CoChat

Proteome constraints reveal targets for improving microbial fitness in nutrient‐rich environments

Open access

Published version of a preprint

The DOI matches its Crossref record; 54 of the 54 references it cites were found in a registry or an index.

Abstract

Cells adapt to different conditions via gene expression that tunes metabolism for maximal fitness. Constraints on cellular proteome may limit such expression strategies and introduce trade‐offs. Resource allocation under proteome constraints has explained regulatory strategies in bacteria. It is unclear, however, to what extent these constraints can predict evolutionary changes, especially for microorganisms that evolved under nutrient‐rich conditions, i.e., multiple available nitrogen sources, such as Lactococcus lactis . Here, we present a proteome‐constrained genome‐scale metabolic model of L. lactis (pcLactis) to interpret growth on multiple nutrients. Through integration of proteomics and flux data, in glucose‐limited chemostats, the model predicted glucose and arginine uptake as dominant constraints at low growth rates. Indeed, glucose and arginine catabolism were found upregulated in evolved mutants. At high growth rates, pcLactis correctly predicted the observed shutdown of arginine catabolism because limited proteome availability favored lactate for ATP production. Thus, our model‐based analysis is able to identify and explain the proteome constraints that limit growth rate in nutrient‐rich environments and thus form targets of fitness improvement.

Overview

The claims and statistics of this paper haven’t been checked yet.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.15252/msb.202010093

Cites

54 of 54 references verified0 citation errors · 0 not checkable

Statistics

Figures

Figure 1 Overview of pcLactis and simulations of glucose‐limited conditions The model explicitly accounts for reactions of metabolism and gene expression processes. Metabolic reactions produce metabolites and energy for not only biomass formation but also gene expression processes. The reactions of gene expression, on the other hand, synthesize RNA and proteins, which catalyze reactions of metabolism and gene expression processes as machineries or enzymes. In addition, pcLactis accounts for degradation of mRNA and proteins as well as dilution of biomass constituents during cell division.Coupling constraints in pcLactis. The coupling constraint allows for relating the reaction rate to the synthesis rate of its catalyst based on the reaction rate equation and steady‐state assumption, where turnover rates k cat of metabolic enzymes, catalytic rates k of machineries, and degradation constants k deg of the catalysts are needed.Simulated exchange fluxes compared with experimentally measured data (Goel et al, 2015).Simulated saturation of glucose transporter.Simulated inactive enzyme. The inactive enzyme is the sum of the enzymes that are synthesized but do not carry fluxes. The production of the inactive enzyme indicates that total proteome is not constrained. The gray area represents where the proteome switch occurs in experiments (Goel et al, 2015), which is between 0.5 and 0.6/h.Sensitivity analysis for glucose transporter and modeled proteome at different growth rates. Color represents the sensitivity score. A higher score indicates a greater impact of a given increase in the constraint on growth rate.
Figure 1 Overview of pcLactis and simulations of glucose‐limited conditions The model explicitly accounts for reactions of metabolism and gene expression processes. Metabolic reactions produce metabolites and energy for not only biomass formation but also gene expression processes. The reactions of gene expression, on the other hand, synthesize RNA and proteins, which catalyze reactions of metabolism and gene expression processes as machineries or enzymes. In addition, pcLactis accounts for degradation of mRNA and proteins as well as dilution of biomass constituents during cell division.Coupling constraints in pcLactis. The coupling constraint allows for relating the reaction rate to the synthesis rate of its catalyst based on the reaction rate equation and steady‐state assumption, where turnover rates k cat of metabolic enzymes, catalytic rates k of machineries, and degradation constants k deg of the catalysts are needed.Simulated exchange fluxes compared with experimentally measured data (Goel et al, 2015).Simulated saturation of glucose transporter.Simulated inactive enzyme. The inactive enzyme is the sum of the enzymes that are synthesized but do not carry fluxes. The production of the inactive enzyme indicates that total proteome is not constrained. The gray area represents where the proteome switch occurs in experiments (Goel et al, 2015), which is between 0.5 and 0.6/h.Sensitivity analysis for glucose transporter and modeled proteome at different growth rates. Color represents the sensitivity score. A higher score indicates a greater impact of a given increase in the constraint on growth rate.CC BY · PubMed Central
Figure 2 Amino acid analysis with pcLactis shows importance of amino acid catabolism for growth as a function of active constraints Simulated fluxes of amino acids compared with experimentally measured data. For each amino acid, simulated fluxes of uptake and tRNA charging are displayed together with the measured uptake flux. The upper bound on the uptake is also displayed, which is the linear trendline of the experimentally measured data if the amino acid is consumed rather than being secreted. The measured data are from the study (Goel et al, 2015).Scaled reduced cost analysis for uptake rates of amino acids at different growth rates. Color represents the scaled reduced cost value. A higher value indicates a greater impact of a given increase in the amino acid uptake on growth rate.
Figure 2 Amino acid analysis with pcLactis shows importance of amino acid catabolism for growth as a function of active constraints Simulated fluxes of amino acids compared with experimentally measured data. For each amino acid, simulated fluxes of uptake and tRNA charging are displayed together with the measured uptake flux. The upper bound on the uptake is also displayed, which is the linear trendline of the experimentally measured data if the amino acid is consumed rather than being secreted. The measured data are from the study (Goel et al, 2015).Scaled reduced cost analysis for uptake rates of amino acids at different growth rates. Color represents the scaled reduced cost value. A higher value indicates a greater impact of a given increase in the amino acid uptake on growth rate.CC BY · PubMed Central
Figure 3 The switches between ATP yielding pathways in L. lactis can be explained from a protein efficiency point of view A small model derived from pcLactis to investigate primary ATP‐producing pathways in L. lactis. The small model consists of three independent pathways, including glycolysis pathway followed by mixed acid fermentation, glycolysis pathway followed by lactic acid fermentation, and arginine catabolism.Parameters of ATP‐producing pathways inferred from pcLactis. ATP yield is ATP generated per glucose (for pathway 1 and 2) or arginine (for pathway 3) consumed. Note that the ATP yield of the arginine pathway is 1.67, which is due to the fact that besides one mole of ATP produced by the carbamate kinase reaction, by degrading one mole of arginine L. lactis does not have to export two moles of proton, which equivalently saves 0.67 mole of ATP that should have been consumed by ATP synthase to balance protons. Protein cost is protein mass (unit: g/gCDW) required per substrate flux (unit: mmol/gCDW/h) through the pathway. Protein efficiency is ATP generated per protein mass per time with the unit of mmol of ATP per gram of protein per hour.Linear programming to solve the small model. The objective function is to maximize the total ATP production flux JATP, which is the sum of ATP production fluxes of three pathways calculated by substrate uptake flux Ji times ATP yield YATP,i. The model is subject to two constraints. One is the total glucose uptake flux Jglc, which is the sum of the glucose fluxes toward pathway 1 J 1 and 2 J 2. The other is the total proteome constraint, which means that the sum of protein requirements of all pathways should be not greater than the proteome allocation in the model. For each pathway, the protein requirement is calculated by the protein cost pi times substrate flux Ji. In addition, there are lower and upper bounds on the uptake flux of each pathway. Specifically, J 1 and J 2 are unlimited while J 3 has a limited upper bound, which is assumed to be linearly correlated with total glucose uptake flux Jglc based on experimental observation.Small model simulations for a range of glucose uptake flux. The top three plots show that with increasing glucose uptake flux the decline in arginine uptake occurs firstly once the inactive enzyme disappears, and subsequently the switch from mixed to lactic acid fermentation. The inactive enzyme is the difference between the total protein requirements and the total proteome. The labels “A”, “B”, and “C” represent distinct phases based on changes in simulated fluxes through pathways with increasing glucose uptake flux: “A” corresponds to increase in both pathway 1 and 3, “B” decrease in pathway 3, and “C” decrease in pathway 1 while increase in pathway 2. The bottom three plots show sensitivities of glucose uptake, total proteome, and arginine uptake to the ATP production. The simulation data are reused as transparent lines in Appendix Fig S4.
Figure 3 The switches between ATP yielding pathways in L. lactis can be explained from a protein efficiency point of view A small model derived from pcLactis to investigate primary ATP‐producing pathways in L. lactis. The small model consists of three independent pathways, including glycolysis pathway followed by mixed acid fermentation, glycolysis pathway followed by lactic acid fermentation, and arginine catabolism.Parameters of ATP‐producing pathways inferred from pcLactis. ATP yield is ATP generated per glucose (for pathway 1 and 2) or arginine (for pathway 3) consumed. Note that the ATP yield of the arginine pathway is 1.67, which is due to the fact that besides one mole of ATP produced by the carbamate kinase reaction, by degrading one mole of arginine L. lactis does not have to export two moles of proton, which equivalently saves 0.67 mole of ATP that should have been consumed by ATP synthase to balance protons. Protein cost is protein mass (unit: g/gCDW) required per substrate flux (unit: mmol/gCDW/h) through the pathway. Protein efficiency is ATP generated per protein mass per time with the unit of mmol of ATP per gram of protein per hour.Linear programming to solve the small model. The objective function is to maximize the total ATP production flux JATP, which is the sum of ATP production fluxes of three pathways calculated by substrate uptake flux Ji times ATP yield YATP,i. The model is subject to two constraints. One is the total glucose uptake flux Jglc, which is the sum of the glucose fluxes toward pathway 1 J 1 and 2 J 2. The other is the total proteome constraint, which means that the sum of protein requirements of all pathways should be not greater than the proteome allocation in the model. For each pathway, the protein requirement is calculated by the protein cost pi times substrate flux Ji. In addition, there are lower and upper bounds on the uptake flux of each pathway. Specifically, J 1 and J 2 are unlimited while J 3 has a limited upper bound, which is assumed to be linearly correlated with total glucose uptake flux Jglc based on experimental observation.Small model simulations for a range of glucose uptake flux. The top three plots show that with increasing glucose uptake flux the decline in arginine uptake occurs firstly once the inactive enzyme disappears, and subsequently the switch from mixed to lactic acid fermentation. The inactive enzyme is the difference between the total protein requirements and the total proteome. The labels “A”, “B”, and “C” represent distinct phases based on changes in simulated fluxes through pathways with increasing glucose uptake flux: “A” corresponds to increase in both pathway 1 and 3, “B” decrease in pathway 3, and “C” decrease in pathway 1 while increase in pathway 2. The bottom three plots show sensitivities of glucose uptake, total proteome, and arginine uptake to the ATP production. The simulation data are reused as transparent lines in Appendix Fig S4.CC BY · PubMed Central
Figure 4 Proteome of the CcpA mutant, which has a higher fitness than wild‐type L. lactis, is changed in the direction of optimality AOverview of experiments for comparing wild‐type L. lactis MG1363 and CcpA‐mutant 445C1 in glucose‐limited chemostats at D = 0.5/h. Samples were taken for dry weight measurements, external metabolite and amino acid analysis, and proteome measurements.BProtein fold changes of the CcpA mutant over wild type. Log2 fold changes were calculated from LFQ values. P values were calculated based on 2‐sided two‐sample t‐tests with a FDR threshold of 0.05 and S0 equal to 1 (Tusher et al, 2001). Proteins outside the significance lines (FDR = 0.05, S0 = 1) are significantly changed in expression. Example sets of proteins are highlighted and are labeled with their gene names.C, D(C) Amino acid fluxes (mmol/gCDW/h) and (D) Residual concentration (mM) of arginine. The arginine uptake flux is increased to such extent that the residual arginine concentration became undetectable in the CcpA mutant. Data points represent the individual chemostat fluxes and concentrations; bars are average values (n = 4). Statistically significant differences between amino acid fluxes were calculated with two‐sided t‐tests and marked with * (P < 0.05).EProtein fractions. Protein fractions were obtained by normalizing iBAQ values by the total sum of all values. The obtained protein fractions were then averaged over the biological replicates (n = 4) for visualization. The fractions glucose transport, glycolysis, arginine pathway, and remaining fraction are significantly different between the two strains (P < 0.001, calculated with two‐sided t‐tests).
Figure 4 Proteome of the CcpA mutant, which has a higher fitness than wild‐type L. lactis, is changed in the direction of optimality AOverview of experiments for comparing wild‐type L. lactis MG1363 and CcpA‐mutant 445C1 in glucose‐limited chemostats at D = 0.5/h. Samples were taken for dry weight measurements, external metabolite and amino acid analysis, and proteome measurements.BProtein fold changes of the CcpA mutant over wild type. Log2 fold changes were calculated from LFQ values. P values were calculated based on 2‐sided two‐sample t‐tests with a FDR threshold of 0.05 and S0 equal to 1 (Tusher et al, 2001). Proteins outside the significance lines (FDR = 0.05, S0 = 1) are significantly changed in expression. Example sets of proteins are highlighted and are labeled with their gene names.C, D(C) Amino acid fluxes (mmol/gCDW/h) and (D) Residual concentration (mM) of arginine. The arginine uptake flux is increased to such extent that the residual arginine concentration became undetectable in the CcpA mutant. Data points represent the individual chemostat fluxes and concentrations; bars are average values (n = 4). Statistically significant differences between amino acid fluxes were calculated with two‐sided t‐tests and marked with * (P < 0.05).EProtein fractions. Protein fractions were obtained by normalizing iBAQ values by the total sum of all values. The obtained protein fractions were then averaged over the biological replicates (n = 4) for visualization. The fractions glucose transport, glycolysis, arginine pathway, and remaining fraction are significantly different between the two strains (P < 0.001, calculated with two‐sided t‐tests).CC BY · PubMed Central

Verification

Claims the paper makes

The claims and statistics of this paper haven’t been checked yet.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

A few free checks a day · papers already verified open instantly · results are public

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

The self-check needs the paper’s full text.

Citations

Cites

Cited by

See all 73

Metadata

Authors
Chen, Y., van Pelt‐KleinJan, E., van Olst, B., Douwenga, S., Boeren, S., Bachmann, H., Molenaar, D., Nielsen, J., & Teusink, B.
Affiliations
  • Chalmers University of Technology (SE)
  • TiFN (NL)
  • Novo Nordisk Foundation (DK)
Corresponding author
Teusink, Bas
Venue
Molecular Systems Biology
Published
2021-04-01
Volume
17
Issue
4
Pages
e10093-e10093
ISSN
1744-4292 (ISSN-L)
OpenAlex
W3151558128
Cited by
73
References
54
Citations by year
2026: 8 · 2025: 18 · 2024: 9 · 2023: 13 · 2022: 14 · 2021: 11
FWCI
3.67
Type
journal article
Language
en
Open access
Licence
Locations
  • Molecular Systems Biology (publishedVersion, other-oa)
  • bioRxiv (Cold Spring Harbor Laboratory) (acceptedVersion, cc-by)
  • PubMed Central (publishedVersion, cc-by)
  • PubMed (publishedVersion)
  • VU Research Portal (publishedVersion, other-oa)
  • Socio-Environmental Systems Modeling (acceptedVersion, cc-by)
  • Chalmers Research (Chalmers University of Technology) (submittedVersion, other-oa)
  • DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
  • Technical University of Denmark, DTU Orbit (Technical University of Denmark, DTU) (submittedVersion, other-oa)
  • Data Archiving and Networked Services (DANS) (submittedVersion, cc-by)
Funders
  • European Union's Horizon 2020 Research and Innovation Program (686070)
  • Novo Nordisk Foundation (NNF10CC1016517)
  • Netherlands Organisation for Scientific Research (ALWTF.2015.4)
  • Top‐sector Agri&Food (AF‐15503)
Topics
  • Life Sciences › Biochemistry, Genetics and Molecular Biology › Molecular Biology › Microbial Metabolic Engineering and Bioproduction
  • Life Sciences › Biochemistry, Genetics and Molecular Biology › Genetics › Bacterial Genetics and Biotechnology
  • Life Sciences › Biochemistry, Genetics and Molecular Biology › Molecular Biology › Gene Regulatory Network Analysis
Keywords
Lactococcus lactis, arginine catabolism, glucose catabolism, lactic acid production, growth rate, resource allocation trade-offs, genome-scale metabolic models, ATP production
MeSH terms
Adenosine Triphosphate, Arginine, Bacterial Proteins, Biological Evolution, Glucose, Models, Biological, Mutation, Lactococcus lactis, Reproducibility of Results, Proteome, Genetic Fitness

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review

Data sources

Bibliographic data: Semantic Scholar, OpenAlex, Crossref.

Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).

Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.