Biosynthesis of fosfomycin, re-examination and re-confirmation of a unique Fe(II)- and NAD(P)H-dependent epoxidation reaction

Abstract

(S)-2-Hydroxypropylphosphonic acid epoxidase (HppE) catalyzes the epoxide ring closure of (S)-HPP to form fosfomycin, a clinically useful antibiotic. Early investigation showed that its activity can be reconstituted with Fe(II), FMN, NADH, and O2 and identified HppE as a new type of mononuclear non-heme iron-dependent oxygenase involving high-valent iron-oxo species in the catalysis. However, a recent study showed that the Zn(II)-reconstituted HppE is active, and HppE exhibits modest affinity for FMN. Thus, a new mechanism is proposed in which the active site-bound Fe2+ or Zn2+ serves as a Lewis acid to activate the 2-OH group of (S)-HPP and the epoxide ring is formed by the attack of the 2-OH group at C-1 coupled with the transfer of the C-1 hydrogen as a hydride ion to the bound FMN. To distinguish between these mechanistic discrepancies, we re-examined the bioautography assay, the basis for the alternative mechanism, and showed that Zn(II) cannot replace Fe(II) in the HppE reaction and NADH is indispensable. Moreover, we demonstrated that the proposed role for FMN as a hydride acceptor is inconsistent with the finding that FMN cannot bind to HppE in the presence of substrate. In addition, using a newly developed HPLC assay, we showed that several non-flavin electron mediators could replace FMN in the HppE-catalyzed epoxidation. Taken together, these results do not support the newly proposed "nucleophilic displacement-hydride transfer" mechanism but are fully consistent with the previously proposed iron-redox mechanism for HppE catalysis, which is unique within the mononuclear non-heme iron enzyme superfamily.

Figure 1

Bioautography assay results of unfiltered reaction samples. A. (1) apo-HppE + Zn(II) + FMN + (S)-HPP; (2) fosfomycin standard; (3) apo-HppE + FMN + (S)-HPP; (4) apo-HppE + Fe(II) + FMN + (S)-HPP. B. (1) apo-HppE + Zn(II) + FMN + (S)-HPP + NADH; (2) fosfomycin standard; (3) apo-HppE + FMN + (S)-HPP + NADH; (4) apo-HppE + Fe(II) + FMN + (S)-HPP + NADH. A typical assay mixture (200 μL) contained 20 mM (S)-HPP (2), 90 μM apo-HppE, 90 μM FMN, and 150 μM of metal ions, in 20 mM Tris·HCl buffer, pH 7.5, in the presence or absence of 22.5 mM NADH (see Materials and Methods for details).

Figure 2

Bioautography assay results of filtered reaction samples. A. (1) apo-HppE + Zn(II) + FMN + (S)-HPP; (2) fosfomycin standard; (3) apo-HppE + Fe(II) + FMN + (S)-HPP. B. (1) apo-HppE + Zn(II) + FMN + (S)-HPP + NADH; (2) fosfomycin standard; (3) apo-HppE + Fe(II) + FMN + (S)-HPP + NADH. A typical assay mixture (200 μL) contained 20 mM (S)-HPP (2), 90 μM apo-HppE, 90 μM FMN, and 150 μM of metal ions, in 20 mM Tris·HCl buffer, pH 7.5, in the presence or absence of 22.5 mM NADH(see Materials and Methods for details).

Figure 3

Bioautography assay results of filtered reaction samples carried out in dark and under sunlight. (1) fosfomycin standard; (2) apo-HppE + Fe(II) + FMN + (S)-HPP, under sunlight; (3) apo-HppE + Fe(II) + FMN + (S)-HPP, in dark. A typical assay mixture (200 μL) contained 20 mM (S)-HPP (2), 90 μM apo-HppE, 90 μM FMN, and 150 μM of metal ions, in 20 mM Tris·HCl buffer, pH 7.5 (see Materials and Methods for details).

Figure 4

HPLC chromatogram of HppE catalyzed conversion of (S)-HPP (2) to fosfomycin (1). Inset shows the increasing of fosfomycin peak after 10, 20, 40, 60, and 80 min (from bottom to top). The starting concentrations were 50 μM HppE, 50 μM Fe²⁺, 75 μM riboflavin, 10 mM (S)-HPP (2), and 15 mM NADH (see Materials and Methods for details).

Figure 5

The NADH-dependence of the conversion of (S)-HPP (2) to fosfomycin (1) catalyzed by HppE (see Materials and Methods for details).

Scheme 1

Scheme 2

Scheme 3