Computational prediction and experimental verification of the gene encoding the NAD+/NADP+-dependent succinate semialdehyde dehydrogenase in Escherichia coli

Abstract

Although NAD(+)-dependent succinate semialdehyde dehydrogenase activity was first described in Escherichia coli more than 25 years ago, the responsible gene has remained elusive so far. As an experimental proof of concept for a gap-filling algorithm for metabolic networks developed earlier, we demonstrate here that the E. coli gene yneI is responsible for this activity. Our biochemical results demonstrate that the yneI-encoded succinate semialdehyde dehydrogenase can use either NAD(+) or NADP(+) to oxidize succinate semialdehyde to succinate. The gene is induced by succinate semialdehyde, and expression data indicate that yneI plays a unique physiological role in the general nitrogen metabolism of E. coli. In particular, we demonstrate using mutant growth experiments that the yneI gene has an important, but not essential, role during growth on arginine and probably has an essential function during growth on putrescine as the nitrogen source. The NADP(+)-dependent succinate semialdehyde dehydrogenase activity encoded by the functional homolog gabD appears to be important for nitrogen metabolism under N limitation conditions. The yneI-encoded activity, in contrast, functions primarily as a valve to prevent toxic accumulation of succinate semialdehyde. Analysis of available genome sequences demonstrated that orthologs of both yneI and gabD are broadly distributed across phylogenetic space.

FIG. 1.

Biosynthesis and degradation metabolism of arginine and putrescine in E. coli. The EC 1.2.1.24 orphan activity is indicated by a shaded box. Genes highly expressed with the yneI (sad) gene under stress conditions in which the DNA-stabilizing gene hupB is knocked out (hupB_KOPG) (see text for details) (17) are indicated by bold italics. Genes highly coexpressed with the gabD gene under nitrogen limitation conditions (40, 47) are indicated by underlined italics. TCA, tricarboxylic acid.

FIG. 2.

Growth of E. coli wild-type strain BW25113 (▵) and the gabD (□) and yneI (○) mutants on M9 minimal medium containing 1% (vol/vol) LB, 20 mM glycerol as the carbon source, and 5 mM GABA as the nitrogen source. Growth of E. coli wild-type strain BW25113 without GABA was used as a control (▴). OD₆₀₀, optical density at 600 nm.

FIG. 3.

Growth rates (A) and approximate lag phases (B) of E. coli wild-type strain BW25113 (black bars) and the gabD (light gray bars) and yneI (dark gray bars) mutants on M9 minimal medium containing glycerol as the carbon source and 0, 1, 2, 3, 4, or 5 mM exogenous SSA. The lag phases are expressed relative to the results obtained without exogenous SSA (Reference). ND, not determined due to the absence of growth.

FIG. 4.

SSADH activities in crude cell extracts harvested from growth on M9 minimal medium with glycerol and 0 or 2 mM exogenous SSA. The values are means ± standard deviations for both NAD⁺ and NADP⁺.

FIG. 5.

Sodium dodecyl sulfate-polyacrylamide gel electrophoresis gel showing purification of the His-tagged YneI and GabD dehydrogenases. Lanes M, molecular weight marker; lane 1, wash fraction of nonspecific bound cell protein from GabD overexpression; lane 2, elution fraction containing purified GabD (predicted molecular mass of the monomer, 51.7 kDa); lane, 3, wash fraction of nonspecific bound cell protein from YneI overexpression; lane 4, elution fraction containing purified YneI (predicted molecular mass of the monomer, 49.7 kDa).

FIG. 6.

Growth of E. coli wild-type strain BW25113 (•) and the gabD (▾) and yneI (▪) mutants on M9 minimal medium containing 5 g/liter glucose as the carbon source and 5 mM arginine as the nitrogen source. OD₆₀₀, optical density at 600 nm.