Evolutionarily conserved structural motifs in bacterial GST (glutathione S-transferase) are involved in protein folding and stability

Abstract

The bacterium Proteus mirabilis expresses a cytosolic class beta glutathione S-transferase (PmGST B1-1) that is part of a family of multifunctional detoxication enzymes. Like other cytosolic GSTs, PmGST B1-1 possesses two local structural motifs, an N-capping box and a hydrophobic staple motif, both of which are located between amino acids 151 and 156. The N-capping box consists of a reciprocal hydrogen bonding interaction of Thr152 with Asp155, whereas the hydrophobic staple motif consists of a hydrophobic interaction between Phe151 and Ala156. By contrast with other GSTs, PmGST B1-1 displays distinct hydrogen bond interactions in the N-capping box. In mammalian GSTs these structural elements are critical for protein folding and stability. To investigate the role played by these two motifs in a distantly related organism on the evolutionary scale, site-directed mutagenesis was used to generate several mutants of both motifs in PmGST B1-1. All mutants were efficiently overexpressed and purified, but they were quite unstable, although at different levels, indicating that protein folding was significantly destabilized. The analysis of the T152A and D155G variants indicated that the N-capping box motif plays an important role in the stability and correct folding of the enzyme. The analysis of F151A and A156G mutants revealed that the hydrophobic staple motif influences the structural maintenance of the protein and is implicated in the folding process of PmGST B1-1. Finally, the replacement of Thr152 and Asp155, as well as Phe151 and Ala156 residues influences the catalytic efficiency of the enzyme.

Figure 1. Comparison of the amino acid sequence of PmGST B1-1 with representative eukaryotic and bacterial members of the cytosolic class GST

Sequences were aligned manually. The position of the first residue shown relative to the full-length protein is indicated to the left of each sequence. Sequences were retrieved from the NCBI data bases. The N-capping box residues are shown in grey boxes. The hydrophobic staple motif residues are shown in black boxes. The glycine residue located four residues before the N-cap is shown in grey. The nomenclature is N″-N′-Ncap-N1-N2-N3-N4 as proposed by Richardson and Richardson [14].

Figure 2. Stereoscopic view of the PmGST B1-1 region with residues that form the N-capping box and the hydrophobic staple motif

α1-helix (in blue) and α6-helix (in green) are shown in loop representation and key residues are shown as sticks. Thr¹⁵² and Asp¹⁵⁵ correspond to the N-cap and N3 residues, respectively. Broken lines indicate hydrogen bonds (in yellow). Oxygen and nitrogen atoms are indicated in red and blue respectively. Hydrophobic interactions formed by the Phe¹⁵¹ (N′) and Ala¹⁵⁶ (N4) residues, as well as between Ala¹⁵⁶ and α1-helix residues are emphasized by the Connolly surface representation (dot frame).

Figure 3. SDS/PAGE of total-cellular extracts from wild-type and mutant enzymes

Proteins were detected by Coomassie Brilliant Blue R-250 staining. Lane 1, molecular mass markers; lane 2, purified wild-type enzyme; lane 3, total cellular extract of non-transformed E. coli XL1Blue; lane 4, wild-type enzyme; lane 5, D155A; lane 6, D155G; lane7, T152A; lane 8, F151A; lane 9, A156G; lane 10, D155A/T152A; lane 11, F151A/A156G.

Figure 4. Fluorescence emission spectra of wild-type and mutant enzymes

Wild-type (●), D155G (▽), T152A (△), A156G (○) and F151A (□). Protein concentration was 3 μM in 10 mM phosphate buffer (pH 7.0) containing 1 mM EDTA. Excitation was at 280 nm.

Figure 5. Effects of temperature on the stability of wild-type and mutants

The enzyme activity at 25 °C was taken as 100%. Wild-type (●), D155G (▽), T152A (△), A156G (○) and F151A (□).

Figure 6. Refolding kinetics of wild-type (●) and D155G (▽), T152A (△), A156G (○) and F151A (□) mutants monitored by changes in enzymatic activity at different temperatures

All enzymes were unfolded in 4 M GdnHCl in 0.1 M potassium phosphate buffer, 1 mM EDTA and 5 mM DTT (pH 6.5) for 30 min. Refolding was initiated by rapid dilution in the same buffer. The residual concentration of GdnHCl was 0.1 M. Aliquots of the refolding reaction mixture were taken at different times and assayed for GST activity.