Identification of residues involved in substrate specificity and cytotoxicity of two closely related cutinases from Mycobacterium tuberculosis

Abstract

The enzymes belonging to the cutinase family are serine enzymes active on a large panel of substrates such as cutin, triacylglycerols, and phospholipids. In the M. tuberculosis H37Rv genome, seven genes coding for cutinase-like proteins have been identified with strong immunogenic properties suggesting a potential role as vaccine candidates. Two of these enzymes which are secreted and highly homologous, possess distinct substrates specificities. Cfp21 is a lipase and Cut4 is a phospholipase A2, which has cytotoxic effects on macrophages. Structural overlay of their three-dimensional models allowed us to identify three areas involved in the substrate binding process and to shed light on this substrate specificity. By site-directed mutagenesis, residues present in these Cfp21 areas were replaced by residues occurring in Cut4 at the same location. Three mutants acquired phospholipase A1 and A2 activities and the lipase activities of two mutants were 3 and 15 fold greater than the Cfp21 wild type enzyme. In addition, contrary to mutants with enhanced lipase activity, mutants that acquired phospholipase B activities induced macrophage lysis as efficiently as Cut4 which emphasizes the relationship between apparent phospholipase A2 activity and cytotoxicity. Modification of areas involved in substrate specificity, generate recombinant enzymes with higher activity, which may be more immunogenic than the wild type enzymes and could therefore constitute promising candidates for antituberculous vaccine production.

Figure 1. Mapping of the main differences between the Cfp21 and Cut4 structural models in the vicinity of the active site.

(A) Cfp21 model. (B) Cut4 model. (C) Amino acid sequences in the 3 mutated areas. Only the amino acids in the three mutagenized areas (A1, A2, and A3) are coloured. Hydrophobic amino acids are in yellow. Negatively charged amino acids are in red and positively charged amino acids are in blue. Residues Asp⁵⁴ in area 1 (A1), Asp⁸³, Tyr⁸⁵, Arg⁸⁶ in area 2 (A2), and Asn¹⁸⁹, ^Ile190, Met¹⁹¹ in area 3 (A3) present on Cfp21 were replaced using site-directed mutagenesis methods by residues found to exist in Cut4 at the same location in the structural models. The amino acids are numbered starting with the initial methionine in the same way as for a protein precursor including the signal sequence.

Figure 2. SDS-PAGE and CD spectra analysis of purified proteins.

A) Protein purity assessed on SDS-PAGE. MW: molecular weight standard from Euromedex; 1: Cfp21 WT; 2: Cfp21-A1; 3: Cfp21-A2; 4: Cfp21-A3; 5: Cfp21-A1A2; 6: Cfp21-A2A3; 7: Cfp21-A1A3; 8: Cfp21-TM. 1 µg of each protein was loaded onto a 15% polyacrylamide gel and stained with Coomassie Blue. B) Far UV spectra of the Cfp21 mutant proteins.

Figure 3. Lipase Specific Activities (SAs) of the Cfp21 mutants on fluorescent triglycerides.

SAs were calculated from the velocity slope obtained for 10 minutes using 10 µg of enzymes. Results are the means obtained in at least 3 independent experiments. Only the mutants with lipase activity are presented. Cfp21 and Cut4 were used as positive and negative standards, respectively .

Figure 4. Phospholipase A activities of the Cfp21 mutants.

A) phospholipase A activity of the Cfp21 mutants: Hydrolysis of ¹⁴C-DPPC. FA: fatty acid; DPPC: dipalmitoyl-phosphatidylcholine; LysoPC: palmitoyl-lyso-phosphatidylcholine. Pure LysoPC, DPPC and Palmitic acid (FA) were used as standards (lanes 1, 2, and 3). Pancreatic porcine phospholipase A₂ (ppPLA₂) was used as a positive standard (lane 4). Products resulting from a 24-h enzymatic reaction with wild-type Cfp21 and its mutants were loaded into lanes 5 to 11. Rates of hydrolysis were calculated as described in the experimental section. B) Specific activities of phospholipase A₁ and A₂ on fluorescent phospholipids. The PLA₁ from Thermomyces lanuginosus was used as a positive standard to determine the PLA₁ activities, and ppPLA₂ was used as a positive standard to determine the PLA₂ activities. The specific fluorescent substrates used to measure the PLA₁ and PLA₂ activities were BODIPY® dye-labeled PED-A1 and the red/green BODIPY® PC-A2, respectively, as described in the Material and Methods section. Specific activities were calculated from the velocity slope obtained for 10 minutes using 10 µg of enzymes. Results are means obtained in at least 3 independent experiments.

Figure 5. Cytotoxic effects of Cfp21 mutants on mouse macrophage cells.

Cytotoxic effects causing macrophage lysis were monitored by measuring the release of LDH into the culture media. LDH activities were measured after 16 h and 24 h of incubation with 150 µg of M. tuberculosis enzymes or 10 µg of ppPLA₂. The rate (%) of macrophage cell lysis was calculated as described in the experimental section. Similar experiments were performed using THL inhibited enzymes to determine the involvement of active serines in the cytotoxic process.