N-terminal engineering of glutamyl-tRNA reductase with positive charge arginine to increase 5-aminolevulinic acid biosynthesis

ABSTRACT

Five-Aminolevulinic acid (ALA), the universal precursor of all tetrapyrroles, has various applications in medicine and agriculture industries. Glutamyl-tRNA reductase (GluTR) as the first key enzyme of C5 pathway is feedback regulated by heme, and its N-terminus plays a critical role on its stability control. Here, the GluTR N-terminus was engineered by inserting different numbers of positively charged lysine and arginine residues. The results confirmed that insertion of lysine or arginine residues (especially one arginine residue) behind Thr2 significantly increased the stability of GluTR. By co-expression of the GluTR variant R1 and the glutamate-1-semialdehyde aminotransferase, ALA production was improved 1.76-fold to 1220 mg/L. The GluTR variant R1 constructed here could be used for engineering the C5 pathway to enhance ALA and other products.

KEYWORDS:

• Five-aminolevulinic acid
• escherichia coli
• glutamyl-tRNA reductase
• heme
• N-terminal engineering

Five-aminolevulinic acid (ALA) is a natural non-amino acid with 5 carbons. ALA is an important intermediate for biosynthesis of tetrapyrroles, such as heme, chlorophyll, cytochrome and vitamin B₁₂.¹ In fact, ALA has been widely used as photodynamic therapy, plant growth regulator, herbicide and insecticide in medicine and agriculture industry.^2-5 There are 2 different pathways for ALA biosynthesis.⁶ One pathway is the C4 pathway, in which glycine and succinyl-CoA are condensed to ALA catalyzed by ALA synthase (ALAS).⁷ The other pathway is the C5 pathway, in which glutamate as carbon skeleton is transformed into ALA through 3 enzymatic reactions that are catalyzed by glutamyl-tRNA synthetase (GluRS), glutamyl-tRNA reductase (GluTR) and glutamate-1-semialdehyde aminotransferase (GSA-AT).⁸

Many ALAS encoding sequences have been cloned, expressed and studied.¹ On this basis, the industrial workhorses Escherichia coli^9-16 and Corynebacterium glutamicum^17-18 have been used for ALA production by expressing ALAS with the C4 pathway. Specifically, after optimization of ALAS expression, cultivation process and ALA transport, the titer was eventually increased to 14.7 g/L.¹⁹ Even so, due to the addition of precursor glycine and succinic acid, it is still more attractive to produce ALA directly from glucose with a metabolic engineering strategy.^20-21

Previously, it has been demonstrated that the key enzyme GluTR of the C5 pathway is feed-back inhibited by heme.^22-23 Further studies confirmed that N-terminal domain of GluTR plays an important effect on the protein stability and insertion of 2 lysine residues with positive charge to the third and forth positions at the N-terminus of GluTR improved protein stabilization.²⁴ Accordingly, overexpression of the GluTR mutant with GSA-AT substantially enhanced ALA production.²⁰ Then, based on the fact that ALA dehydratase is feedback inhibited by the intermediate protoporphyrinogen IX,²⁵ both cell growth and ALA production was further improved with addition of Fe²⁺ in the minimal medium.²⁶ However, the stability of GluTR was still a rate-limiting step for achieving high titer of ALA.

In nature, GluTR interacts with GSA-AT to form a tight complex to catalyze glutamyl-tRNA to ALA via the unstable intermediate GSA.²⁷ Clearly, N-terminus of GluTR involves in interaction with GSA-AT and previous studies have confirmed that the N-terminal peptide plays a critical role on regulating the stability of GluTR during catalysis process (Fig. 1).²⁸ Consequently, in order to further improve the stability of the enzyme, different numbers of lysine were introduced at the N-terminus of GluTR, respectively (Fig. 2A). As shown in Fig. 3A, compared with the wild-type GluTR (690 mg/L), all the variants with inserting different numbers of lysine resulted in higher accumulation of ALA. Especially, the variant K2 with introduction of 2 lysine residues produced a higher titer (1120 mg/L) which was in consistent with previous results.²⁴

Figure 1. The model complex of GluTR and GSA-AT. The V-shaped dimer protein is GluTR while the other dimer is GSA-AT.

Figure 2. The illustration of GluTR variants with insertion of different numbers of lysine residues (A) and arginine residues (B). WT represented the wild-type GluTR.

Figure 3. ALA production of the variants with inserting different lysine residues (A) and arginine residues (B).

To further investigate the potential reasons, the structure of the variant K2 was simulated and compared with the wild-type GluTR. Clearly, insertion of 2 lysine residues generated more hydrogen bonds (Fig. 4AB), suggesting that the added positive amino acid residues might weaken the feedback-regulation caused by heme. Thus, the other positively charged amino acid residue arginine was also investigated (Fig. 2B). As expected, insertion of arginine with different numbers also resulted in similar results while the variant R1 produced the highest titer (1220 mg/L, Fig. 3B), which was 176.8% of that of the wild-type GluTR. At the same time, the expression level of GluTR with different arginine residues was investigated with SDS-PAGE analysis (Fig. 5). With the increase of arginine residues, the protein expression level was consistently enhanced. In combination with ALA production and structure analysis (Fig. 4C), it could be speculated that insertion of more than one arginine residues might weaken the interaction with GSA-AT or decrease the activity although with more stability. As a result, the variant R1 could be used to further optimize the heme biosynthesis pathway for enhancing ALA production.

Figure 4. Structural analysis of N-terminal domain of GluTR. (A) Wild-type GluTR; (B) the variant GluTR K2 with insertion of 2 lysine residues; (C) the variant GluTR R1 with insertion of one arginine residue. The green dotted lines indicate the hydrogen bonds.

Figure 5. SDS-PAGE analysis of the GluTR variants with insertion of arginine residues.

Disclosure of potential conflicts of interest

No potential conflicts of interest were disclosed.

Funding

This study was funded by the National Natural Science Foundation of China (31200020), the Program for Changjiang Scholars and Innovative Research Team in University (No. IRT_15R26), and the 111 Project.