size of 3. The abbreviations used in this figure are defined in the list of abbreviations. Download figure Open in new tab Figure 3: Fermentation profile of Z. mobilis ZM4. A) Concentration of extracellular metabolites B) Concentration of intracellular metabolites C) g6p to 6pgn ratio D) kdpg to 3pg ratio E) Nicotinamide cofactor ratio F) Adenylate energy charge ratio Fermentation was performed with 160 g/L glucose in ZMM-2 medium at 37°C with pH-maintained at 6.0 +\- 0.05 by addition of 4 M potassium hydroxide. The shaded background represents different phases of fermentation: dark gray, growth coupled fermentation; medium gray, growth uncoupled fermentation; and light gray, no ethanol production phases. One representative fermentation profile is shown (n=2). Figure S4 shows a biological duplicate of this experiment. Additional File 1 (zmm1_ext) shows additional extracellular metabolites data for this experiment. Additional File 1 (zmm1_int) shows intracellular metabolites concentration. In subplots B–H, each circle represents an individual measurement, and the line plot represents a trendline, which is smoothed using a rolling average with a window size of 3. The abbreviations used in this figure are defined in the list of abbreviations. For E. coli , the final ethanol titer reached 448 ± 2 mM (20.6 ± 0.1 g/L), with 358 ± 8 mM (64.5 ± 1.4 g/L) glucose remaining at the end of fermentation (panel A, Figure 1 ). For T. saccharolyticum , the final ethanol titer reached 949 ± 6 mM (43.7 ± 0.3 g/L), with 203 ± 2 mM (36.5 ± 0.4 g/L) of glucose accumulating in the medium ( Figure 2 , panel A). Although cellobiose was fully consumed by ∼90 hours ( Figure 2 , panel A), fermentation was not substrate-limited because glucose, which is also a utilizable carbon source for T. saccharolyticum ( Tsakraklides et al., 2012 ), remained available. For Z. mobilis fermentation, the final ethanol titer reached 1553 ± 15 mM (71.6 ± 0.7 g/L) with 174 ± 2 mM (31.3 ± 0.4 g/L) glucose still present at the end of fermentation ( Figure 3 , Panel A). Analysis of intracellular metabolites revealed that pyruvate accumulated progressively in E. coli and T. saccharolyticum , indicating inhibition of reactions downstream of pyruvate ( Figures 1 – 2 , panel B). To further probe glycolytic flux, we examined metabolite ratios that serve as indicators of bottlenecks between upper and lower glycolysis. In E. coli , the ratio of f6p to fbp increased during the late growth-uncoupled phase ( Figure 1 , panel C), consistent with reduced activity at the PFK step. In T. saccharolyticum , f6p was not detected, so the g6p to fbp ratio was used as a proxy and showed a similar upwards trend ( Figure 2 , panel C). The dhap to 3pg ratio also increased as fermentation progressed in both microbes ( Figure 1 - 2 , panel D), consistent with restricted flux in lower glycolysis downstream of triose phosphate metabolism. In T. saccharolyticum , the pyr to accoa ratio began increasing during the early–mid growth-uncoupled phase ( Figure 2 , panel D), suggesting reduced conversion of pyruvate to acetyl-CoA. Together, these data indicate that pyruvate accumulation occurs first, followed by buildup of upstream intermediates extending to hexose phosphates. In T. saccharolyticum , the accoa to coa ratio decreased further ( Figure 2 , panel F), reinforcing the interpretation of a bottleneck at the pyruvate-to-acetyl-CoA step. Additionally, the ratios of oxidized to reduced nicotinamide cofactors (nad to nadh and nadp to nadph) decreased as fermentation progressed ( Figure 2 , panel G). By the end of fermentation, the average ratios were 2.5 for nad to nadh and 1.6 for nadp to nadph, values close to those reported previously for mid-log cells grown at lower substrate concentrations (2.1 and 1.4, respectively) ( Beri et al., 2016 ). This further supports the conclusion that fermentation is inhibited at the pyruvate-to-ethanol pathway. In E. coli and Z. mobilis , metabolite ratios for the pyruvate to acetaldehyde step and nicotinamide cofactors could not be quantified because acetaldehyde and nadh were not detected by LC-MS. By contrast, Z. mobilis did not accumulate pyruvate as fermentation progressed ( Figure 3 , panel B). The g6p to 6pgn ratio increased slightly after the growth-coupled phase ( Figure 3 , panel C), while other ratios showed transient increases followed by declines ( Figure 3 , panels D-E). The absence of sustained accumulation of intracellular metabolites suggests that the limitation of ethanol titer in this organism may not stem from central metabolic bottlenecks, but rather from other factors such as inhibition of substrate uptake transporters. Across all three microbes, the adenylate energy charge remained relatively constant during fermentation ( Figure 1 , panel E; Figure 2 , panel H; Figure 3 , panel F). 3.3 Global overview of metabolite changes during growth and fermentation In addition to specific intracellular metabolites with absolute quantification, our LC-MS platform gives us information about relative concentrations of dozens of other metabolites. By clustering metabolites based on relative abundance changes over time, we can observe patterns of metabolite changes ( Figure 4 ). Download figure Open in new tab Figure 4: Relative abundance of metabolites during fermentation. Untargeted metabolite analysis was performed for all samples. Metabolites were identified by M/Z ratio and retention time. For each metabolite, peak area values were normalized by calculating z-scores across all time points, allowing comparison of temporal dynamics independent of absolute abundance levels. Metabolites were hierarchically clustered using average linkage clustering with correlation distance as the similarity metric. Red shading indicates high abundance. Blue shading indicates low abundance. The growth phase is indicated by shaded bars at the top of the columns: dark grey corresponds to growth-coupled fermentation, medium grey corresponds to growth-uncoupled fermentation, and light grey corresponds to no fermentation. Only one replicate is shown here. Both replicates and all metabolites are shown in Supporting Figure S5. Underlying data is available as Additional File 3. 3.3.1 In E. coli RL3019, there is a relative lack of central carbon metabolites, suggesting that many of them are present at very low levels. In pyruvate metabolism, there is an initial accumulation of acetyl-CoA and asparagine, but as growth slows, a variety of pyruvate-derived amino acids start to accumulate. In nucleotide metabolism, there is a gradual transition from high-energy phosphates (ATP, GTP) to low-energy phosphates (AMP, GMP) to nucleotide breakdown products (uracil, inosine, hypoxanthine, xanthine, guanine) ( Figure 4 ). Note that these changes are emphasized by the row-wise normalization of the heatmap ( Figure 4 ). The heatmap emphasizes trends and changes. The absolute metabolite concentration measurements ( Figures 1 - 3 ) provide better comparison to known physiological states (e.g. adenylate charge) and between organisms. 3.3.2 In T. saccharolyticum M1442, glycolytic intermediates accumulate during log-phase. Immediately after growth stops, there is a large flux shift to the TCA cycle, and many glycolytic intermediates are depleted. As fermentation slows, there is a large accumulation of intermediates in the valine, leucine, and isoleucine biosynthesis pathway, however there appears to be some kind of bottleneck as levels of the endproduct amino acids are low. In energy cofactor metabolism, there is a spike in high-energy phosphates (ATP, GTP) right before growth stops, and a spike of low-energy phosphates (AMP, GMP) right after growth stops. As fermentation slows and stops, there is a moderate accumulation of nucleotide degradation products (thymidine, uridine, uracil, and xanthosine) ( Figure 4 ). 3.3.3 In Z. mobilis ZM4, there is a long lag phase where intracellular metabolites are generally low. Similar to T. saccharolyticum , log-phase growth is associated with high levels of central carbon metabolites. After growth stops, there is an accumulation of upper glycolysis intermediates (fructose-6-phosphate, glucose-6-phosphate, and glucose-1-phosphate), as well as some pyruvate-derived metabolites (leucine/isoleucine, valine, and succinate). In energy cofactor metabolism, a similar pattern is observed compared to T. saccharolyticum with a spike in high-energy phosphates (ATP, GTP) right before growth stops, and a spike of low-energy phosphates (AMP, GMP) right after growth stops, as well as accumulation of nucleotide degradation products as fermentation stops ( Figure 4 ). 3.4 Thermodynamic constraints highlight PYK as a limiting step in ethanologen E. coli and T. saccharolyticum , with Z. mobilis showing no clear bottleneck We hypothesized that thermodynamic equilibrium might be causing fermentation to stop. To evaluate this hypothesis, we applied the Max–Min Driving Force (MDF) framework ( Noor et al., 2014 ) to measured metabolite concentrations. A pathway is considered thermodynamically feasible if its MDF score is positive, which also reflects the degree of kinetic constraint from backward flux. Positive MDF values indicate that all reactions can maintain sufficiently high driving forces to support forward flux, whereas MDF values near or below zero suggest that at least one reaction is at thermodynamic equilibrium or thermodynamically infeasible ( Noor et al., 2014 ). At each fermentation time point, MDF scores were calculated using the eQuilibrator Python library ( Flamholz et al., 2012 ) [ https://gitlab.com/equilibrator/equilibrator-pathway ]. The reactions were set up as shown in Figure 5 and the reaction stoichiometries are provided in Additional File 2. Download figure Open in new tab Figure 5: Graphical representation of E. coli RL3019, T. saccharolyticum M1442 and Z. mobilis ZM4 fermentation pathway. Note that T. saccharolyticum does not have FNORP reaction. However, the reactions involved in that specific location have stoichiometry similar to canonical FNORP ( De Souza et al., 2025 ). Blue capitalized text represents reactions, red text indicates cofactors and black text represents metabolites. The abbreviations used in this figure are defined in the list of abbreviations. MDF analysis yields two outputs. The first is the overall MDF score for the pathway, which should be positive whenever forward flux occurs. The second is the identification of bottleneck reactions (highlighted in red in Figures 6 ), representing the reactions that are determining the MDF value (i.e. the most thermodynamically limiting steps) under the measured metabolite concentrations. As expected, during the growth-coupled and growth-uncoupled fermentation phases, MDF scores were largely positive in E. coli , aside from a few early points with high variability due to low cell density. In Z. mobilis , MDF scores decreased slightly as fermentation progressed but remained positive throughout, including during the no-ethanol-production phase. In contrast, T. saccharolyticum exhibited several intervals of negative MDF scores despite active substrate conversion to ethanol, suggesting these values likely reflect experimental error and prompting additional analyses to identify possible sources of this discrepancy. Download figure Open in new tab Figure 6. Max-Min Driving Force (MDF) scores during fermentations of E . coli RL3019 (left), T. saccharolyticum M1442 (middle), and Z. mobilis ZM4 (right). MDF scores were calculated at each time point during the course of fermentation. The blue circles represent the specific time points for which MDF scores are plotted. In the top row of panels, the shaded backgrounds indicate different phases of fermentation: dark gray, growth-coupled fermentation; medium gray, growth-uncoupled fermentation; and light gray, no ethanol production. For each organism, the four lower subplots represent cumulative ΔrGʹ for the blue-circle time points in the MDF score chart. The green dotted line shows cumulative ΔrGʹ when metabolite concentrations are fixed at 1 mM, the gray line shows values based on measured concentrations, and the red line highlights the bottleneck reactions responsible for changes in MDF scores. Biological replicates of these experiments are shown in Figure S6. Abbreviations used in this figure are defined in the list of abbreviations. 3.5 Error analysis of MDF scores in E. coli and T. saccharolyticum We considered two sources of error for our MDF measurements for negative scores in engineered strains of E. coli and T. saccharolyticum : (1) random measurement error (equally distributed across all metabolites), or (2) specific errors for a single metabolite. For the first approach, we investigated whether relaxing the concentrations of all quantified metabolites simultaneously by applying multiplicative factors to the concentration bounds would alter the MDF scores. The lower bound for each metabolite was set to the measured value divided by the chosen factor, while the upper bound was set to the measured value multiplied by the same factor. For T. saccharolyticum , MDF scores remained negative at the measured concentrations but became positive once the bounds were expanded beyond a 1.5-fold change (Figure S7). At later stages of fermentation, larger relaxations were required, and progressively increasing the bounds (up to 100-fold) raised the MDF values but did not change the qualitative trend of declining thermodynamic feasibility after ethanol production ceased (Figure S7). In the case of E. coli , MDF scores during the later phase of fermentation became positive with as little as a 2-fold relaxation. With more than a 5-fold change in metabolite concentrations, even the late fermentation points were rendered thermodynamically feasible. Since only large deviations from measured concentrations were sufficient to reverse the negative MDF scores, we next examined the effect of measurement errors in single metabolites. For the second approach, we investigated whether relaxing the concentration bounds of individual measured metabolites would affect the MDF scores. For each metabolite, the lower bound was set to 1 μM and the upper bound to 100 mM. As shown for E. coli ( Figure 7 ), relaxing the bounds of pep, pyr, atp, or adp individually was sufficient to raise the MDF score to higher values even during the no-ethanol-production phase. These metabolites are all associated with the PYK reaction, and the observation in this relaxation experiment is consistent with our identification of the PYK reaction being in local thermodynamic equilibrium ( Figure 6, 146 hour timepoint sub-panel). Download figure Open in new tab Figure 7: Max-Min Driving Force (MDF) score by relaxing the bounds of measured metabolites during E. coli RL3019 fermentation. The shaded background represents different phases of fermentation: dark gray, growth coupled fermentation; medium gray, growth uncoupled fermentation; and light gray, no ethanol production phases. The abbreviations used in this figure are defined in the list of abbreviations. In the case of T. saccharolyticum , we can assess the effect on metabolite bound relaxation on the two negative-MDF (i.e. thermodynamically infeasible) regions separately ( Figure 8 ): The reaction infeasibility observed between 29 and 57 hours, which could be resolved by changing the bounds for etoh, accoa, coa, nad, nadh, nadp, and nadph. As mentioned above (at 43 hrs, Figure 8 for T. saccharolyticum ), the metabolic bottleneck is seen at the reactions downstream of pyruvate, and changing the bounds of any of these metabolites alleviates the issue. The reaction infeasibility after ethanol production stops. Although we could not resolve the negative MDF values for all time points as we did for E. coli , we were able to improve the feasibility at the first time point by relaxing constraints on the bounds for pep, pyr, atp, and adp. This improvement is also attributed to the fact that these metabolites are involved in the PYK reaction and changing any of them would increase the feasibility of this reaction. Download figure Open in new tab Figure 8: Max-Min Driving Force (MDF) score by relaxing the bounds of measured metabolites during T. saccharolyticum M1442 fermentation. The shaded background represents different phases of fermentation: dark gray, growth coupled fermentation; medium gray, growth uncoupled fermentation; and light gray, no ethanol production phases. The abbreviations used in this figure are defined in the list of abbreviations. DISCUSSION The goal of this study was to understand factors that limit ethanol titer in microbial fermentations. A key limitation in much prior work on this topic is a lack of high-quality intracellular metabolite data collected during fermentations where ethanol was produced at high titer. By combining both high-titer fermentations and quantitative intracellular metabolite quantification, the data collected here therefore represents a unique window into the metabolic changes that occur as fermentation slows and eventually stops. Furthermore, by observing this phenomenon in several different species ( E. coli , T. saccharolyticum , and Z. mobilis ) we can get hints about the generalizability of our insights across different species. This topic has also been extensively studied in yeast ( Saccharomyces cerevisiae ). In the brewing literature, the phenomenon of a fermentation slowing or stopping before the substrate is consumed is known as a “stuck” fermentation ( Alexandre & Charpentier, 1998 ; Bisson, 1999 ). Several factors have been proposed as causes of stuck fermentations, including deficiencies in nitrogen, oxygen, vitamins and minerals, ethanol inhibition, and membrane damage ( Alexandre & Charpentier, 1998 ), however the mechanisms by which these factors affect metabolism are not well understood. It has been shown that ethanol inhibits key glycolytic enzymes such as phosphofructokinase, phosphoglycerate kinase, and pyruvate decarboxylase, leading to a gradual slowdown in fermentation ( Millar et al., 1982 ), but the extent to which this is responsible for the stuck fermentation phenotype is not known. One of the more well-supported mechanisms explaining stuck fermentations is a lack of substrate transport. Sugar transporters in yeast are known to have a low half-life, requiring rapid biosynthesis to maintain transport capacity. Thus a cessation of protein synthesis can rapidly eliminate substrate transport ( Bisson, 1999 ). Previously, cessation of ethanol production in Z. mobilis was suggested to result from enzyme inhibition, membrane disruptions, and metabolic imbalances ( Millar et al., 1982 ; Osman & Ingram, 1985 ). Although all of these factors may be present at high enough ethanol concentrations, we are most interested in identifying the factor(s) that first cause fermentation to stop. Our observation of a relative lack of accumulation of all intracellular metabolites in Z. mobilis when fermentation stops suggests that inhibition of individual enzymes or imbalances of metabolite ratios is not a likely cause of this phenotype ( Figure 3 ). Instead, we propose that a lack of substrate transport is the primary cause. However, we cannot completely rule out the possibility of metabolite leakage playing a role. Previous studies on ethanologenic strains of E. coli have suggested that the Z. mobilis Pdc enzyme is inhibited by ethanol ( Millar et al., 1982 ; Trinh et al., 2010 ), resulting in accumulation of pyruvate at high ethanol titers and eventually the cessation of fermentation. Our observations in E. coli are largely consistent with prior studies, and we also conclude that a lack of flux of the PDC reaction is responsible for the stopping phenotype ( Figure 1 ). Interestingly, however, Z. mobilis itself does not display strong inhibition of the pyruvate-to-ethanol pathway under our tested conditions, as no pyruvate accumulation was observed during fermentation ( Figure 3 ). This is particularly puzzling, since the heterologous ethanol production pathway in E. coli RL3019 consists of the pdc and adh genes from Z. mobilis . One possible explanation may be related to cofactor biosynthesis. The Pdc enzyme requires thiamine pyrophosphate, and its native biosynthesis pathways may be differently inhibited by ethanol in E. coli vs. Z. mobilis . This could also explain the stimulatory effect of yeast extract (which contains high levels of thiamine) on ethanol titer ( Table 2 ). Understanding the reason for this observed difference between E. coli and Z. mobilis is an important area for future investigation. Furthermore, the direct comparison of E. coli and Z. mobilis provides an indication of the extent to which the industrially-relevant ethanol production phenotype has been successfully transferred. At high substrate concentrations in chemically defined media, Z. mobilis exhibits more than a 3-fold higher ethanol titer, compared to E. coli ( Table 2 ). Therefore, it appears that simply transferring the pyruvate to ethanol pathway ( pdc and adhB genes) is only sufficient to partly transfer the ethanol producing ability of Z. mobilis ( Millar et al., 1982 ; Trinh et al., 2010 ). Compared to either E. coli or Z. mobilis , the metabolism of T. saccharolyticum is relatively poorly understood. The wild type organism is known to be relatively ethanol sensitive. In a comparison among six other bacteria, it was the most ethanol sensitive, and the only one unable to initiate growth at ethanol concentrations of 25 g/L or higher ( Huffer et al., 2011 ). Interestingly, it appears to have a latent ethanol production pathway that allows for both high yield and titer production. This pathway can be activated by disrupting enzymes associated with acetate and lactate production ( Shaw et al., 2008 , 2011 ). Subsequently, mutations are observed in the hfs gene cluster that appear to increase ethanol yield by increasing electron transfer from ferredoxin to NAD(P) + ( Eminoğlu et al., 2017 ; Fabri, De Souza, et al., 2025 ). Additional mutations in the adhE gene disrupt activity of the ADH domain of AdhE (Fabri, Pech-Canul, et al., 2025), forcing it to rely on the NADPH-linked AdhA enzyme instead ( Zheng et al., 2017 ), and effectively switching the cofactor specificity of the ADH reaction from NADH to NADPH. Previously we found that in wild type T. saccharolyticum , adding ethanol during fermentation causes a substantial increase in the dhap to 3pg ratio, indicative of a metabolic bottleneck at the GAPDH reaction ( Tian et al., 2017 ), however that effect was decreased in this work, suggesting that the mutations involved in the development of the ethanologen strain tested here (strain M1442) substantially reduced or eliminated that bottleneck ( Figure 2 ). And in fact, that outcome is consistent with our previous thermodynamic analysis of the effects of changing the cofactor specificity of the ADH reaction ( Dash et al., 2019 ). For the ethanologenic strain of T. saccharolyticum (strain M1442), there have been no prior studies of changes in its metabolism during high titer fermentation. Here, we observe an increase in pyruvate that is associated with the cessation of fermentation. The most likely explanation is that fermentation stopped due to loss of function of the pyruvate to ethanol pathway, which is similar to what was observed in E. coli . The relatively high increase in the pyruvate to acetyl-CoA, compared to the relatively steady ratio of acetyl-CoA to CoA suggest that a lack of flux through the PFOR reaction may be the immediate cause of fermentation cessation. Previously we have shown that Pfor enzymes from different organisms differ in their ability to enable high titer ethanol production ( Cui et al., 2019 ). However, the PforA enzyme from T. saccharolyticum was the one that enabled the highest ethanol titers. It is also possible that the lack of flux through the PFOR reaction is due to inhibition of reactions necessary for electron transfer from ferredoxin to NAD(P) + . The role of energy charge in cessation of fermentation is still ambiguous. For all three organisms we studied, the adenylate energy charge was relatively constant throughout fermentation, indicating that cessation was not driven by depletion of cellular energy ( Figures 1 - 3 ). On the other hand, the heatmap data shows trends in energy cofactors associated with changes in growth phase, suggesting the possibility that relatively small changes in adenylate charge can have a large effect on the physiological response of the cell ( Figure 4 ). A critical factor that we identified via our error analysis of our MDF calculations was the importance of accurate cofactor quantification, particularly the nicotinamide cofactors (NADH, NAD+, NADPH, and NADP+). The large number of reactions in which these metabolites participate often gives them a large effect on MDF calculations. Such errors may create the appearance of pathway infeasibility even when flux is clearly occurring. Thus, rigorous metabolite sampling and analytical precision are essential when linking intracellular metabolite levels to thermodynamic constraints. Furthermore, these cofactors can also be rapidly interconverted during the metabolite quenching process. Since there are practical limits on the speed at which metabolite quenching can be performed, future experiments may benefit from the inclusion of orthogonal methods of nicotinamide cofactor quantification, such as redox-active sensors ( Brekasis, 2003 ; Henry & Crosson, 2011 ), or dedicated redox reporter enzymes ( Bekers et al., 2015 ; Canelas et al., 2008 ). Looking at all three strains together, several consistent patterns emerge. During rapid growth, relative concentrations of central metabolites are high. These pools start to deplete almost immediately after growth stops ( Figure 4 , Supporting Figure S5), potentially due to a lack of substrate transport. Thus, one of the most profound shifts in metabolism during batch fermentations is the shift from growth to non-growth. Once the cells have shifted to non-growth fermentation, the rate of product formation gradually slows down until it eventually stops completely. Unlike with the growth/no-growth transition, the metabolic signature of this transition is less abrupt (except perhaps for E. coli ). Fermentation appears to stop due to a lack of flux through the pyruvate-to-ethanol pathways. In E. coli and T. saccharolyticum , this pyruvate accumulation then leads to a local thermodynamic equilibrium at the upstream (PYK) reaction, and this blockage eventually propagates upstream to upper glycolysis. In all three organisms, the final stage of cessation of fermentation is associated with an accumulation of nucleotide degradation products. The reasons for this are not clear, but one potential explanation is a loss of proton motive force, leading to acidification of the cytoplasm. The key benefit of the thermodynamic analysis is that it allows us to build a causal chain between the observed changes in metabolite levels and a fundamental physical constraint (i.e. local thermodynamic equilibrium) that could explain the cessation of fermentation. The use of intracellular metabolite quantification under high titer production conditions, combined with thermodynamic analysis, is a promising technique to provide insights into metabolic mechanisms of product titer limitation. Important areas for future research include more accurate quantification of energy and redox cofactors (ATP/ADP/AMP, GTP/GDP/GMP, NAD(P) + /NAD(P)H), developing more robust pyruvate to ethanol production pathways, and understanding the remaining factors that allow Z. mobilis to produce higher ethanol titers compared to E. coli . Authors contributions BDS: Conceptualization; Data curation; Formal analysis; Investigation; Methodology; Validation; Visualization; Writing - original draft ET: Data acquisition and curation; Investigation; DMS: Data acquisition DAN: Data curation, Review and Editing LL: Funding Acquisition, Project Administration, Writing - Review and Editing DGO: Writing - Review and Editing, Supervision, Project Administration, Funding Acquisition Possible journals: Molecular systems biology, mBio, Metabolic engineering, PNAS, nature metabolism, nature microbiology LIST OF METABOLITE ABBREVIATIONS 2pg 2-phosphoglycerate 3pg 3-phosphoglycerate 6pgln 6-phosphogluconolactone 6pgn 6-phosphogluconate acald acetaldehyde accoa acetyl coenzyme A adp adenosine diphosphate atp adenosine triphosphate bpg bisphosphoglycerate cb cellobiose coa coenzyme A dhap dihydroxyacetone phosphate etoh ethanol f6p fructose 6-phosphate fbp fructose 1,6-bisphosphate fdox oxidized ferredoxin fdred reduced ferredoxin g3p glycerol-3-phosphate g6p glucose 6-phosphate glc glucose kdpg 2-keto-3-deoxy-6-phosphogluconate nad oxidized nicotinamide adenine dinucleotide nadh reduced nicotinamide adenine dinucleotide nadp oxidized nicotinamide adenine dinucleotide phosphate nadph reduced nicotinamide adenine dinucleotide phosphate pep phosphoenolpyruvate pyr pyruvate LIST OF ENZYME ABBREVIATIONS ADH alcohol dehydrogenase reaction (nad-dependent) ADHP alcohol dehydrogenase reaction (nadp-dependent) ALDH aldehyde dehydrogenase reaction CBP cellobiose phosphorylase reaction EDA 2-keto-3-deoxy-6-phosphogluconate aldolase EDD 6-phosphogluconate dehydrogenase ENO enolase reaction FBA fructose-1,6-bisphosphate aldolase reaction FNORP ferredoxin:nadp oxidoreductase reaction G6PDH glucose 6-phosphate dehydrogenase reaction GAP glyceraldehyde-3-phosphate dehydrogenase reaction GLK glucokinase reaction GPM phosphoglycerate mutase reaction PDC pyruvate decarboxylase reaction PFOR pyruvate:ferredoxin oxidoreductase reaction PGI phosphoglucose isomerase reaction PGK phosphoglycerate kinase reaction PGL 6-phosphogluconolactonase PGM phosphoglucomutase reaction PFK phosphofructokinase reaction PYK pyruvate kinase reaction TPI triosephosphate isomerase reaction ACKNOWLEDGEMENTS This work was supported by the Center for Bioenergy Innovation (CBI), U.S. Department of Energy, Office of Science, Biological and Environmental Research Program under Award Number ERKP886. REFERENCES 1. ↵ A. D’Alessandro Agrawal , S. , Kumar , S. , Sehgal , R. , George , S. , Gupta , R. , Poddar , S. , Jha , A. , & Pathak , S. ( 2019 ). El-MAVEN: A Fast, Robust, and User-Friendly Mass Spectrometry Data Processing Engine for Metabolomics . In A. D’Alessandro (Ed.), High-Throughput Metabolomics (Vol. 1978, pp. 301 – 321 ). Springer New York . doi: 10.1007/978-1-4939-9236-2_19 OpenUrl CrossRef PubMed 2. ↵ Alexandre , H. , & Charpentier , C . ( 1998 ). Biochemical aspects of stuck and sluggish fermentation in grape must . Journal of Industrial Microbiology and Biotechnology , 20 ( 1 ), 20 – 27 . doi: 10.1038/sj.jim.2900442 OpenUrl CrossRef Web of Science 3. ↵ Bekers , K. M. , Heijnen , J. J. , & Van Gulik , W. M. ( 2015 ). Determination of the in vivo NAD:NADH ratio in Saccharomyces cerevisiae under anaerobic conditions, using alcohol dehydrogenase as sensor reaction: Determination of in vivo NAD:NADH ratio in S. cerevisiae . Yeast , 32 ( 8 ), 541 – 557 . doi: 10.1002/yea.3078 OpenUrl CrossRef PubMed 4. ↵ Beri , D. , Olson , D. G. , Holwerda , E. K. , & Lynd , L. R . ( 2016 ). Nicotinamide cofactor ratios in engineered strains of Clostridium thermocellum and Thermoanaerobacterium saccharolyticum . FEMS Microbiology Letters , 363 ( 11 ), 1 – 11 . doi: 10.1093/femsle/fnw091 OpenUrl CrossRef PubMed 5. ↵ Bisson , L. F . ( 1999 ). Stuck and Sluggish Fermentations . American Journal of Enology and Viticulture , 50 ( 1 ), 107 – 119 . doi: 10.5344/ajev.1999.50.1.107 OpenUrl Abstract / FREE Full Text 6. ↵ Brekasis , D . ( 2003 ). A novel sensor of NADH/NAD+ redox poise in Streptomyces coelicolor A3(2) . The EMBO Journal , 22 ( 18 ), 4856 – 4865 . doi: 10.1093/emboj/cdg453 OpenUrl Abstract / FREE Full Text 7. ↵ Callaghan , M. M. , Thusoo , E. , Sharma , B. D. , Getahun , F. , Stevenson , D. M. , Maranas , C. , Olson , D. G. , Lynd , L. R. , & Amador-Noguez , D . ( 2023 ). Deuterated water as a substrate-agnostic isotope tracer for investigating reversibility and thermodynamics of reactions in central carbon metabolism . Metabolic Engineering , 80 , 254 – 266 . doi: 10.1016/j.ymben.2023.10.006 OpenUrl CrossRef PubMed 8. ↵ Canelas , A. B. , Van Gulik , W. M. , & Heijnen , J. J. ( 2008 ). Determination of the cytosolic free NAD/NADH ratio in Saccharomyces cerevisiae under steady- state and highly dynamic conditions . Biotechnology and Bioengineering , 100 ( 4 ), 734 – 743 . doi: 10.1002/bit.21813 OpenUrl CrossRef PubMed Web of Science 9. ↵ Carreón-Rodríguez , O. E. , Gutiérrez-Ríos , R. M. , Acosta , J. L. , Martinez , A. , & Cevallos , M. A . ( 2019 ). Phenotypic and genomic analysis of Zymomonas mobilis ZM4 mutants with enhanced ethanol tolerance . Biotechnology Reports , 23 , e00328 . doi: 10.1016/j.btre.2019.e00328 OpenUrl CrossRef 10. ↵ Clasquin , M. F. , Melamud , E. , & Rabinowitz , J. D . ( 2012 ). LC- MS Data Processing with MAVEN: A Metabolomic Analysis and Visualization Engine . Current Protocols in Bioinformatics , 37 ( 1 ). doi: 10.1002/0471250953.bi1411s37 OpenUrl CrossRef 11. ↵ Cui , J. , Olson , D. G. , & Lynd , L. R . ( 2019 ). Characterization of the Clostridium thermocellum AdhE, NfnAB, ferredoxin and Pfor proteins for their ability to support high titer ethanol production in Thermoanaerobacterium saccharolyticum . Metabolic Engineering , 51 , 32 – 42 . doi: 10.1016/j.ymben.2018.09.006 OpenUrl CrossRef PubMed 12. ↵ Dash , S. , Olson , D. G. , Hung , S. , Chan , J. , Amador-noguez , D. , Lynd , L. R. , & Maranas , C. D . ( 2019 ). Thermodynamic analysis of the pathway for ethanol production from cellobiose in Clostridium thermocellum . Metabolic Engineering , 55 ( March ), 161 – 169 . doi: 10.1016/j.ymben.2019.06.006 OpenUrl CrossRef PubMed 13. ↵ De Souza , L. C. , Herring , C. D. , & Lynd , L. R. ( 2025 ). Genetic investigation of hydrogenases in Thermoanaerobacterium thermosaccharolyticum suggests that redox balance via hydrogen cycling enables high ethanol yield . Applied and Environmental Microbiology , e01109 – 24 . doi: 10.1128/aem.01109-24 OpenUrl CrossRef 14. ↵ Dien , B. S. , Cotta , M. A. , & Jeffries , T. W . ( 2003 ). Bacteria engineered for fuel ethanol production: Current status . Applied Microbiology and Biotechnology , 63 ( 3 ), 258 – 266 . doi: 10.1007/s00253-003-1444-y OpenUrl CrossRef PubMed Web of Science 15. ↵ Edwards , J. S. , & Palsson , B. O . ( 2000 ). The Escherichia coli MG1655 in silico metabolic genotype: Its definition, characteristics, and capabilities . Proceedings of the National Academy of Sciences , 97 ( 10 ), 5528 – 5533 . doi: 10.1073/pnas.97.10.5528 OpenUrl Abstract / FREE Full Text 16. ↵ Eminoğlu , A. , Murphy , S. J.-L. , Maloney , M. , Lanahan , A. , Giannone , R. J. , Hettich , R. L. , Tripathi , S. A. , Beldüz , A. O. , Lynd , L. R. , & Olson , D. G . ( 2017 ). Deletion of the hfsB gene increases ethanol production in Thermoanaerobacterium saccharolyticum and several other thermophilic anaerobic bacteria . Biotechnology for Biofuels , 10 ( 1 ), 282 . doi: 10.1186/s13068-017-0968-9 OpenUrl CrossRef PubMed 17. Fabri , J. H. T. M. , De Souza , L. C. , Bergamo , L. W. , Lynd , L. R. , & Olson , D. G. ( 2025 ). Identification of electron transfer enzymes in Thermoanaerobacterium saccharolyticum . Journal of Bacteriology , 207 ( 7 ), e00107 – 25 . doi: 10.1128/jb.00107-25 OpenUrl CrossRef PubMed 18. Fabri , J. H. T. M. , Pech-Canul , A. , Ziegler , S. J. , Burgin , T. E. , Richardson , I. D. , Maloney , M. I. , Bomble , Y. J. , Lynd , L. R. , & Olson , D. G . ( 2025 ). The role of AdhE mutations in Thermoanaerobacterium saccharolyticum . Journal of Bacteriology , 207 ( 5 ), e00015 – 25 . doi: 10.1128/jb.00015-25 OpenUrl CrossRef PubMed 19. Fein , JaredE ., Charley , RobertC ., Hopkins , KimA ., Lavers , B. , & Lawford , HughG . ( 1983 ). Development of a simple defined medium for continuous ethanol production by Zymomonas mobilis . Biotechnology Letters , 5 ( 1 ). doi: 10.1007/BF00189955 OpenUrl CrossRef 20. ↵ Flamholz , A. , Noor , E. , Bar-Even , A. , & Milo , R . ( 2012 ). EQuilibrator—The biochemical thermodynamics calculator . Nucleic Acids Research , 40 ( D1 ), 770 – 775 . doi: 10.1093/nar/gkr874 OpenUrl CrossRef 21. ↵ Fulton , L. M. , Lynd , L. R. , Körner , A. , Greene , N. , & Tonachel , L. R . ( 2015 ). The need for biofuels as part of a low carbon energy future . Biofuels, Bioproducts and Biorefining , 9 ( 5 ), 476 – 483 . doi: 10.1002/bbb.1559 OpenUrl CrossRef 22. Ghosh , I. N. , & Landick , R. ( 2016 ). OptSSeq: High-Throughput Sequencing Readout of Growth Enrichment Defines Optimal Gene Expression Elements for Homoethanologenesis . ACS Synthetic Biology , 5 ( 12 ), 1519 – 1534 . doi: 10.1021/acssynbio.6b00121 OpenUrl CrossRef PubMed 23. ↵ Henry , J. T. , & Crosson , S. ( 2011 ). Ligand-Binding PAS Domains in a Genomic, Cellular, and Structural Context . Annual Review of Microbiology , 65 ( 1 ), 261 – 286 . doi: 10.1146/annurev-micro-121809-151631 OpenUrl CrossRef PubMed Web of Science 24. ↵ Herring , C. D. , Kenealy , W. R. , Joe Shaw , A. , Covalla , S. F. , Olson , D. G. , Zhang , J. , Ryan Sillers , W. , Tsakraklides , V. , Bardsley , J. S. , Rogers , S. R. , Thorne , P. G. , Johnson , J. P. , Foster , A. , Shikhare , I. D. , Klingeman , D. M. , Brown , S. D. , Davison , B. H. , Lynd , L. R. , & Hogsett , D. A . ( 2016 ). Strain and bioprocess improvement of a thermophilic anaerobe for the production of ethanol from wood . Biotechnology for Biofuels , 9 ( 1 ), 125 . doi: 10.1186/s13068-016-0536-8 OpenUrl CrossRef PubMed 25. ↵ Holwerda , E. K. , Thorne , P. G. , Olson , D. G. , Amador-Noguez , D. , Engle , N. L. , Tschaplinski , T. J. , Van Dijken , J. P. , & Lynd , L. R. ( 2014 ). The exometabolome of Clostridium thermocellum reveals overflow metabolism at high cellulose loading . Biotechnology for Biofuels , 7 ( 1 ), 155 . doi: 10.1186/s13068-014-0155-1 OpenUrl CrossRef PubMed 26. ↵ Huffer , S. , Clark , M. E. , Ning , J. C. , Blanch , H. W. , & Clark , D. S . ( 2011 ). Role of Alcohols in Growth, Lipid Composition, and Membrane Fluidity of Yeasts, Bacteria, and Archaea . Applied and Environmental Microbiology , 77 ( 18 ), 6400 – 6408 . doi: 10.1128/AEM.00694-11 OpenUrl Abstract / FREE Full Text 27. ↵ Ingram , L. O. , Conway , T. , Clark , D. P. , Sewell , G. W. , & Preston , J. F . ( 1987 ). Genetic engineering of ethanol production in Escherichia coli . Applied and Environmental Microbiology , 53 ( 10 ), 2420 – 2425 . doi: 10.1128/aem.53.10.2420-2425.1987 OpenUrl Abstract / FREE Full Text 28. ↵ Inui , M. , Kawaguchi , H. , Murakami , S. , Vertès , A. A. , & Yukawa , H . ( 2004 ). Metabolic Engineering of Corynebacterium glutamicum for Fuel Ethanol Production under Oxygen-Deprivation Conditions . Microbial Physiology , 8 ( 4 ), 243 – 254 . doi: 10.1159/000086705 OpenUrl CrossRef PubMed 29. ↵ Jacobson , T. B. , Adamczyk , P. A. , Stevenson , D. M. , Regner , M. , Ralph , J. , Reed , J. L. , & Amador-noguez , D . ( 2019 ). 2H and 13 C metabolic flux analysis elucidates in vivo thermodynamics of the ED pathway in Zymomonas mobilis . Metabolic Engineering , 54 ( May ), 301 – 316 . doi: 10.1016/j.ymben.2019.05.006 OpenUrl CrossRef PubMed 30. ↵ Jacobson , T. B. , Korosh , T. K. , Stevenson , D. M. , Foster , C. , Maranas , C. , Olson , D. G. , Lynd , L. R. , & Amador-Noguez , D . ( 2020 ). In Vivo Thermodynamic Analysis of Glycolysis in Clostridium thermocellum and Thermoanaerobacterium saccharolyticum Using 13 C and 2 H Tracers . mSystems , 5 ( 2 ), e00736 – 19 . doi: 10.1128/mSystems.00736-19 OpenUrl CrossRef PubMed 31. ↵ Jilani , S. B. , Venigalla , S. S. K. , Mattam , A. J. , Dev , C. , & Yazdani , S. S . ( 2017 ). Improvement in ethanol productivity of engineered E. coli strain SSY13 in defined medium via adaptive evolution . Journal of Industrial Microbiology and Biotechnology , 44 ( 9 ), 1375 – 1384 . doi: 10.1007/s10295-017-1966-4 OpenUrl CrossRef 32. ↵ Jojima , T. , Noburyu , R. , Sasaki , M. , Tajima , T. , Suda , M. , Yukawa , H. , & Inui , M . ( 2015 ). Metabolic engineering for improved production of ethanol by Corynebacterium glutamicum . Applied Microbiology and Biotechnology , 99 ( 3 ), 1165 – 1172 . doi: 10.1007/s00253-014-6223-4 OpenUrl CrossRef 33. ↵ Kalnenieks , U . ( 2006 ). Physiology of Zymomonas mobilis: Some Unanswered Questions . In Advances in Microbial Physiology (Vol. 51 , pp. 73 – 117 ). Elsevier. doi: 10.1016/S0065-2911(06)51002-1 OpenUrl CrossRef PubMed 34. ↵ Kanehisa , M . ( 2000 ). KEGG: Kyoto Encyclopedia of Genes and Genomes . Nucleic Acids Research , 28 ( 1 ), 27 – 30 . doi: 10.1093/nar/28.1.27 OpenUrl CrossRef PubMed Web of Science 35. ↵ Kanehisa , M . ( 2019 ). Toward understanding the origin and evolution of cellular organisms . Protein Science , 28 ( 11 ), 1947 – 1951 . doi: 10.1002/pro.3715 OpenUrl CrossRef PubMed 36. ↵ Kanehisa , M. , Furumichi , M. , Sato , Y. , Matsuura , Y. , & Ishiguro-Watanabe , M . ( 2025 ). KEGG: Biological systems database as a model of the real world . Nucleic Acids Research , 53 ( D1 ), D672 – D677 . doi: 10.1093/nar/gkae909 OpenUrl CrossRef PubMed 37. ↵ Lee , Y. , Jain , M. K. , Susan , C. E. , & Gregory Zeikus , J . ( 1993 ). Taxonomic Distinction of Saccharolytic Thermophilic Anaerobes . INTERNATIONAL JOURNAL OF SYSTEMATIC BACTERIOLOGY , 43 ( 1 ), 41 – 51 . OpenUrl CrossRef 38. ↵ Lynd , L. R. , Beckham , G. T. , Guss , A. M. , Jayakody , L. N. , Karp , E. M. , Maranas , C. , McCormick , R. L. , Amador-Noguez , D. , Bomble , Y. J. , Davison , B. H. , Foster , C. , Himmel , M. E. , Holwerda , E. K. , Laser , M. S. , Ng , C. Y. , Olson , D. G. , Román-Leshkov , Y. , Trinh , C. T. , Tuskan , G. A. , … Wyman , C. E . ( 2022 ). Toward low-cost biological and hybrid biological/catalytic conversion of cellulosic biomass to fuels . Energy & Environmental Science , 15 ( 3 ), 938 – 990 . doi: 10.1039/D1EE02540F OpenUrl CrossRef 39. ↵ Martinez , A. , Grabar , T. B. , Shanmugam , K. T. , Yomano , L. P. , York , S. W. , & Ingram , L. O . ( 2007 ). Low salt medium for lactate and ethanol production by recombinant Escherichia coli B . Biotechnology Letters , 29 ( 3 ), 397 – 404 . doi: 10.1007/s10529-006-9252-y OpenUrl CrossRef PubMed 40. ↵ Millar , D. G. , Griffiths-Smith , K. , Algar , E. , & Scopes , R. K . ( 1982 ). Activity and stability of glycolytic enzymes in the presence of ethanol . Biotechnology Letters , 4 ( 9 ), 601 – 606 . doi: 10.1007/BF00127792 OpenUrl CrossRef 41. ↵ Milo , R. , Jorgensen , P. , Moran , U. , Weber , G. , & Springer , M . ( 2009 ). BioNumbers The database of key numbers in molecular and cell biology . Nucleic Acids Research , 38 , 750 – 753 . doi: 10.1093/nar/gkp889 OpenUrl CrossRef PubMed Web of Science 42. ↵ Noor , E. , Bar-Even , A. , Flamholz , A. , Reznik , E. , Liebermeister , W. , & Milo , R . ( 2014 ). Pathway Thermodynamics Highlights Kinetic Obstacles in Central Metabolism . PLoS Computational Biology , 10 ( 2 ), e1003483 . doi: 10.1371/journal.pcbi.1003483 OpenUrl CrossRef 43. ↵ Ohta , K. , Beall , D. S. , Mejia , J. P. , Shanmugam , K. T. , & Ingram , L. O . ( 1991 ). Genetic improvement of Escherichia coli for ethanol production: Chromosomal integration of Zymomonas mobilis genes encoding pyruvate decarboxylase and alcohol dehydrogenase II . Applied and Environmental Microbiology , 57 ( 4 ), 893 – 900 . doi: 10.1128/aem.57.4.893-900.1991 OpenUrl Abstract / FREE Full Text 44. ↵ Osman , Y. A. , & Ingram , L. O . ( 1985 ). Mechanism of ethanol inhibition of fermentation in Zymomonas mobilis CP4 . Journal of Bacteriology , 164 ( 1 ), 173 – 180 . doi: 10.1128/jb.164.1.173-180.1985 OpenUrl Abstract / FREE Full Text 45. ↵ Panesar , P. S. , Marwaha , S. S. , & Kennedy , J. F . ( 2006 ). Zymomonas mobilis : An alternative ethanol producer . Journal of Chemical Technology & Biotechnology , 81 ( 4 ), 623 – 635 . doi: 10.1002/jctb.1448 OpenUrl CrossRef Web of Science 46. ↵ L. Olsson Rogers , P. L. , Jeon , Y. J. , Lee , K. J. , & Lawford , H. G. ( 2007 ). Z ymomonas mobilis for Fuel Ethanol and Higher Value Products . In L. Olsson (Ed.), Biofuels (Vol. 108, pp. 263 – 288 ). Springer Berlin Heidelberg . doi: 10.1007/10_2007_060 OpenUrl CrossRef PubMed 47. ↵ Rogers , P. L. , Lee , K. J. , Skotnicki , M. L. , & Tribe , D. E . ( 1982 ). Ethanol Production by Zymomonas mobilis . Advances in Biochemical Engineering , 23 , 37 – 84 . OpenUrl 48. Rogers , P. L. , Lee , K. J. , & Tribe , D. E . ( 1979 ). Kinetics of alcohol production by zymomonas mobilis at high sugar concentrations . Biotechnology Letters , 1 ( 4 ), 165 – 170 . doi: 10.1007/BF01388142 OpenUrl CrossRef 49. ↵ Seo , J.-S. , Chong , H. , Park , H. S. , Yoon , K.-O. , Jung , C. , Kim , J. J. , Hong , J. H. , Kim , H. , Kim , J.-H. , Kil , J.-I. , Park , C. J. , Oh , H.-M. , Lee , J.-S. , Jin , S.-J. , Um , H.-W. , Lee , H.-J. , Oh , S.-J. , Kim , J. Y. , Kang , H. L. , … Kang , H. S . ( 2005 ). The genome sequence of the ethanologenic bacterium Zymomonas mobilis ZM4 . Nature Biotechnology , 23 ( 1 ), 63 – 68 . doi: 10.1038/nbt1045 OpenUrl CrossRef PubMed Web of Science 50. ↵ Sharma , B. D. , Hon , S. , Thusoo , E. , Stevenson , D. M. , Amador-Noguez , D. , Guss , A. M. , Lynd , L. R. , & Olson , D. G . ( 2024 ). Pyrophosphate-free glycolysis in Clostridium thermocellum increases both thermodynamic driving force and ethanol titers . Biotechnology for Biofuels and Bioproducts , 17 ( 1 ), 146 . doi: 10.1186/s13068-024-02591-5 OpenUrl CrossRef 51. ↵ Sharma , B. D. , & Olson , D. G . ( 2025 ). Intracellular Metabolite Extraction , Peak Detection with El-MAVEN and Quantification Using Python V . 2 . doi: 10.17504/protocols.io.kqdg3wjbev25/v2 OpenUrl CrossRef 52. ↵ Sharma , B. D. , Olson , D. G. , Giannone , R. J. , Hettich , R. L. , & Lynd , L. R . ( 2023 ). Characterization and Amelioration of Filtration Difficulties Encountered in Metabolomic Studies of Clostridium thermocellum at Elevated Sugar Concentrations . Applied and Environmental Microbiology , 89 ( 4 ), e00406 – 23 . doi: 10.1128/aem.00406-23 OpenUrl CrossRef PubMed 53. ↵ Shaw , A. J. , Covalla , S. F. , Hogsett , D. A. , & Herring , C. D . ( 2011 ). Marker Removal System for Thermoanaerobacterium saccharolyticum and Development of a Markerless Ethanologen . Applied and Environmental Microbiology , 77 ( 7 ), 2534 – 2536 . doi: 10.1128/AEM.01731-10 OpenUrl Abstract / FREE Full Text 54. ↵ Shaw , A. J. , Podkaminer , K. K. , Desai , S. G. , Bardsley , J. S. , Rogers , S. R. , Thorne , P. G. , Hogsett , D. A. , & Lynd , L. R . ( 2008 ). Metabolic engineering of a thermophilic bacterium to produce ethanol at high yield . Proceedings of the National Academy of Sciences , 105 ( 37 ), 13769 – 13774 . doi: 10.1073/pnas.0801266105 OpenUrl Abstract / FREE Full Text 55. ↵ Sprenger , G. A . ( 1996 ). Carbohydrate metabolism in Zymomonas mobilis : A catabolic highway with some scenic routes . FEMS Microbiology Letters , 145 ( 3 ), 301 – 307 . doi: 10.1111/j.1574-6968.1996.tb08593.x OpenUrl CrossRef Web of Science 56. ↵ Tao , Z. , Yuan , H. , Liu , M. , Liu , Q. , Zhang , S. , Liu , H. , Jiang , Y. , Huang , D. , & Wang , T . ( 2023 ). Yeast Extract: Characteristics, Production, Applications and Future Perspectives . Journal of Microbiology and Biotechnology , 33 ( 2 ), 151 – 166 . doi: 10.4014/jmb.2207.07057 OpenUrl CrossRef PubMed 57. ↵ Tian , L. , Perot , S. J. , Stevenson , D. , Jacobson , T. , Lanahan , A. A. , Amador-Noguez , D. , Olson , D. G. , & Lynd , L. R . ( 2017 ). Metabolome analysis reveals a role for glyceraldehyde 3-phosphate dehydrogenase in the inhibition of C. thermocellum by ethanol . Biotechnology for Biofuels , 10 ( 1 ), 276 . doi: 10.1186/s13068-017-0961-3 OpenUrl CrossRef PubMed 58. ↵ Trinh , C. T. , Huffer , S. , Clark , M. E. , Blanch , H. W. , & Clark , D. S . ( 2010 ). Elucidating Mechanisms of Solvent Toxicity in Ethanologenic Escherichia coli . 106 ( 5 ), 721 – 730 . doi: 10.1002/bit.22743 OpenUrl CrossRef PubMed 59. ↵ Trinh , C. T. , & Srienc , F . ( 2009 ). Metabolic Engineering of Escherichia coli for Efficient Conversion of Glycerol to Ethanol . Applied and Environmental Microbiology , 75 ( 21 ), 6696 – 6705 . doi: 10.1128/AEM.00670-09 OpenUrl Abstract / FREE Full Text 60. ↵ Trinh , C. T. , Unrean , P. , & Srienc , F . ( 2008 ). Minimal Escherichia coli Cell for the Most Efficient Production of Ethanol from Hexoses and Pentoses . Applied and Environmental Microbiology , 74 ( 12 ), 3634 – 3643 . doi: 10.1128/AEM.02708-07 OpenUrl Abstract / FREE Full Text 61. ↵ Tsakraklides , V. , Shaw , A. J. , Miller , B. B. , Hogsett , D. A. , & Herring , C. D . ( 2012 ). Carbon catabolite repression in Thermoanaerobacterium saccharolyticum . Biotechnology for Biofuels , 5 ( 1 ), 85 . doi: 10.1186/1754-6834-5-85 OpenUrl CrossRef PubMed 62. ↵ Yomano , L. P. , York , S. W. , & Ingram , L. O . ( 1998 ). Isolation and characterization of ethanol-tolerant mutants of Escherichia coli KO11 for fuel ethanol production . Journal of Industrial Microbiology and Biotechnology , 20 ( 2 ), 132 – 138 . doi: 10.1038/sj.jim.2900496 OpenUrl CrossRef PubMed Web of Science 63. ↵ Yomano , L. P. , York , S. W. , Zhou , S. , Shanmugam , K. T. , & Ingram , L. O . ( 2008 ). Re-engineering Escherichia coli for ethanol production . Biotechnology Letters , 30 ( 12 ), 2097 – 2103 . doi: 10.1007/s10529-008-9821-3 OpenUrl CrossRef PubMed 64. ↵ Zheng , T. , Olson , D. G. , Murphy , S. J. , Shao , X. , Tian , L. , & Lynd , L. R . ( 2017 ). Both adhE and a Separate NADPH-Dependent Alcohol Dehydrogenase Gene, adhA , Are Necessary for High Ethanol Production in Thermoanaerobacterium saccharolyticum . Journal of Bacteriology , 199 ( 3 ). doi: 10.1128/JB.00542-16 OpenUrl Abstract / FREE Full Text View the discussion thread. Back to top Previous Next Posted November 24, 2025. Download PDF 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 Metabolic imbalance limits fermentation in microbes engineered for high-titer ethanol production 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 Metabolic imbalance limits fermentation in microbes engineered for high-titer ethanol production Bishal Dev Sharma , Eashant Thusoo , David M. Stevenson , Daniel Amador-Noguez , Lee R. Lynd , Daniel G. Olson bioRxiv 2025.11.21.689677; doi: https://doi.org/10.1101/2025.11.21.689677 Share This Article: Copy Citation Tools Metabolic imbalance limits fermentation in microbes engineered for high-titer ethanol production Bishal Dev Sharma , Eashant Thusoo , David M. Stevenson , Daniel Amador-Noguez , Lee R. Lynd , Daniel G. Olson bioRxiv 2025.11.21.689677; doi: https://doi.org/10.1101/2025.11.21.689677 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 Systems Biology Subject Areas All Articles Animal Behavior and Cognition (7618) Biochemistry (17633) Bioengineering (13856) Bioinformatics (41841) Biophysics (21399) Cancer Biology (18529) Cell Biology (25422) Clinical Trials (138) Developmental Biology (13352) Ecology (19860) Epidemiology (2067) Evolutionary Biology (24282) Genetics (15582) Genomics (22462) Immunology (17700) Microbiology (40295) Molecular Biology (17140) Neuroscience (88419) Paleontology (666) Pathology (2823) Pharmacology and Toxicology (4813) Physiology (7632) Plant Biology (15107) Scientific Communication and Education (2042) Synthetic Biology (4284) Systems Biology (9808) Zoology (2267)