Human African Trypanosomiasis Is Caused Biology Essay

Published: Last Edited:

This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.

Currently the drugs available in the market for treating African sleeping sickness are very few. The drugs like Pentamidine, Suramin and Melarsoprol are very expensive, ineffective and at times use of these drugs can cause harmful side effects.[3]

Therefore there is a need for a new, cost-effective, non-toxic and efficacious drug to treat this fatal parasitic disease.

Protein kinases -Potential drug targets

Parasite protein kinases are considered as the potential drug targets for the disease African trypanosomiasis. The protein kinases have significant role in various intracellular signalling pathways involved in proliferation, differentiation, stress response and motility. Therefore disrupting or inhibiting the activity of the protein kinase will be detrimental for the parasite Trypanosoma brucei.[2]

Protein kinase (Gene identifier: Tb927.3.690) and serine/threonine-protein kinase (Gene identifier: Tb927.3.2440) are the two potential drug targets and these belong to the Trypanosoma brucei brucei strain. The suitability of these two proteins as novel therapeutic targets was analysed and it is reported below.

Evaluation of the target protein (Tb927.3.690) - As a potential drug target.

Selection of template

The sequence of the target protein Tb927.3.690 with unknown 3D protein structure was obtained from NCBI. For homology modelling, the templates 4ERW and 1GZ8 were selected among the top hits because they both had the highest sequence similarity of 40% with the target protein sequence. Both 4ERW and 1GZ8 are human proteins. Also other factors like resolution and R-factor of the crystallographic structure were taken into account for the selection.

The resolution and R- factor of the template 4ERW was 2 Å and 0.196 respectively.

The resolution and R-factor of the template 1GZ8 was 1.30 Å and 0.153 respectively.

Both the templates were used for homology modelling as use of two or more templates increases the model accuracy. Additionally, to assess the quality of the templates Ramachandran plot was considered.

The Ramachandran plot of 4ERW showed that 99.3% of all the residues were in the allowed regions and there were 2 outliers: 154 VAL, 162 GLU.

The Ramachandran plot of 1GZ8 showed that all (100%) the residues were in the allowed regions.

Next the target protein sequence was successfully aligned with both the template sequences using the T-Coffee program and the 3D model of the target protein was obtained using MODELLER.

Quality Assessment of the model

For a model to be accurate, its atomic co-ordinates should be within 0.5 Å rmsd of a control experimental structure.[4] Using pymol, when the template structures 4ERW and 1GZ8 were structurally aligned to the modelled target protein their root mean square deviation values (RMSD) were found to be 0.305 Å and 0.323 Å respectively. This indicates that the produced model is of good quality.

Ligand Docking and Analysis of the active site

Ligand Docking - Using the virtual screening program LIDAEUS a small set of potential ligands that would most probably bind to the target protein (protein kinase) were identified from a library of known chemical molecules.

Based on the visual analysis, scoring function, ligand efficiency and interaction of the ligand with the active site of the target protein, the following two ligands were selected from the top 50 hits.

Ligand 1: 3-phenyl-4,5-dihydroisoxazol-5-one. The ligand efficiency was found to be -1.74845 and the ligand interacted with the active site of the target protein via 2 hydrogen bonds and electrostatic interaction.

Ligand 3: 7-methylbenzo[d]isoxazol-3-ol. The ligand efficiency was found to be -1.77698 and the ligand interacted with the active site of the target protein via 2 hydrogen bonds and many (3) electrostatic interactions.

The electrostatic interactions are possibly due to the presence of two ASP's, one GLU and one LYS in the binding pocket of the protein kinase (target) and these residues provide carboxyl groups and amine terminals for interaction with the ligand.

Analysis of the binding pocket of the parasite protein and the human protein (4ERW).

The amino acid composition in the active site's of the target and the template were compared and GLY, VAL,ALA and ASN residues were found to be in the same positions in both the pockets and few differences like PHE instead of TYR and LEU instead of MET were seen in the active site of protein from human (Fig 1).

From the comparative analysis it's understood that the binding pockets of the human protein template and parasite protein kinase have many similarities. Therefore it is possible for the ligand to interact with the protein (4ERW) in human.

In conclusion, the chosen templates led to the generation of a good 3D model of the target protein kinase. In terms of interaction with the parasite kinase, the ligands 3-phenyl-4,5-dihydroisoxazol-5-one and 7-methylbenzo[d]isoxazol-3-ol showed polar contacts and electrostatic interaction. But it is not advisable to use this parasite protein kinase as a drug target because its active site is very similar to the active site of a human protein. There are high chances of the designed ligand for the parasite kinase interacting with the human protein.

Evaluation of the second target protein (Tb927.3.2440) - As a potential drug target.

Selection of Template

The sequence of the target protein Tb927.3.2440 with unknown 3D protein structure was acquired from NCBI. For homology modelling, the templates 1GZN and 1GZK were selected among the top hits because they both had the highest sequence similarity of 39% with the target protein sequence. Both 1GZN and 1GZK are human proteins. Also the other factors like resolution and R-factor of the crystallographic structure were taken into account for the selection.

The resolution and R- factor of the template 1GZN was 2.5 Å and 0.268 respectively.

The resolution and R-factor of the template 1GZK was 2.30 Å and 0.248 respectively.

Both the templates were used for homology modelling of the serine/threonine-protein kinase (target protein) to enhance the model accuracy. Furthermore, to assess the quality of the templates Ramachandran plot analysis was considered.

The Ramachandran plot of 1GZN showed that 98.9% of all the residues were in the allowed regions and there were 3 outliers: 354 ASP, 355 HIS, 411 SER.

The Ramachandran plot of 1GZK showed that 98.9% of all the residues were in the allowed regions and there were 3 outliers: 294 PHE, 354 ASP, 355 HIS.

Next the target protein sequence was successfully aligned with both the template sequences using the T-Coffee program and the 3D model of the target protein was obtained using MODELLER.

Quality Assessment of the model

For a model to be accurate, its atomic co-ordinates should be within 0.5 Å rmsd of a control experimental structure.[4] To check for the accuracy and quality, the template structures 1GZN and 1GZK were structurally aligned to the modelled target protein and the root mean square deviation values (RMSD) were found to be 0.226 Å and 0.235 Å respectively. This shows that the generated model is of good quality.

Ligand Docking and Analysis of the active site

Using the virtual screening program LIDAEUS a small set of potential ligands that would most probably bind to the target protein (serine/threonine-protein kinase) were identified from a database of known chemical molecules.

Based on the visual analysis, scoring function, ligand efficiency and interaction of the ligand with the active site of the target protein kinase, the following two ligands were selected from the top 50 hits.

Ligand 1: 1,3-benzodioxol-5-ol. The ligand efficiency was found to be -2.247 and the ligand interacted with the active site of the target protein via hydrogen bonds and electrostatic interaction.

Ligand 17: 3-methylisoxazol-5-amine. The ligand efficiency was found to be -1.704 and the ligand interacted with the active site of the target protein via 2 hydrogen bonds, electrostatic interactions and a PI bond.

The electrostatic interaction is due to the presence of ASP and GLU residues in the active site of the serine/threonine protein kinase ( target) and these residues provide carboxyl groups for the electrostatic interaction.

PI-PI ring stacking interaction was seen between the ligand 17 and the target protein. This non-bonded interaction can be observed between two aromatic rings and it plays an important role in ligand - protein interaction.

Next the binding pockets of the target protein and the human template protein were examined. The amino acid composition in the active site's of the target and the template were compared. The amino acid residues LEU, PHE, TYR (amino acids with hydrophobic side chains) and GLU (amino acid with hydrophilic side chain) were found to be in the same positions in both the pockets and few differences like the presence of MET instead of LEU and GLU instead of ASP were seen in the active site of human (Fig 2).

From the comparative analysis it's understood that the binding pockets of the human protein template and parasite protein kinase have many similarities. From the analysis it was also observed that the TYR residue in the active site of human protein formed hydrogen bond with ligand1 and also PHE residue in the active site of the human protein formed hydrogen bond with ligand 17.

In conclusion, the chosen templates 1GZN and 1GZK led to the generation of a good 3D model of the target serine/threonine-protein kinase. In terms of interaction with the parasite kinase, the ligands 1,3-benzodioxol-5-ol and 3-methylisoxazol-5-amine showed polar contacts, electrostatic interaction and PI-PI interaction. But this protein kinase target cannot be used as a drug target because its active site is very similar to the active site of a human protein and as observed in the analysis, there are high chances that the ligand designed for this parasite kinase will interact with the human protein.