Identification of glyA (encoding serine hydroxymethyltransferase) and its use together with the exporter ThrE to increase L-threonine accumulation by Corynebacterium glutamicum

Abstract

L-threonine can be made by the amino acid-producing bacterium Corynebacterium glutamicum. However, in the course of this process, some of the L-threonine is degraded to glycine. We detected an aldole cleavage activity of L-threonine in crude extracts with an activity of 2.2 nmol min(-1) (mg of protein)(-1). In order to discover the molecular reason for this activity, we cloned glyA, encoding serine hydroxymethyltransferase (SHMT). By using affinity-tagged glyA, SHMT was isolated and its substrate specificity was determined. The aldole cleavage activity of purified SHMT with L-threonine as the substrate was 1.3 micromol min(-1) (mg of protein)(-1), which was 4% of that with L-serine as substrate. Reduction of SHMT activity in vivo was obtained by placing the essential glyA gene in the chromosome under the control of P(tac), making glyA expression isopropylthiogalactopyranoside dependent. In this way, the SHMT activity in an L-threonine producer was reduced to 8% of the initial activity, which led to a 41% reduction in glycine, while L-threonine was simultaneously increased by 49%. The intracellular availability of L-threonine to aldole cleavage was also reduced by overexpressing the L-threonine exporter thrE. In C. glutamicum DR-17, which overexpresses thrE, accumulation of 67 mM instead of 49 mM L-threonine was obtained. This shows that the potential for amino acid formation can be considerably improved by reducing its intracellular degradation and increasing its export.

FIG. 1.

Pathways for l-threonine degradation. (A) Pathway initiated by threonine-3-dehydrogenase (TDH) activity, with subsequent CoA-dependent conversion of 2-amino-3-ketobutyrate by amino-ketobutyrate lyase (AKB-CoA lyase) or spontaneous decarboxylation of 2-amino-3-ketobutyrate. (B) l-Threonine cleavage by transaldolase (TA) or SHMT activity and the main activity of SHMT generating 5,10-methylene tetrahydrofolate (THF).

FIG. 2.

Purification of SHMT. A sodium dodecyl sulfate gel is shown, containing an extract of E. coli without induced glyA expression (lane 1), with induced expression (lane 2), or the flowthrough (lanes 3 and 4). Lane St, size standards, with sizes indicated on the right (in kilodaltons). Lanes E1 to E4, eluted SHMT protein.

FIG. 3.

Depletion construct pK18mobglyA′ to reduce expression of SHMT. Shown is the construct itself with the inducible tac promoter and the 3′ part of glyA. Furthermore, the recombination of pK18mobglyA′ with the chromosomal glyA sequences of strain DM326-2 is shown, and at the bottom is shown the resulting genomic organization in the strain enabling IPTG-dependent P_tac-driven expression of glyA.

FIG. 4.

Growth of strain DM326-2::pK18mobglyA′ on minimal medium without IPTG (○) or with 10 μM (×) or 100 μM (▪) IPTG and the control strain DM326-2 pZ1 without IPTG (•).

FIG. 5.

l-Threonine (solid bars) and glycine accumulation (open bars) with IPTG concentrations as indicated at 24, 48, and 72 h. (A) Product accumulation with DM326-2::pK18mobglyA′. (B) Productaccumulation with DM326-2::pK18mobglyA′/pEC-T18mob2thrE,overexpressing the exporter gene. Mean values and standard deviations for three independent cultures are presented.

FIG. 6.

Amino acid accumulation at 24, 48, and 72 h as a function of thrE expression with strain DR-17. Solid bars, strain DR-17/pEC-T18mob2thrE; open bars, control DR-17/pEC-T18mob2. Each value represents the average of two independent experiments.