Structural and functional perturbation of Giardia lamblia triosephosphate isomerase by modification of a non-catalytic, non-conserved region

Abstract

Background: We have previously proposed triosephosphate isomerase of Giardia lamblia (GlTIM) as a target for rational drug design against giardiasis, one of the most common parasitic infections in humans. Since the enzyme exists in the parasite and the host, selective inhibition is a major challenge because essential regions that could be considered molecular targets are highly conserved. Previous biochemical evidence showed that chemical modification of the non-conserved non-catalytic cysteine 222 (C222) inactivates specifically GlTIM. The inactivation correlates with the physicochemical properties of the modifying agent: addition of a non-polar, small chemical group at C222 reduces the enzyme activity by one half, whereas negatively charged, large chemical groups cause full inactivation.

Results: In this work we used mutagenesis to extend our understanding of the functional and structural effects triggered by modification of C222. To this end, six GlTIM C222 mutants with side chains having diverse physicochemical characteristics were characterized. We found that the polarity, charge and volume of the side chain in the mutant amino acid differentially alter the activity, the affinity, the stability and the structure of the enzyme. The data show that mutagenesis of C222 mimics the effects of chemical modification. The crystallographic structure of C222D GlTIM shows the disruptive effects of introducing a negative charge at position 222: the mutation perturbs loop 7, a region of the enzyme whose interactions with the catalytic loop 6 are essential for TIM stability, ligand binding and catalysis. The amino acid sequence of TIM in phylogenetic diverse groups indicates that C222 and its surrounding residues are poorly conserved, supporting the proposal that this region is a good target for specific drug design.

Conclusions: The results demonstrate that it is possible to inhibit species-specifically a ubiquitous, structurally highly conserved enzyme by modification of a non-conserved, non-catalytic residue through long-range perturbation of essential regions.

Figure 1. Ribbon representation of WT GlTIM.

The catalytic residues (K13, H96, E174) and the substrate analog 2-PG are shown as stick models; loop 6 is depicted in magenta and loop 7 in yellow. C222 is shown as dark blue sticks.

Figure 2. Chemical structure of the modifications of C222 introduced by site directed mutagenesis or chemical modification.

Values in parentheses indicate the volume added by the modification (volume of the lateral chain of the amino acid or chemical group introduced minus the volume of the lateral chain of the cysteine residue). For simplicity, only the structure of the lateral chain is shown; R indicates the alpha carbon.

Figure 3. Binding of 2-PG to WT GlTIM and C222 mutants.

(A) Fluorescence emission spectra of WT GlTIM in the absence and in the presence of increasing concentrations of 2-PG; for clarity, not all the spectra obtained in the experiment are shown. (B) Plot of maximal fluorescence intensity at 332 nm as a function of the 2-PG concentration for each mutant. For WT GlTIM, C222V, C222M and C222N, solid lines represent the fit of the data to equation y = (α/2Et)(Et+x+Kd) - √ (Et+x+Kd)²– (4xEt). For C222F, C222D and C222K the changes of fluorescence were minimal and could not be reasonably fitted.

Figure 4. Thermostability of WT GlTIM and C222 mutants.

The thermal unfolding of 0.1 mg/ml GlTIM in TED buffer was monitored by recording the change of the circular dichroism signal at 222 nm in a scanning from 25 to 70°C, at a rate of 1°C/min. The fraction of unfolded protein and the Tm values (inset) were calculated as previously described . Experiments were performed by duplicate; in all cases standard errors were less than 5%.

Figure 5. Asymmetric unit of C222D^C.

Each subunit in the asymmetric unit of the GlTIM C222D^C crystal is shown in different color.

Figure 6. Structural comparison of GlTIM WT^C and C222N^C.

(A) Structural alignment of GlTIM WT^C (green) and C222N^C (cyan); the overall RMSD for these structures is 0.38 Ų. (B) Per residue Cα RMSD of GlTIM WT^C versus C222N^C; for comparison, the same scale is used in Fig. 7C. (C) Active site comparison of GlTIM WT^C (green) and C222N^C (cyan).

Figure 7. Structural analysis of GlTIM C222D^C.

(A) Structural superposition of the 20 monomers in the crystallographic structure of GlTIM C222D^C; each chain is shown in different color. (B) Close-up of the superposed loop 6 and 7 regions in GlTIM C222D^C, which show the major conformational differences between the different chains; the orientation is the same as in panel A. (C) Per residue Cα RMSD values of the 20 monomers present in the crystallographic structure of GlTIM C222D^C; each chain is shown in a different color.

Figure 8. Structural divergence of loops 6 and 7 in the crystal structure of GlTIM C222D^C.

(A) Detailed view of the structural differences occurring in loops 6 and 7 in the chain F of C222D^C (red), in comparison with canonical closed (cyan) and open (yellow) states. For the closed state, WT GlTIM crystalized with 2-PG was chosen (4BI7). As GlTIM structure in the open conformation has not been obtained, the closely related structure of T. vaginalis TIM (3QST) was used as representative of the TIM-open state. In panels (B) and (C), the same comparison is shown for GlTIM C222D^C chains E and H, respectively. The substrate analog 2-PG and lateral chains of the YGGS motif are shown as stick models.