L-asparaginase (L-ASP) is a therapeutically significant enzyme, catalyzes the hydrolysis of L-asparagine (L-Asn) to produce aspartic acid and ammonia (
The enzyme asparaginase is present in most of the microbes like
Different tumour cells require L-Asn for the synthesis of proteins but they are deprived of this essential amino acid in the presence of L-ASP (13). Thus, cancer cells cannot grow in the absence or diminished expression of L-Asn. The inability to synthesize L-Asn by these tumour cells entirely depends on L-Asn from the circulating blood (14).
Therefore, depletion of L-Asn results in the death of leukaemic cells. A tumour inhibitory activity was observed only with asparaginases isolated from
Many research organizations have studied asparaginases production and purification in order to minimize impurities which may cause hypersensitivity or allergic reactions (18). The prolonged use of these enzymes may serve as antigens to the host body and elicit the production of antibodies in the tissues, which may interfere with the efficacy of the drug. Therefore, it is essential to search for novel L-ASP with medical properties. Several crystal structures of asparaginases from a wide variety of organisms have been determined earlier, (19). but still there is a paucity of structural information. Hence, the present study is focussed on modeling of L-ASP I enzyme from
The amino acid sequence of asparaginase enzyme was searched using the UniProtKB at the ExPASy (
Physical and chemical parameters of the entire L-ASP I enzyme was computed by ProtParam tool (
BLASTp search of L-ASP I enzyme with default parameter was performed against the Brook Haven Protein Data Bank (PDB) to find the suitable template for comparative or homology modelling (
The predicted model was assessed by PROCHECK, Structural analysis and verification server (SAVES) (
The model quality was evaluated for different parameters like Root Mean Square Deviation (RMSD) using TM-score (
The functional assessment of the predicted model was carried out by searching against the PROSITE database and the active site residues of the modeled L-ASP I enzyme was predicted by Galaxy server (25).
L-Asn was extracted from the Library of 3-D Molecular Structures in PDB format (
The L- Asp I enzyme has a molecular weight of 36.9 kDa with a theoretical isoelectric point (pI) 5.25. The Grand average of hydropathicity value for protein is -0.142 and the instability index (II) computed to be 43.21. The enzyme is predicted as a non-allergen.
The secondary structure prediction of the entire sequence of L-ASP I enzyme using SOPMA (with default parameters) showed that the protein having the composition of Helix = 35.61%, Coil = 32.34%, Strand = 20.77% and Beta-turn = 11.28% respectively. It is clear from this secondary structure prediction that this enzyme belongs to α/β class of proteins.
The three- dimensional (3D) structure of L-ASP I enzyme was determined by homology modeling. BLASTp analysis (
Blastp report of amino acid sequences producing significant alignments with L-ASP I of
PDB ID | Query coverage (%) | Max Identity (%) | E value | Gap (%) | Positive (%) |
---|---|---|---|---|---|
2OCD_A | 100 | 85 | 8e-168 | 0 | 91 |
2P2D_A | 99 | 73 | 4e-139 | 0 | 84 |
2HIM_A | 99 | 73 | 2e-138 | 0 | 84 |
3NTX_A | 99 | 70 | 2e-132 | 0 | 82 |
The modeled enzyme subunit contains two α/β domains connected by a linker region (
The linker sequence (magenta) connects the two monomers. The smaller C-terminal domain (residues 213-337) comprising of four right-handed α-helices with α-1 and α-2 present within the distal region of the domain interface. There are three β-sheets occupying the interior region namely β-1, β-2 and β-3 which are arranged parallel while the helices α-3 and α-4 are located on the inter-domain side of the enzyme. The linker sequence is made of coil protein which connects the two structural domains of the dimer from the amino acid residues 190-212. The topology of L-ASP I structure showing α -helices, β - sheets, turns and linker region is represented in a cartoon form (
The geometrical and structural consistency of the modeled enzyme was evaluated by different. The Φ and Ψ distributions of Ramachandran plot analysis using PROCHECK revealed that 93.4% and 5.6% amino acid residues are present in the most favored and additionally allowed regions, while 0.7% and 0.3% amino acid residues in the generously allowed and disallowed regions. This apparently indicates the geometric fitness of the modeled enzyme (
The L-ASP I model quality was assessed with the template PDB ID: 2OCD_A (29), based on high sequence identity (85%). Calculation of RMSD between the modeled enzyme and 2OCD_A was found to be 0.83 Å. It determines the best-aligning pair of chains between the template and predicted structure. ERRAT plot shows an overall quality factor of 83.90%, which is indicative of good resolution structure. The Z score was predicted to be -9.75 in L-ASP I and -8.08 in PDB ID: 2OCD_A (
The newly generated 3D model of L-ASP I after validation was deposited in Protein Model Database (PMDB). The L-ASP can be accessed from the archive using the Accession No.PM0080443 (
The dimeric form of L-ASP I contain several sites in both the structural domains predicted by Prosite. The residues of each active site from the N-terminal domain show proximity with the C-terminal domain.
The larger N-terminal domain contains two signature motifs for L-ASP I from the residues 8-16 (IAYTGGTIG) and 84-94 (GFVILHGTDTM) respectively. It also possesses two N-glycosylation sites positioned from the residues 159-162 (NRST) and 192-195 (NITV). The other motifs include Tyrosine sulfation site (residues 42-56, 47-61 and 75-89), Protein kinase C phosphorylation site (residues 2-4, 161-163), Casein kinase II phosphorylation site (residues 37-40, 50-53, 64-67 and 164167) and N-myristoylation site (residues 12-17, 16-21 and 158163).
The smaller C-terminal also contains N-glycosylation site (residues 270-273) NLTQ, Tyrosine sulfation site (residues 313-327), Protein kinase C phosphorylation site (residues 230232 and 262-264), Casein kinase II phosphorylation site (residues 297-300 and 319-322) and N-myristoylation site (residues 282-287, 287-292 and 294-299). The active site residues of the modeled enzyme were predicted to be 14T, 60S, 91T, 92D, 117S, 118Q and 170N by Galaxy server.
When molecular docking runs were performed, the enzyme L-ASP I displayed high binding affinity and significant preference to its substrate L-Asn from its larger N-terminal domain (
The surrounding residues of L-Asn and L-ASP I enzyme are shown within the sphere of the radius 8 Å. The enzyme residues T14, T91, S117, Q118, D92 and N170 are close to L-Asn (central residue) along the N and C direction contribute to shortrange interactions, while residue S60 show medium range interaction with the bound L-Asn (
The distance between the hydrogen bond lengths were measured between ligand substrate L-Asn and the modeled enzyme L-ASP I (
Molecular interactions between ligand substrate L-Asn and L-ASP I of
S.No. | Ligand | Residue and atom | Ligand atom | Bond length (A°) |
---|---|---|---|---|
1 | L-Asn | T14 & OG1 | OC | 2.4 |
2 | S60 & OG | O | 3.5 | |
3 | T91 & OG1 | OD1 | 1.7 | |
4 | D92 & OD1 | 1H | 2.4 | |
5 | S 117 & O | 2HD2 | 2.3 | |
6 | Q118 & O | 1HD2 | 2.3 | |
7 | N170 & ND2 | OD1 | 2.4 |
The potential uses of L-ASP as anticancer or anti-leukemic agent for the past few decades have completely revolutionized research in a new direction. These natural bioactive compounds isolated from a variety of microbes are considered to be safe and less toxic compared to synthetic drugs. In order to study the efficacy of these compounds, it is essential to understand its structure and function. The prediction of ligand binding sites is important in drug discovery to facilitate the optimization process. Most desirable therapeutic agents are usually required in micro quantities but expressing a high degree of purity and specificity.
In the present study, molecular docking results provided detailed structural insights between L-ASP I and ligand substrate L-Asn interactions. The enzyme L-ASP I was predicted to be dimeric and have a molecular weight of 36.9 kDa. The molecular weight of enzymes plays an important role in the mechanism of enzyme action. The theoretical isoelectric point demonstrates its acidic nature (pI <7) and probable role in buffer systems of blood. The GRAVY value is ≤0 indicates hydrophilic nature having an affinity to the surrounding water molecules (30). The enzyme predicted to be as non-allergen and close to Apolipoprotein B, a major protein constituent of chylomicrons. Apolipoprotein B-100 functions as a recognition signal for cellular binding (31). However, several adverse effects of L-ASP applications have been reported earlier due to immunogenic reactions (32). In order to alleviate side effects in the treatment process drugs such as Oncaspar and Pegasparase are used in patients whose cancer has not been treated or cannot be treated with asparaginase. These are pegylated asparaginase (poly ethylene glycol linked to asparaginase) which enhances stability, safety and efficacy of the drugs (33).
Interestingly, the predicted 3D structure of L-ASP I has the characteristic features of L-ASP family, designated as Type I cytosolic enzyme. The Φ and Ψ distributions of Ramachandran plot analysis revealed that 93.4% and 5.6% amino acid residues are present in the most favoured and additionally allowed regions, while 0.7% and 0.3% amino acid residues in the generously allowed and disallowed regions. This is in conformity with V. cholerae L-ASP I which exhibited 89.2% and 10.2% amino acid residues in the most favoured and additionally allowed regions, while 0.4% and 0.3% amino acid residues found in the generously allowed and disallowed regions (PDB: 2OCD). Similarly, the Escherichia coli L-ASP (EcA I) showed the presence of 92.2% residues in the most favourable region, 7.4% in the allowed region and none in the disallowed region (34). In addition, PROCHECK analysis revealed the main chain and side chain parameters were found to be within the acceptable range indicating the reliability of L-ASP I structure.
The structural convergence includes root mean square deviation (RMSD), low RMSD value between L-ASP I and template PDB ID: 2OCD showed the stability of the predicted structure, while Errat value indicated the good resolution of the structure. ProSA analysis predicted negative interaction energies (-9.75) for most of the residues in L-ASP I, confirming a stable structure. The point of the structure was within a range as demonstrated by X-ray and NMR studies. The model quality and location of the Z-score is in conformity with the template PDB ID: 2OCD.
The modeled enzyme belongs to α/β class of proteins characterized by alternating α-helices/β-sheets running along the backbone of the N-terminal domain. The larger N-terminal domain has an unusual left-handed β-α-β crossover normally found in a bacterial protein, flavodoxin. This type of arrangement serves as a cavity or cleft for the active site, involved in ligand binding (35). The enzyme is dimeric containing two structural domains with several active sites. The two monomers are held together by a linker amino acid sequence comprising of coils. The coil proteins are long insoluble fibres involved in the extracellular matrix, leading to enzyme-substrate interactions. The larger N-terminal structural domain forms the outer surface of the enzyme while the smaller C-terminal domain is located at the distal region of the domain interface. The present study is in conformity with the crystal structure of Escherichia coli asparaginase II, showing similar structural details (36).
The modeled enzyme showed two signature motifs for L-ASP I activity. It also represented three N-glycosylation sites in the larger and smaller structural domains, where carbohydrates can be linked to proteins through asparagines (N-linked). An asparagine residue can accept an oligosaccharide only if the residue is part of an L-Asn-X-Ser or L-Asn-X-Thr sequence in which ‘X’ can be any residue except proline. Glycosylation enhances the stability of the protein in the blood. Unglycosylated protein has only about 10% of the bioactivity of the glycosylated form because the protein is rapidly removed from the blood by the kidneys. Thus, potential glycosylation sites can be detected with amino acid sequences. Many proteins, especially those that are present on the surfaces of cells are secreted; acquire carbohydrate units on specific asparagine residues. The addition of sugars makes the proteins more hydrophilic and able to participate in interactions with other proteins (37).
The structure-activity relationship is well demonstrated in the present investigation between the enzyme L-ASP I and ligand substrate L-Asn interactions. A number of distinguishing characteristics are associated with a binding pocket/cleft in L-ASP I. L-Asn access to the binding pocket is facilitated through the N-terminal region of the enzyme. Access is gained by the L-Asn entering through α1-β1, α2-β2 and β5-β6 within the larger structural domain of the enzyme. This is in accordance with the crystal structure of E. coli L-asparaginase II showing the presence of active sites in the larger structural domain of the tetramer (38).
In the present investigation, L-ASP I of
The side chains of enzyme residues T14, S60, T91, S117, Q118, D92 and N170 showed a close affinity to the bound L-Asn substrate using ArgusLab v4.0. The residues T14, T91, S117 and S60 are considered to be potential nucleophiles, whereas S60 and S117 serve as a base while D92 is an acid and Q118 is neutral. It is observed from the docked complex that amino acids T14, S117 and D92 come together in a sequential configuration due to the folding of protein residues of the enzyme. These three residues form a catalytic triad (Thr-Ser-Asp) and facilitate binding with the substrate to perform covalent catalysis with the help of T14, which targets the carbonyl carbon of L-Asn and pressurizes the carbonyl oxygen to accept an electron. The T14 is polarized and oriented by S117, which may contribute to the action of T14 as a nucleophile and gets stabilized by D92 within the active site. Though T91 found to be in close contact with L-Asn it is worth mentioning here that OG1of T91 interacts with 1H of L-Asn substrate, while OG1 of T14 interacts with OC of L-Asn which is essential for nucleophilic activity. This is in conformity with the catalytic triads commonly found in hydrolase and transferase enzymes (40). Earlier reports of L-ASP crystal structures demonstrated the presence of two catalytic triads in each active site. The triad I is composed of a nucleophile (Thr), a base (Tyr) and an acidic moiety (Glu) which functions on the flexibility of the active site loop (41). A similar kind of mechanism may take place when T14 gets activated along with other two residues S117 and D92 leading to the conversion of substrate L-Asn to aspartic acid and ammonia.
Based on the structural evidence, it is therefore derived that T14 is the most potential candidate which acts as a primary nucleophile. The present investigation is in agreement with the modelled crystal structure of the inactive T89V mutant, refined with the aspartate ligand which was covalently bound to T12 by its side chain carboxy group indicating that T12 may function as a primary nucleophile, while this activity may not occur due to the lack of a nucleophilic group on residue T89 (42). In the crystal structure of Escherichia coli L-asparaginase, active sites from N to C-terminal domains contain diverse subunits. The presence of two threonine residues of the enzyme T12 and T89 may act as primary nucleophiles. However, T89 exhibited as an essential nucleophile, which is fundamental for the enzymatic action (43).
The biological activity of any drug is dependent on the binding between protein and ligand (44). The present investigation has clearly demonstrated a close affinity between the enzyme and its ligand substrate. Hence, L-ASP I of
The predicted structure of L-ASP I of