Substitution of the Methionine Axial Ligand of the T1 Copper for the Fungal-like Phenylalanine Ligand (M298F) Causes Local Structural Perturbations that Lead to Thermal Instability and Reduced Catalytic Efficiency of the Small Laccase from Streptomyces coelicolor A3(2)

Abstract

Many industrial processes operate at elevated temperatures or within broad pH and salinity ranges. However, the utilization of enzymes to carry out biocatalysis in such processes is often impractical or even impossible. Laccases (EC 1.10.3.2), which constitute a large family of multicopper oxidases, have long been used in the industrial setting. Although fungal laccases are in many respects considered superior to their bacterial counterparts, the bacterial laccases have been receiving greater attention recently. Albeit lower in redox potential than fungal laccases, bacterial laccases are commonly thermally more stable, act within broader pH ranges, do not contain posttranslational modifications, and could therefore serve as a high potential scaffold for directed evolution for the production of enzymes with enhanced properties. Several examples focusing on the axial ligand mutations of the T1 copper site have been published in the past. However, structural evidence on the local and global changes induced by those mutations have thus far been of computational nature only. In this study, we set out to structurally and kinetically characterize a few of the most commonly reported axial ligand mutations of a bacterial small laccase (SLAC) from Streptomyces coelicolor. While one of the mutations (Met to Leu) equips the enzyme with better thermal stability, the other (Met to Phe) induces an opposite effect. These mutations cause local structural rearrangement of the T1 site as demonstrated by X-ray crystallography. Our analysis confirms past findings that for SLACs, single point mutations that change the identity of the axial ligand of the T1 copper are not enough to provide a substantial increase in the catalytic efficiency but can in some cases have a detrimental effect on the enzyme's thermal stability parameters instead.

Figure 1

(A) Partial structure-based multiple sequence alignment of the mononuclear copper site of two bacterial (2-domain S. coelicolor and 3-domain B. subtilis) and two fungal laccases (2-domain from T. versicolor and C. cinereus). The axial ligand to the T1 copper is highlighted with a red box and its stabilizing interaction partner with a green box; (B) T1 copper interaction partners for the small and large laccases from bacteria and fungi; axial ligand and their stabilizing interaction partner are in red and green, respectively.

Figure 2

(A) From left to right: omit maps from the T1 copper site (contoured at 4σ above background) calculated with the Fourier coefficients (F_obs – F_calc) with phases from the final models but with the coordinates of the M198/M298/Cu (wild type, beige), L298/Cu (M289L mutant, magenta), F298/Cu (M298F mutant, green), and F198/M298/Cu (M198F/M298F double mutant, gold) omitted prior to calculations, respectively. The coordinates of the final models of mutants are superimposed with corresponding maps and with the coordinates of wild-type ScSLAC for comparison accuracy. Displacement distance between the C_α-s of the mutated axial ligand are marked with a contoured line. (B) Overlays of the poses for the T1 site of (1) TvLAC with M298F mutant (cyan-green), (2) CcLAC with M298L mutant (cyan-magenta), (3) TvLAC with M198F/M298F double mutant (cyan/gold), and (4) BsLAC with TvLAC (cyan-green).

Figure 3

Heat inactivation profiles as determined for wild type and axial ligands in the presence of ABTS substrate at 70 °C. Some level of protection against heat-inactivation can be achieved by mutating the axial Met (blue circles) to a Leu residue (pink circles). Double mutant of M198F/M298F (MFMF) is the most susceptible to heat inactivation.

Figure 4

Radii of gyration (R_g) with ± 1 σ for error bars for wild-type (WT) and mutants (M298F, M298L) with a side-by-side comparison of sonication (S) and the French press (F) treatments. The M298L mutant, which appears to have a slightly higher R_g than the other samples, responded differently to the two treatments.

Figure 5

Kratky plot of the wild type French pressed protein (WTF) compared to the M298L mutant prepared by sonication (MLS). Though the two curves differ slightly according to the radius and gyration and χ² (1.36), the profiles are difficult to distinguish at the current level of experimental noise.