Source-linked AI summary
Predicting synthetic rescues in metabolic networks
Adilson E. Motter, Natali Gulbahce, Eivind Almaas, Albert-Laszlo Barabasi
TL;DR
The paper asks whether additional network changes can restore growth lost after enzyme-gene deletions. Using flux-balance-based analysis, it identifies rescue deletions in microbial metabolic networks, including mutants whose predicted suboptimal growth is zero. The reported candidates and growth recovery vary with the tested medium.
Problem
The paper asks how network-based strategies might recover growth lost through defective genes.
Source: Introduction · p. 3Method
The authors use flux-balance-based approaches to identify additional enzyme-gene deletions that can restore or improve growth in gene-deficient mutants.
Source: Introduction · p. 3, Constraint-based approach · p. 9Results
In E. coli, 6 suboptimally essential genes and 17 suboptimal-recovery candidates were identified, and targeted deletions raised the growth rate of tested pfk, fbaA, and tpiA mutants from zero to more than 45% of wild-type growth.
Source: Results · pp. 4–5Takeaways & Limitations
Additional gene deletions are a candidate network-based route to restoring growth in mutants initially classified as nonviable.
Source: Introduction · p. 3The mutants identified and rescue deletions required depend on the tested medium.
Source: Results · p. 5
Abstract
from arXiv · showhide
An important goal of medical research is to develop methods to recover the loss of cellular function due to mutations and other defects. Many approaches based on gene therapy aim to repair the defective gene or to insert genes with compensatory function. Here, we propose an alternative, network-based strategy that aims to restore biological function by forcing the cell to either bypass the functions affected by the defective gene, or to compensate for the lost function. Focusing on the metabolism of single-cell organisms, we computationally study mutants that lack an essential enzyme, and thus are unable to grow or have a significantly reduced growth rate. We show that several of these mutants can be turned into viable organisms through additional gene deletions that restore their growth rate. In a rather counterintuitive fashion, this is achieved via additional damage to the metabolic network. Using flux balance-based approaches, we identify a number of synthetically viable gene pairs, in which the removal of one enzyme-encoding gene results in a nonviable phenotype, while the deletion of a second enzyme-encoding gene rescues the organism. The systematic network-based identification of compensatory rescue effects may open new avenues for genetic interventions.
Introduction
Cellular functions depend on networks of components, so genetic defects can disrupt modules rather than act locally. The paper proposes restoring growth in single-cell mutants through additional gene deletions, distinguishing mutants that cannot grow even at the predicted optimum from those that may recover after adaptation.
Introduction
- Network-based research motivates asking whether function lost through defective genes can be recovered, rather than focusing only on cell-killing effects of network damage.The authors frame this as a question about predicting recovery strategies for lost cellular function.Source: Introduction · p. 2
- A gene is optimally essential when its deletion yields zero growth at the new optimal steady state predicted by Flux Balance Analysis (FBA).After adaptation, mutants often converge toward an optimal growth rate predicted by FBA.Source: Introduction · p. 2
- A gene is suboptimally essential when FBA predicts nonzero optimal growth after its deletion but MOMA predicts zero growth shortly after deletion.Such mutants may be experimentally unable to grow initially, yet viable after their metabolism and regulatory system readjust.Source: Introduction · pp. 2–3
- The proposed network-based strategy restores growth in mutants lacking a suboptimally essential gene by deleting other enzyme-encoding genes, an effect termed the Lazarus effect.The authors also describe suboptimal recovery, in which additional deletions increase growth in mutants that remain viable.Source: Introduction · p. 3
Results
The results show that targeted gene deletions can restore or improve growth in enzyme-deficient mutants, including experimentally supported E. coli cases. The authors identify candidates across media and organisms and relate predicted rescue to measured growth and flux behavior.
Results
- The method compares suboptimal and optimal flux patterns to select deletions that suppress pathways active only in the suboptimal state.The proposed approach uses MOMA for the mutant’s suboptimal fluxes and FBA for its predicted optimal fluxes; rescue deletions can raise suboptimal growth without changing the optimal growth rate.Source: Results · pp. 3–4
- In arabinose-grown E. coli, deleting aceA and sucAB reroutes the non-growing fbaA mutant through the pentose phosphate pathway and restores biomass production.MOMA predicts zero suboptimal growth, while FBA predicts a nonzero optimal growth state.Source: Results · p. 4
- After aceA rescue, further deletions raised fbaA-mutant growth to about 67% of wild-type after 40 deletions; nuoA recovery approached 59% of its wild-type optimum.The tnaB deletion produced a large enhancement in the fbaA case.Source: Results · p. 4
- More than 45% of wild-type growth was restored from zero in pfk, fbaA, and tpiA mutants by targeted deletions.The analysis identified 6 suboptimally essential genes and 17 candidates for suboptimal recovery in E. coli grown in glucose minimal medium.Source: Results · pp. 4–5
- For nonviable tpiA mutants, six concurrent deletions shut down the glyoxylate pathway and restore growth; subsequent deletions can increase it further.The pathway’s initially nonzero flux in viable but incompletely adapted mutants converges to zero over weeks of adaptive evolution.Source: Results · p. 5
- 8, 21, and 25 E. coli mutants gained over 10% of wild-type growth after rescue in minimal acetate, minimal glucose, and rich media, respectively.Single-gene rescue combinations numbered 58 in H. pylori, 94 in E. coli, and 58 in S. cerevisiae; the candidate lists and required deletions depend on the medium.Source: Results · p. 5
- In glucose minimal medium, the authors predicted the Lazarus effect for 3 S. cerevisiae mutants and suboptimal recovery for 11 others.Two predicted Lazarus mutants were experimentally verified as nonviable; the third showed positive growth in experiments, possibly reflecting model incompleteness or incomplete adaptation.Source: Results · p. 5
- Experimental data support growth improvement after selected gene deletions, and the flux-based assumption correctly predicted 20 of 22 analyzed E. coli mutants.The growth improvements were observed in multiple environments, including E. coli and B. subtilis experiments.Source: Results · p. 6
Discussion
The discussion presents flux measurement and robustness as ways to support experimental testing of predicted rescue deletions. It connects synthetic rescue to examples beyond metabolism and suggests that targeted pathway blocking could offer an alternative to repairing faulty genes.
Discussion
- Suboptimal and near-optimal fluxes can be measured experimentally, for example with 13C-tracer techniques, to identify candidate rescue deletions while reducing modeling bias.Candidates typically involve reactions active in the suboptimal state but inactive in the optimal state.Source: Discussion · p. 6
- Rescue sets are not unique, and the number of combinations producing the same effect generally increases with set size; the authors report robustness to parameter choices.This flexibility may help address mismatches between modeled and real genetic or environmental conditions and side effects from unknown gene functions.Source: Discussion · p. 7
- Gene-deletion rescue also occurs beyond metabolism, and synthetically viable gene pairs may revise how gene essentiality is interpreted.The paper describes examples involving bacterial growth and mammalian knockouts; a lethal single deletion does not necessarily mean the gene is essential if additional deletions can restore growth.Source: Discussion · p. 7
- Blocking selected pathways may offer an alternative to repairing faulty genes, and the authors propose systematically finding rescue gene pairs and sets for therapeutic strategies.The discussion relates this possibility to examples of rescue through pathway rerouting or suppression of activity associated with a defective gene.Source: Discussion · pp. 7–8
Methods
The method models metabolic fluxes under steady-state stoichiometric and reaction-bound constraints, then compares FBA and MOMA states to identify rescue deletions. FBA supplies the optimal biomass-producing reference, while MOMA models a mutant state close to the original flux pattern.
Methods
- For m metabolites and n reactions, steady-state fluxes satisfy stoichiometric balance, while thermodynamics, substrate availability, and enzyme capacity bound each reaction.Unavailable substrates have zero uptake bounds.Source: Constraint-based approach · p. 9
- FBA maximizes biomass flux under these constraints, and deleting genes for enzymes in a reaction sets that reaction’s bounds to zero.Biomass is represented as an additional reaction.Source: Constraint-based approach · p. 9
- MOMA selects a constraint-compatible mutant flux state closest in Euclidean distance to the wild-type FBA state.Source: Constraint-based approach · p. 9
- Rescue deletions target genes that deactivate reactions active in the MOMA mutant state but inactive in the FBA optimum, shifting mutant fluxes closer to the optimum.The deletions are not predicted to change FBA fluxes or steady-state biomass production.Source: Identifying rescue gene deletions · p. 9
MOMA to G2
The procedure compares wild-type and mutant flux states, then selects and prunes additional gene deletions intended to improve suboptimal biomass production. Its target is biomass production itself, which the paper distinguishes from biomass yield.
MOMA to G2
- The approach increases biomass production, which the paper notes is not necessarily the same as biomass yield used in metabolic engineering studies.Source: MOMA to G2 · p. 10
- The procedure calculates wild-type and mutant FBA fluxes, then uses the wild-type FBA state as the reference for the mutant’s MOMA flux prediction.Source: MOMA to G2 · p. 10
- It continues to rescue selection when the mutant’s MOMA biomass flux is below its FBA-predicted biomass flux.Source: MOMA to G2 · p. 10
- The method targets reactions with nonzero MOMA flux but zero FBA flux, then finds a gene-consistent deletion set that deactivates those reactions.Source: MOMA to G2 · p. 10
- The rescue set is incrementally reduced by restoring the gene whose deletion contributes least to increased biomass flux.Source: MOMA to G2 · pp. 9–10