Predicting synthetic rescues in metabolic networks

Abstract

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 non-viable 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.

Figure 1

Schematic illustration of the consequences of gene deletion on the organism's growth rate. (A) The growth rate following the deletion of an enzyme-encoding gene often drops, but after many generations may recover to a new optimal value not very different from the original one (red line). The optimal growth rate before and after the deletion is predicted by FBA (black and green dotted lines). The blue line indicates the predicted buffering effect of additional gene deletions: by deleting appropriately selected additional genes, the suboptimal growth rate shortly after gene deletions is higher than without the rescue deletions. (B–E) The effect of rescue deletions on the fluxes of a metabolic network, where M₁ … M₄ represent metabolites and the width of the arrows represents the strength of individual fluxes.

Figure 2

Distribution of metabolic fluxes in the E. coli 's TCA cycle in arabinose minimal medium for (A) wild-type organism predicted by FBA, (B) fbaA mutant predicted by MOMA, (C) optimal state of fbaA mutant predicted by FBA, and (D) fbaA mutant with the rescue deletions of genes aceA and sucAB, predicted by MOMA. Key flux changes are highlighted in orange. Note that the metabolic flux pattern predicted by MOMA after the fbaA deletion (B) is similar to the wild-type fluxes (A). With the rescue deletions, however, MOMA-predicted fluxes (D) are brought closer to the FBA-predicted fluxes (C), restoring the organisms' ability to produce biomass. While we show a double deletion for its pedagogical value, we note that the deletion of aceA alone is sufficient to rescue the mutant (see Figure 3A) and that the mutant can also be rescued with other single-gene deletions (see Figure 4B and Supplementary Information).

Figure 3

The impact of rescue deletions. (A) Predicted biomass production for the fbaA mutant of E. coli in arabinose minimal medium as a function of the number of rescue deletions when starting with aceA and sucAB. Deleted rescue genes are indicated in the figure. (B) Biomass production of tpiA- and nuoA-deficient mutants in glucose minimal medium as a function of the number of individual rescue deletions. Deleted genes are indicated in the figure. The optimal biomass flux remains unchanged with the addition of rescue deletions. The biomass fluxes are normalized by the wild-type flux G^wt_FBA=0.745 mmol/g DW-h in (A) and 0.908 mmol/g DW-h in (B).

Figure 4

The impact of rescue deletions for E. coli (A, B) and S. cerevisiae (C, D) gene-deficient mutants. (A, C) Predicted biomass production before (○) and after (•) rescue deletions in glucose minimal media. The mutants are generated through the deletion of the genes shown at the x-axis. We show the results for all mutants with G¹_MOMA<G¹_FBA such that G¹_MOMA≤0.8 G^wt_FBA and G¹_FBA≥0.2 G^wt_FBA. If the rescue deletion changes the growth rate from zero to some positive value, we observe the Lazarus effect, applying to suboptimally essential genes (left). If the rescue deletion only enhances the growth rate, we observe a suboptimal recovery (right). The experimental information on the lethality of the original E. coli (Edwards and Palsson, 2000; Gerdes et al, 2003; Baba et al, 2006; PEC, 2007) and S. cerevisiae (Giaever et al, 2002; Steinmetz et al, 2002; SDG, 2007) gene-deficient mutants is indicated with (+) for viable mutants, (−) for non-viable mutants, and (^a) for a gene absent in the databases. (B, D) Same as in (A, C) for single-gene rescue deletions in various media. We show selected mutants with significant biomass improvements after the rescue deletion of a single gene. The rescue deletion is indicated at the top, and the tested media are indicated at the bottom. The abbreviations stand for acetate (Ac), α-ketoglutarate (Akg), arabinose (Ara), ethanol (Eth), galactose (Gal), glucose (Glc), glucose anaerobic (Glc^a), glycerol (Gly), lactate (Lac), malate (Mal), mannose (Man), pyruvate (Pyr), rich medium (Rich) (see Supplementary Information), sorbitol (Sor), succinate (Succ), sucrose (Suc), and xylose (Xyl). The biomass fluxes are normalized by the wild-type flux G^wt_FBA in all panels. In units of mmol/g DW-h, the wild-type fluxes for E. coli are 0.187 (Ac), 0.535 (Akg), 0.745 (Ara), 0.908 (Glc), 0.367 (Lac), 0.388 (Mal), 0.908 (Man), 0.303 (Pyr), 2.87 (Rich), 0.418 (Succ), and 1.37 (Suc), while for S. cerevisiae they are 0.189 (Ac), 0.311 (Eth), 0.703 (Gal), 0.819 (Glc), 0.180 (Glc^a), 0.532 (Gly), 1.34 (Rich), 0.798 (Sor), and 0.742 (Xyl). All the genes involved in the rescues of (A, C) are listed in Supplementary Information, while the minimum rescue sets are listed in Supplementary Tables SII and SIII, respectively. The alternative rescue genes for each media in (B, D) are listed along with the corresponding recoveries in Supplementary Information.

Figure 5

Experimental evidence that gene deletions can enhance suboptimal growth rates: growth rate before (○) and after (•) gene deletions for (A) E. coli MG1655 (Fong and Palsson, 2004) and (B) B. subtilis 168 (Fischer and Sauer, 2005). The deleted genes are indicated at the top. All genes in (A) are involved in the catalysis of central metabolic reactions, and growth is measured after 10 days in α-ketoglutarate (Akg), glucose (Glc), glycerol (Gly), lactose (Lac), malate (Mal), and ribose (Rib) media. The carbon source in (B) is glucose.