Computational insights into Candida albicans Malate synthase: Impact of cofactor and substrates on enzyme conformation and tunnel formation

Computational insights into Candida albicans Malate synthase: Impact of cofactor and substrates on enzyme conformation and tunnel formation | Authorea try { document.documentElement.classList.add('js'); } catch (e) { } var _gaq = _gaq || []; _gaq.push(['_setAccount', 'G-8VDV14Y67G']); _gaq.push(['_trackPageview']); (function() { var ga = document.createElement('script'); ga.type = 'text/javascript'; ga.async = true; ga.src = ('https:' == document.location.protocol ? 'https://ssl' : 'http://www') + '.google-analytics.com/ga.js'; var s = document.getElementsByTagName('script')[0]; s.parentNode.insertBefore(ga, s); })(); Skip to main content Preprints Collections Wiley Open Research IET Open Research Ecological Society of Japan All Collections About About Authorea FAQs Contact Us Quick Search anywhere Search for preprint articles, keywords, etc. Search Search ADVANCED SEARCH SCROLL This is a preprint and has not been peer reviewed. Data may be preliminary. 28 June 2025 V1 Latest version Share on Computational insights into Candida albicans Malate synthase: Impact of cofactor and substrates on enzyme conformation and tunnel formation Authors : Lukkani Laxman Kumar 0000-0003-1168-2708 and Ayaluru Murali 0000-0001-6406-6840 [email protected] Authors Info & Affiliations https://doi.org/10.22541/au.175115182.24846590/v1 316 views 202 downloads Contents Abstract Information & Authors Metrics & Citations View Options References Figures Tables Media Share Abstract Candida albicans , an opportunistic pathogen, uptakes host-derived carbon sources for its survival and causes candidiasis infection. However, host defences such as macrophages restrict the C. albicans’ sugar uptake via phagocytosis, triggering the glyoxylate pathway for adaptation in nutrient-limited conditions. Within macrophages, the C. albicans malate synthase (CaMLS1) exploits the available acetyl-CoA (ACOA), which condenses in presence of Mg 2+ with glyoxylate (GOXL) to form malate, supporting its survival. The absence of structural studies highlights the need for in silico interventions to better understand CaMLS1 functional architecture thereby aiding in design of the therapeutic drug agents. In this study, computational approaches such as MD simulations, correlations studies and binding affinity analysis were employed to delineate the structural dynamics and to investigate the interactions of cofactors and substrates with CaMLS1. From the results, template-based 3D modeling revealed that CaMLS1 has three domains: N-terminal, TIM-barrel and C-terminal domain, of which TIM-barrel found to be dynamic through MD studies. However, substrates (GOXL and ACOA) binding to CaMLS1 markedly reduced this dynamism, indicating the stability of the complex. The binding free energy calculations of the complex showed that ACOA binds with notably higher affinity (-921 kJ/mol) than GOXL (-6 kJ/mol), consistent with previous observations reported in literature. Furthermore, presence of cofactors and substrates modulated the tunnel availability at the active site, thereby affecting substrate’s access and product release through changes in the enzyme’s conformational dynamics. The findings from the structure and dynamics studies of CaMLS1, as well as the key residues involved in substrate binding offer scope for design of molecules targeting this enzyme for antifungal therapy. Computational insights into Candida albicans Malate synthase: Impact of cofactor and substrates on enzyme conformation and tunnel formation Lukkani Laxman Kumar , Ayaluru Murali* Department of Bioinformatics, Pondicherry University, Pondicherry – 605014, India *Corresponding author (E-mail: [email protected] ) ORCID: Lukkani Laxman Kumar (0000-0003-1168-2708); Ayaluru Murali (0000-0001-6406-6840) Abstract Candida albicans , an opportunistic pathogen, uptakes host-derived carbon sources for its survival and causes candidiasis infection. However, host defences such as macrophages restrict the C. albicans’ sugar uptake via phagocytosis, triggering the glyoxylate pathway for adaptation in nutrient-limited conditions. Within macrophages, the C. albicans malate synthase (CaMLS1) exploits the available acetyl-CoA (ACOA), which condenses in presence of Mg 2+ with glyoxylate (GOXL) to form malate, supporting its survival. The absence of structural studies highlights the need for in silico interventions to better understand CaMLS1 functional architecture thereby aiding in design of the therapeutic drug agents. In this study, computational approaches such as MD simulations, correlations studies and binding affinity analysis were employed to delineate the structural dynamics and to investigate the interactions of cofactors and substrates with CaMLS1. From the results, template-based 3D modeling revealed that CaMLS1 has three domains: N-terminal, TIM-barrel and C-terminal domain, of which TIM-barrel found to be dynamic through MD studies. However, substrates (GOXL and ACOA) binding to CaMLS1 markedly reduced this dynamism, indicating the stability of the complex. The binding free energy calculations of the complex showed that ACOA binds with notably higher affinity (-921 kJ/mol) than GOXL (-6 kJ/mol), consistent with previous observations reported in literature. Furthermore, presence of cofactors and substrates modulated the tunnel availability at the active site, thereby affecting substrate’s access and product release through changes in the enzyme’s conformational dynamics. The findings from the structure and dynamics studies of CaMLS1, as well as the key residues involved in substrate binding offer scope for design of molecules targeting this enzyme for antifungal therapy. Keywords: Candida albicans, Candidiasis, Glyoxylate pathway, Malate synthase, Cofactor and Substrates, Homology modeling, Molecular Dynamics Simulations Abbreviations ACOA – Acetyl-CoA; CaMLS1 – Candida albicans Malate synthase; GOXL – Glyoxylate; GROMACS – GROningen MAchine for Chemical Simulations; MMA – CaMLS1-Mg 2+ -ACOA complex, MMAG – CaMLS1-Mg 2+ -ACOA-GOXL complex; MMG – CaMLS1-Mg 2+ -GOXL complex. Highlights • C. albicans’ Malate synthase (CaMLS1) and its substrate complexes were evaluated using molecular dynamics (MD) trajectories • Mg 2+ , required for catalytic activity, was intact throughout the simulations in substrate-bound complexes • Stabilization of the protein backbone as well as TIM-barrel was enhanced in the complex with two substrates compared to the other systems • Presence/absence of the tunnel was altered in response to cofactor and substrates, suggesting their role in regulating the tunnel. • Per-residue energy decomposition analysis revealed that positively charged residues at the substrate binding site contributed significantly with highly negative energies, indicating favorable interactions and identifying the site as promising for ligand design. Introduction Fungal infections in the 21 st century are on a spike and are recognized as serious concern for human health [1]. Over 6.55 million people are affected each year by immediately life-threatening fungal diseases, leading to over 3.75 million deaths. Of these deaths, about 2.55 million are exclusively attributable to the fungal infections [2]. The known fungal infections are quite diverse, and among them, Candida sp . is the most frequently isolated species [3]. Among the Candida sp ., the Candida albicans (a human fungal pathogen) is a public health concern causing life-threatening infections “candidiasis” in India and worldwide [4, 5]. About 1.56 million people have a Candida bloodstream infection or invasive candidiasis each year, with 0.99 million deaths (63.6%) [2]. Candidiasis, a broad spectrum of diseases, is widely prevalent in many countries, especially in immunocompromised individuals [4]. The opportunistic pathogen, C. albicans , is present in diverse host niches such as skin, oral-cavity, mucosal, gastrointestinal tract (GI), reproductive organs, and in the bloodstream [6]. During the COVID-19 pandemic, an association between Candidiasis and COVID-19 cases was reported [7]. Oral candidiasis is often one of the earliest clinical signs of human immunodeficiency virus (HIV) infection. In fact, the initial diagnosis of AIDS in many patients was made based on the presence of oral candidiasis symptoms [8]. In another study, among the 276 HIV/AIDS patients, the overall prevalence of oral candidiasis is 41% [9]. Mohammadi et al have evaluated the significant relationship between diabetes and Candidiasis. Among the isolates, C. albicans is the prominent species found in the oral cavity of patients with Diabetes Mellitus [10]. Being heterotrophic, C. albicans adapt to grow on a wide range of host niches (such as skin, mucosal surfaces, and blood) for their survival and virulence. The transcriptional profiling of C. albicans have shown no significant regulation in blood plasma (in the glycolytic pathway), while in other host environments, a significant upregulation of genes have been observed [11]. C. albicans, a facultative aerobe , can metabolize carbon sources through a variety of pathways which are important for their growth, virulence, and establishment of infection in the host. As C. albicans moves into a host, it adapts its energy metabolism to find alternate ways of providing the energy and substrates needed for cellular processes [12]. In the host blood stream, the availability of glucose is high, which is a preferred carbon source for C. albicans . Among C. albicans and other pathogens, it is reported that carbon source and pathogenicity strongly contribute to virulence, as glycolytic genes are upregulated during infections [12, 13]. In immunocompromised individuals, though the immune system is weak it still combats the pathogen through different innate immune mechanisms. One such is macrophage mediated phagocytosis [14]. When C. albicans is engulfed by a macrophage, it encounters a carbon-limiting environment within macrophage phagosome by creating hostile conditions [12, 15] . In order to survive and adapt within nutrient limiting environment, C. albicans activates the glyoxylate pathway as a key metabolic adaption for utilising the alternative carbon sources inside macrophage [16] . Glyoxylate pathway activation As a part of adaptation, C. albicans undergo morphological changes from yeast to hyphae. These hyphae can differentiate into filamentous hyphae, which elongate and pierce the macrophage membrane, allowing C. albicans to escape [17]. Lorenz and Fink were the first to show that glyoxylate genes are essential for the virulence of C. albicans that can survive within macrophages [18]. The glyoxylate cycle, an anaplerotic pathway, bypasses decarboxylation steps of the tricarboxylic acid (TCA) cycle for the conversion of two carbon compounds when there is a scarce of simple sugars [16]. The glyoxylate pathway is widely distributed across the different kingdoms of life, which include bacteria [19], fungi [20], plants [21], and protozoans [22]. Two key enzymes of the glyoxylate cycle are Isocitrate lyase (ICL) and Malate synthase (MLS1). These enzymes play an important role by facilitating the conversion of fatty acids to carbohydrates which are essential for their survival and adaptation in diverse environments across various species [23]. Interestingly, ICL and MLS1 are absent in humans making them more potential for anti-fungal drug development [24]. Role of MLS1 in C. albicans C. albicans Malate synthase (CaMLS1) catalyzes the condensation of glyoxylate (GOXL) and acetyl-CoA (ACOA) to form Malate and Co-enzyme A (CoA). The catalytic activity of MLS1 takes place in the presence of Mg 2+ ion [25]. Metal ions, bound to amino acid side chains in protein, play a major role in protein folding, regulation, and catalytic process [26]. Metals such as iron (Fe) play an important role in catalytic activity of some enzymes viz., Catalase and Cytochrome oxidase by donating or accepting electrons [27]. Beyond their roles in basic metabolism, metal cofactors are also important in pathogenesis and infection processes [28]. Certain pathogens exploit metal-dependent enzymes to evade host immune responses, enhance their virulence, and survive under stressful conditions within host. Okada et. al has demonstrated the importance of the Mg 2+ for enzymatic activity of C. tropicalis Malate synthase [29]. This illustrates how metal cofactors can influence the metabolic processes that may contribute to the pathogen survival and adaptability. The MLS1 has been shown to play an important role in adherence and host-fungal interactions in Paracoccidioides sp. and has shown inhibition activity with alkaloid compounds [30]. Through in vivo studies, a reduction of virulence of C. albicans was reported in mouse models lacking the mls1 gene [18]. For understanding of any biological process or for developing a drug molecule, the 3D structure plays a prominent role. However, no experimentally resolved structure (such as X-ray crystallography, NMR, or cryo-EM) is currently available for CaMLS1. This study made an attempt to model the CaMLS1 and thereby evaluating the role of metal cofactors and substrates. Unravelling the binding of metal ions and substrates to the active sites of CaMLS1 would provide an understanding of the activation mechanism which could help in designing an effective drug. Materials and Methods Protein sequence retrieval and Phylogeny construction The amino acid sequence of C. albicans malate synthase (CaMLS1) (Q5APD2) was retrieved from UniProt database (https://www.uniprot.org/uniprotkb/). The physico-chemical properties – such as molecular weight (MW), isoelectric point (PI), instability index (II), aliphatic index, and grand average of hydropathicity (GRAVY) were analysed using Protparam [31]. The functional analysis of CaMLS1 was carried out using InterPro (www.expacy.org) database [32]. Using BLASTp [33], the CaMLS1 query sequence was subjected to the NR database and collected the protein sequences of Candida sp. with malate synthase. Further, the multiple sequence alignment was performed for Malate synthase protein sequences of C. albicans and other Candida sp. using the CLUSTALW [34] and visualized using sequence viewer tool in CLC workbench [35]. Based on the multiple sequence alignment, the phylogenetic tree was generated using IQ-TREE v1.6.11 [36], which selects the best substitution model. The phylogenetic tree, and branch length were visualised via I-TOL-webserver v 6 [37]. Secondary structure prediction and 3D protein modeling of CaMLS1 To predict the CaMLS1 secondary structure composition, the Garnier-Osguthorpe-Rabson IV (GOR IV) [38] and PSIPRED [39] tools were utilized. For template identification, the primary CaMLS1 protein sequence was submitted to BLASTp and searched against the Protein Data Bank (PDB) database. Based on the E-value, coverage, and percent identity, the best template was employed for modeling. The CaMLS1 structure was modeled using the Robetta server, which employs comparative modeling [40]. Although state-of-art tools like AlphaFold are available, Robetta was preferred for its effective use of homologous templates in this case. The loop regions in the protein models were optimised using the ModLoop server (https://modbase.compbio.ucsf.edu/modloop/) [41]. The stereochemical properties of the generated models were analysed using ERRAT [42] and PROCHECK [43], hosted at the SAVES server (https://saves.mbi.ucla.edu/). Protein secondary architecture was analysed using PDBSum [44]. Docking and interaction studies of substrates To check the interactions and docking scores of the substrates with the active sites of CaMLS1, molecular docking studies were performed using two substrates - Glyoxylate (GOXL) and Acetyl-CoA (ACOA). The 2D chemical structures - GOXL (ChEBI ID: 36655) and ACOA (ChEBI ID: 57288) were retrieved from the ChEBI database (https://www.ebi.ac.uk/chebi/init.do) [45]. The active residues of CaMLS1, responsible for binding two substrates were identified through available literature on malate synthase from other organism [46], as well as by utilizing the CASTp server [47]. Protein structure refinement, preprocessing, addition of hydrogens and energy minimization were carried out using protein preparation wizard module in Schrödinger suite [48]. Before docking, both the molecules were subjected to energy minimization using LigPrep tool [49] with Optimized Potentials for Liquid Simulations (OPLS) force field [50]. According to the potential binding pocket of the processed protein structure, the receptor grids were placed to dock the appropriate substrates using receptor grid generation module of Schrödinger suite. The protein and the substrates were docked through Glide v9.3 XP (extra precision) of Schrödinger suite [51]. The 2D interactions of protein and substrates were generated with ligand interaction diagram module in Schrodinger Maestro platform v13 and LigPlot + v2.2.5 [52]. The obtained poses from the docking were analysed and the best pose was selected further based on the XP Glide score and interactions formed. Molecular dynamics (MD) simulations MD simulations of C. albicans malate synthase and its complexes were performed in triplicates using GROMACS package [53] version 2019 with GROMOS54A7 forcefield [54]. The ATB server was used to calculate the ligands’ parameters and topologies [55]. The initial models were solvated using an extended simple point charge (SPC/E) water model embedded in a cubic box at a distance of 10 Å from the box edge. According to the system’s net charge, the Na + / Cl - ions were added to neutralize the system by replacing water molecules with these ions. Prior to MD studies, an energy minimization of 50K steps was performed using the steepest descent method for the removal of bad contacts and steric clashes in the system. The system equilibration was carried out in two steps, namely NVT (constant Number of particles, Volume, and Temperature) and NPT (constant Number of particles, Pressure, and Temperature) ensembles. During the NVT and NPT phases, the system was maintained at 300 K and 1 atm for 100 ps using V-rescale thermostat [56] for temperature coupling and Parrinello-Rahman barostat [57] as pressure coupling. The Particle-mesh Ewald method [58] was employed for calculating electrostatic interactions with a coulomb cutoff of 10 Å. Finally, an MD production run was performed for all systems for a time scale of 100 ns. Further, the MD trajectory files were analysed using GROMACS tools such as gmx_rms, gmx_rmsf, and gmx_trjconv (-dump) to obtain the root means square deviation (RMSD), root means square fluctuation (RMSF), and for extracting 3D structures at different time frames. Hydrogen bond plot and residue occupancy with the substrates were calculated using gmx hbond and the readHBmap.py script [59]. The replicates for each RMSD, RMSF, and H-bond analysis were represented as R1, R2, and R3, corresponding to the three independent MD simulations. Replicate selection was made based on RMSD profiles, aiming to choose an intermediate trajectory between the extremes for further analysis. XMGrace was utilized for calculating and plotting all graphs. The protein structure visualization and the tunnel analysis were conducted using multiple tools. UCSF Chimera V1.15 [60] and PyMOL V 2.4 [61], the visualization tool, was utilized for figure generation of protein structures and analysis. For the identification of tunnels in different CaMLS1 systems, the caver web tool (https://loschmidt.chemi.muni.cz/caverweb/) [62] was utilized. The starting point for tunnel detection was kept constant across all systems and was located in the active site region. A probe radius of 0.9 Å and a shell depth of 3 Å were used, while other parameters were maintained at their default values. Essential dynamics studies and dynamic cross correlation analysis To access the protein conformational flexibility, principal component analysis (PCA)-driven free energy landscape (FEL) studies were employed to identify the collective motions and Gibbs free energy of CaMLS1 (apo-protein), CaMLS1-Cofactors, (Mg 2+ ) and CaMLS1-Cofactor with substrates in a simulation trajectory, determining both the direction and magnitude of the motions. The PCA-FEL studies were conducted using GROMACS 2019 package using various utilities. For the analysis of essential dynamics, the covariance matrix was calculated and diagonalized using gmx_covar, yielding a set of eigenvalues and eigenvectors obtained with gmx_anaeig. These eigenvectors (representing motion directions) along with their associated eigenvalues (indicating motion amplitude) collectively illustrated the fluctuation of conformations within the CaMLS1 systems. The majority of the motions observed in the CaMLS1, CaMLS1-cofactor, and CaMLS1-cofactor-substrates were captured within the first ten eigenvectors. Cosine content calculations were performed based on these eigenvectors using gmx_analyze. The FEL plots for each simulated system were generated using principal components (PC) with a cosine content of less than or equal to 0.5 [63]. The free energy profile was generated using gmx_sham, utilizing the projections from the top two principal components (PC2 and PC3) were obtained from the PCA analysis. Further, these FEL plots aided in extracting the stable conformations of each system. The dynamic cross-correlation map (DCCM) analysis was conducted to examine the movements of CaMLS1 Cα atoms and their correlation systems. The MD trajectory file (.xtc) was converted into (.dcd) format using VMD software [64]. The analysis was performed using the bio3d package in R studio [65, 66]. An in-house R script was developed for plotting the positive and negative correlations separately and also to calculate the domain-domain correlations for CaMLS1 complexes. MM/PBSA binding free energy calculation and per-residue decomposition analysis The binding free energy calculation was performed to understand the contribution of the binding energy of each amino acid of the substrate bound complexes. Each simulated complex’s MM/PBSA-based binding free energy (ΔG bind ) was calculated with g_mmpbsa v 5.1.4 package in the GROMACS platform [67]. The 100 ns MD simulation trajectory with 1 ns interval was utilised to calculate the binding energy of the complexes. Furthermore, the energy contribution for each residue in the complex was calculated by using the MmPbSaDecomp.py python script (https://rashmikumari.github.io/g_mmpbsa/) [67]. Phylogeny analysis, Secondary structure prediction and Homology modeling of CaMLS1 The protein sequence of C. albicans malate synthase (CaMLS1) was subjected to Protparam tool in order to evaluate the physico-chemical parameters. The CaMLS1 consisted of 551 residues and possessed a molecular weight of 62.52 kDa. The theoretical pI (8.74) and instability index (46.13) signified that the protein will have a net positive charge at neutral pH and is likely to be unstable. The aliphatic index and GRAVY were computed to be 88.33 and -0.326 respectively, indicating a thermally stable and hydrophobic in nature (GRAVY <0; hydrophobic) [68]. The multiple sequence alignment of amino acid sequence of CaMLS1 was carried out with non-albicans species as shown in Fig. S1. The percent identity was calculated for C. albicans with non-albicans species including B. anthracis . It was observed that C. albicans’ CaMLS1 showed significant identity with malate synthases of C. africana (99.82%), C. dubliniensis (98.55%), C. tropicalis (94.37%), C. viswanathii (93.65%), C. orthopsilosis (88.57%), C. parapsilosis (88.93%), C. auris (81.67%), C. glabrata (54.92%) and B. anthracis (46.58%). Since no better-characterised fungal homolog for malate synthase was available, conservation of cofactor and catalytic sites were examined in Candida sp. using B. anthracis as a reference. From the alignment, the catalytic residues from N281 to Y287 (represented with a black box in Fig. S1) were found to be conserved across all the Candida sp . Likewise, the cofactor binding sites at catalytic site (site-1) (E258 and D286) (marked with star in blue colour in Fig. S1) were also conserved. However, the site-2 binding residues (N297 and H298) are conserved only in some Candida sp. On the other hand, when the multiple sequence analysis of CaMLS1 was performed with malate synthase of other bacterial pathogens (Fig. S2), the P. brasiliensis (55.90%), B. anthracis (46.58%), showed optimal identity but M. tuberculosis (10.92%), P. aeruginosa (11.16%) and E. coli (10.05%) showed much lower identity. Interestingly, the cofactor binding sites (E258 and D286) (shown in blue star) were conserved in MLS1 of bacterial pathogens too. However, the catalytic residues (NCGRWDY) which were conserved among the Candida sp . (Fig. S1) (represented with rectangular box in Fig. S1) were found to be less conserved in other organisms (Fig. S2). Phylogenetic analysis was carried out to establish the evolutionary relationship of CaMLS1 protein with MLS1 of other pathogenic organisms (P. brasiliensis, B. anthracis, E. coli, M. tuberculosis, P. aeruginosa) . A Maximum Likelihood (ML) based Phylogenetic tree was constructed with the best fitting model LG+I+G4 model and 1000 bootstraps. The phylogenetic tree of MLS1 diverged into two clusters, as shown in Fig. 1a. The cluster with clades represented in red are related to fungal species, while the clades in purple corresponded to bacterial species. It was observed that the CaMLS1 is more closely related to other Candida sp. Among the fungal species, P. brasiliensis also showed an evolutionary relationship by sharing a common ancestor with Candida sp. On the otherhand, among the bacterial spp., M. tuberculosis, E. coli and P. aeruginosa were closest relatives. The secondary structure of the CaMLS1 protein was predicted by PSIPRED (Fig. S3a) and GOR IV (Fig. S3b). Fig. 1b shows the comparative secondary structure composition of helices, strands, and coils of CaMLS1 as reported by PSIPRED and GORIV servers. From these predictions, it was observed that the CaMLS1 protein has more coils followed by α-helices and extended strands. Since the X-ray structure of CaMLS1 has not been reported till now, the 3D model was built using Robetta server through a homology modeling approach. The crystal structure of MLS1 from Bacillus anthracis (PDB ID:3CUX_A) was selected as template which has a query coverage of 94%, E-value (0) and an identity of 51.61% with CaMLS1. Five models were built by the Robetta server (employing comparative modeling) and the best model was chosen based on the structural quality and stereochemical properties. This model was subjected to model refinement using ModLoop, which was subsequently validated using SAVES server . The best model was chosen based on the structural quality and stereochemical properties (Fig. 1c). The model, thus identified, showed an ERRAT value of 95.0 while the PROCHECK results revealed 89.4% residues in the most favored region and 10.6% residues in the additionally allowed region (Fig. S3c). The CaMLS1 model was subjected to secondary structure analysis using PDBSum. The CaMLS1 protein model consisted of 27 helices and 13 strands. Other structural elements include two antiparallel and one parallel sheets, 6 β-α-β units, 2 beta hairpins, and one beta bulge. It also exhibited 34 helix-helix interactions, 31 beta turns, and 10 gamma turns. Upon sequence alignment of the template sequence with CaMLS1, high sequence identity was observed. Three domains were reported in the template structure hinting the presence of three domains in CaMLS1. The sequence alignment resulted in the identification of the three domains in CaMLS1, namely, N-terminal (15-67), TIM-barrel (100-394), and C-terminal (429-532) domains (Fig. 1c). The superimposition of template (blue in Fig. 1d) and the modeled structure (orange in Fig. 1d) showed an RMSD of 0.65 Å. The alignment of the three domains was also carried out (Fig. S4). It was observed that the N-terminal domain had an RMSD of 0.71 Å (Fig. S4a), the TIM-barrel with 0.55 Å (Fig. S4b), and C-terminal domain with 0.66 Å (Fig. S4c). Molecular dynamics simulations of CaMLS1 The molecular dynamics simulation for CaMLS1 was carried out in triplicate for 100 ns. The RMSD profiles for the three MDS runs were shown in Fig. S5a. From the RMSD plot, the CaMLS1 was found to be stabilised at 0.44 nm, with an average RMSD of 0.36±0.05 nm across the three replicates, demonstrating consistent behaviour between simulations. The RMSD profile of representative run was shown in Fig. 2a. To analyze the domain specific contribution to the dynamics of the CaMLS1 protein, the RMSDs of all three domains were extracted and plotted (Fig. 2b and Fig. S6(a-c)). Among all these domains, the N-terminal (in red) and C-terminal (in cyan) domains of CaMLS1 showed a lower deviation (Fig. 2b). In case of TIM-barrel (in orange) (which has cofactor binding sites), an ascending RMSD pattern was observed with 0.34 nm at the end of the simulation. The average RMSD for each domain of CaMLS1 was computed. It was observed that the average RMSD values of N-terminal (0.15±0.02 nm) and C-terminal (0.19±0.01 nm) domains were found to be lower compared to TIM-barrel domain (0.31±0.04 nm). These RMSD profiles indicated that an increasing RMSD trend of CaMLS1 can be attributed to the dynamic nature of the TIM barrel domain. The RMSF profiles of Cα atoms present in the N-terminal, TIM-barrel, and C-terminal domains of CaMLS1 were displayed in Fig. 2c and Fig. S7a. The major peaks in the RMSF profiles of CaMLS1 were numbered consecutively. It was observed that the major peaks with flexible residues were located either in the loop regions or loops connected to the helices. Further, the B-factor values for CaMLS1 were computed and projected, providing a 3D representation of structure flexibility (Fig. 2d). It showed the highly fluctuated regions in correlation with the numbered peaks of the RMSF profile. The flexible regions of CaMLS1 were found to be present at the N-terminal domain (indicated by arrow 1) and TIM-barrel domains (arrows 3 and 4) indicated by the thicker regions. Furthermore, the principal component analysis (PCA) based free energy landscape (FEL) studies was performed to extract the lowest energy conformation of CaMLS1 protein. The 2D contour plot (Fig. 2e) for the CaMLS1 protein was obtained from the two principal components. The 2D -FEL map presented three minima energy clusters for CaMLS1. Among the clusters, the lowest energy conformation was identified at 93.57 ns time frame. Role of Mg 2+ ions in CaMLS1 stability The template structure (with PDB ID: 3CUX) has two Mg 2+ metal ions (one at the catalytic site and the other on the surface) . A step-by-step procedure was followed to investigate the role of Mg 2+ metal ions in maintaining the structural stability of CaMLS1 . To model the CaMLS1 with the Mg 2+ metal ions, the metal binding sites of template (GLU247, ASP275, ASN286, and HIS287) were aligned to the modeled CaMLS1 structure (GLU258, ASP286, ASN297, HIS298) using PyMol. The 3D coordinates of the template metal binding sites were matched to the CaMLS1 modeled structure. The final structures were named (Fig. 3a) as follows: CaMLS1 with two Mg 2+ ions (CaMLS1-Mg), CaMLS1 with one Mg 2+ at the catalytic site (CaMLS1-Mg-1), and CaMLS1 with one Mg 2+ on the surface (CaMLS1-Mg-2). The close-up view of Mg 2+ ions for each of these three systems were presented in Fig 3a along with a schematic representation illustrating the presence of metal ions in the CaMLS1 protein. To explore the structural stability of CaMLS1 with Mg 2+ ions, MD simulations were performed for 100 ns in triplicates (shown in fig. S5(b-d)). For examining the stability of CaMLS1, RMSD profiles for CaMLS1 protein backbone was extracted and presented in Fig. 3b. It was observed that the RMSD for all three systems (CaMLS1-Mg, CaMLS1-Mg-1 and CaMLS1-Mg-2) was stabilized during the 100 ns simulation period with different deviations for each system. At the end of simulation, the RMSD of CaMLS1-Mg (purple in Fig. 3b) was 0.45 nm, while CaMLS1-Mg-1 (cyan in Fig. 3b) and CaMLS1-Mg-2 (green in Fig. 3b) showed a deviation of 0.41 nm and 0.40 nm. The 2D-interactions and 3D structures of the three systems were extracted at 100 ns and were presented in Figs. S8 and S9(a-c), respectively. Additionally, the Mg 2+ interactions remained stable across all replicates. The average RMSD for each case was computed. It was observed that, the average RMSD values were found to be 0.39±0.04 nm (CaMLS1-Mg), 0.38±0.04 nm (CaMLS1-Mg-1) and 0.37±0.05 nm (CaMLS1-Mg-2). To assess the impact of Mg 2+ on the cofactor binding sites within the TIM-barrel of CaMLS1, the RMSD profiles of TIM-barrel for three systems were generated (Fig. 3c and Fig. S10(a-c)). The RMSD plot of TIM-barrel in CaMLS1(black in Fig. 3c) showed a deviation of 0.34 nm at 100 ns of time scale. The TIM-barrel domain of CaMLS1-Mg (purple), CaMLS1-Mg-1(cyan), and CaMLS1-Mg-2 (green) showed stable RMSD profile (from 15 ns onwards) with deviation of 0.39, 0.28 and 0.31 nm respectively. A comparison of RMSD profiles of TIM-barrel from each case showed some deviations, indicating minor differences in structural fluctuations among the systems (with and without Mg 2+ ). The RMSF plots of CaMLS1-Mg, CaMLS1-Mg-1, and CaMLS1-Mg-2 were displayed in Fig. 4, while the corresponding triplicate RMSF for each system were presented in Figure S7(b-d) The fluctuations corresponding to the loops present in CaMLS1 were numbered consecutively. The peaks 2 and 3 were consistently present in all three cases studied but varied in degree of fluctuation. The RMSF profile of Cα atoms in CaMLS1-Mg-2 demonstrated less fluctuation in comparison with the Cα atoms of CaMLS1-Mg, CaMLS1-Mg-1. Interestingly, the fluctuations were found to be either in the loops or loops connecting helices. Additionally, the RMSF values of CaMLS1, CaMLS1-Mg, CaMLS1-Mg-1, and CaMLS1-Mg-2 were plotted in black, purple, cyan, and green respectively (in Fig. S11a). The binding sites were represented in the RMSF plot. It was observed that the fluctuations on the cofactor binding sites were decreased when compared with CaMLS1 system without Mg 2+ (in Fig. S11b). Further, the flexibility of CaMLS1-Mg, CaMLS1-Mg-1, and CaMLS1-Mg-2 can be assessed during the MD simulations by analysing their B-factor. The B-factor values for each system were computed and projected, providing a 3D representation of structure flexibility (in Fig. 4(d-f)). Three flexible regions were found in all three systems with varying degrees of fluctuations. For instance, the thickness seen in B-factor representation decreased from CaMLS1-Mg to CaMLS1-Mg-2. Among the Mg 2+ bound complexes, the CaMLS1 with two Mg 2+ ions were considered for further analysis. The protein collective motions and minimum energy conformations for CaMLS1-Mg systems were studied through PCA-based free energy landscape (FEL) analysis. The 2D contour plots were generated for each individual system of CaMLS1-Mg, CaMLS1-Mg-1, and MLS1-Mg-2, which are shown in Fig. S12. The FEL plot of CaMLS1-Mg (Fig. S12a) showed four different clusters, and the corresponding least energy conformation was extracted at the 83.32 ns time scale. Similarly, for the CaMLS1-Mg-1 (one cluster) and CaMLS1-Mg-2 (five clusters), the FEL plots and their corresponding conformations were extracted from clusters at 83.85 ns, and 29.11 ns (Figs. S12b and c) respectively. The extracted structures were validated through ERRAT, Verify3D, and Ramachandran’s plot analysis (Table S1). The structures indicated high structural quality and stereochemical reliability, with residues 81-84% in favoured region. CaMLS1 with cofactors and substrates A proper understanding of interaction and biological activity of a macromolecular complex requires the proper orientation and docking score of ligands towards macromolecule. Molecular docking of substrates (GOXL and ACOA) was performed with CaMLS1 protein in the presence of the Mg 2+ ions. The validated model of CaMLS1 with two Mg 2+ ions (CaMLS1-Mg) was considered for these docking studies. Three different sets of docking were performed: 1) CaMLS1-Mg-ACOA-GOXL (MMAG), 2) CaMLS1-Mg-GOXL (MMG), and 3) CaMLS1-Mg-ACOA (MMA). A sequential docking approach was adopted for docking the MMAG complex. Initially, the substrate GOXL was docked to the MLS1-Mg protein. The best pose of the MLS1-Mg-GOXL complex was used to dock the ACOA substrate. The GOXL substrate showed a glide score of -6.876 kcal/mol, with two hydrogen bonds one each with ARG174 and Mg 2+ ion (in Figs. 5A (b, d)). On the other hand, the ACOA substrate exhibited a high binding efficiency towards CaMLS1-Mg-GOXL with a glide score of -9.313 kcal/mol (in Figs. 5A (c, e)). The substrate ACOA in the complex showed eight H-bond interactions with the target protein. From these results, it was observed that both substrates showed a potential to interact with the key active site residues of CaMLS1 with strong affinity. In the case of the MMG complex, the substrate GOXL was docked to the CaMLS1 and showed a docking score of -6.898 kcal/mol. From the pose, it can be depicted that the GOXL interacted with ARG174 and Mg 2+ ions of CaMLS1 protein (Figs. 5B(a-c)). The detailed interactions of substrates with different residues of CaMLS1 were shown in Table S2. The substrate ACOA in MMA complex showed a significant docking score of -8.279 kcal/mol towards the CaMLS1 protein, with nine H-bonds. A thorough 2D and 3D interactions has been depicted in the Fig. 5C (a-c). Substrate-induced dynamics studies To evaluate the stability of interactions and conformational changes in the substrate-bound complexes (MMAG, MMG, and MMA), a 100 ns molecular dynamics simulation was performed in triplicate, and trajectories were examined for each system. The RMSD values for complex backbone were calculated and presented in Fig. 6a and the triplicates data was shown in Fig. S5(e-g). At first, the RMSD graph of all the systems showed different degrees of deviations (till 40 ns of time scale) and later reached their equilibrium. The RMSD of complex MMG (blue in Fig. 6a) showed a deviation of 0.40 nm, while the MMA complex (orange in Fig. 6a) displayed a deviation of 0.36 nm. The MMAG (green in Fig. 6a) exhibited a deviation of 0.34 nm at the end of simulation. The structures of the MMAG (in dark green), MMG (in blue) and MMA (in orange) at 100 ns time stamp were presented in the Fig. S9(d-f)). The average RMSD values calculated for MMAG, MMG and MMA were 0.32±0.03, 0.38±0.04, and 0.32±0.04 respectively. Among the three complexes, the MMAG and MMA showed a least deviation when compared with the MMG complex. To evaluate the substrate stability in the complex, the RMSD of substrates were plotted (Fig. 6b) for three systems. The RMSD of the substrate GOXL in MMG (blue in Fig. 6b) and GOXL in MMAG complex (green) remained constant at 0.05 nm, the trend was consistently seen in replicates (Fig. S13(a-b)). The RMSD of ACOA (orange) in MMA displayed high deviation with fluctuations (between 35 and 55 ns) and reached equilibrium with an RMSD of 0.47 nm. The ACOA in MMAG complex (dark green) exhibited less deviation with RMSD of 0.3 nm which implied a minimal deviation from the initial pose. From the Fig. S13(c-d), though the ACOA was dynamic in one of the replicates, the ACOA interactions were intact to the MLS1 across the replicates. Further, the RMSD plot for TIM-barrel domain in each system was analysed (Fig. 6c, Fig. S10(d-f)). The TIM-barrel in complex MMG exhibited a deviation of 0.40 nm, whereas in MMA and MMAG complexes, deviation of 0.31 nm was observed for both. These results indicated that the TIM-barrel domain undergone minimal conformational change in case of both substrates (MMAG) and ACOA (MMA) when compared to GOXL (MMG) substrate. To examine the changes at the residue level, the RMSF profiles were calculated (Fig. 6(d-f), Supplementary Fig. S7(e-g)). The RMSF profile of Cα atoms in MMAG demonstrated less fluctuation in comparison with the Cα atoms of MMG and MMA. The peaks in the RMSF profiles were numbered consecutively, and those with major fluctuations were highlighted in the 3D structure using B-factor putty representation. Interestingly, these peaks were found to be either in the loops or loops connecting helices. Additionally, the flexibility of MMAG, MMG, and MMA can be evaluated during the MD simulations by analysing their B-factor. The B-factor values for each complex were calculated and visualised, by providing a 3D representation of structure mobility (in Fig. 6(g-i)). The degree of fluctuation in the flexible regions increased progressively from MMAG to MMG and was highest in MMA. Similarly, the thickness observed in the B-factor representation followed the same trend, with MMAG showing the least and MMA the highest level of flexibility. Upon keen observation of the RMSD profiles, the backbone of MMAG complex as well as its TIM-barrel domain, showed less deviation and minimal conformational change compared to other systems. Similarly, the RMSF profile of MMAG also exhibited considerably less fluctuation and fewer peaks in the TIM-barrel region. The hydrogen bond interactions in CaMLS1-substrate complexes of MMG, MMA, GOXL and ACOA in MMAG complex were analysed throughout 100 ns of MD simulations (Fig. 7(a-d)). The GOXL in MMG (in Fig. 7a) and GOXL in MMAG complex (Fig. 7c) showed an average hydrogen bond of two and one, respectively, with CaMLS1 protein. In contrast, the ACOA substrate demonstrated stronger H-bond interactions, forming an average of four hydrogen bonds in MMA (Fig. 7b) and five in MMAG complex (Fig. 7d). This H-bond profile for both substrates GOXL and ACOA was consistently observed across the replicates, as shown in Fig. S14. The hydrogen bond occupancy of GOXL in MMG & MMAG complex and ACOA in MMA & MMAG complexes were analysed along MD simulation trajectories (Fig. 7(e-f)). From the MD trajectories, it was observed that GOXL substrate in MMG showed hydrogen bond occupancy of 85.8% with residue TRP285 (Fig. 7e). In case of MMAG complex, the GOXL formed a hydrogen bond with the residue ARG306 with a comparably lower occupancy of 21.2% (Fig. 7g). Analysis from the replicate data also revealed that the GOXL in MMG exhibited the stable interactions with TRP285 across all replicates, although with varying occupancy rates (Fig. S15A(a-c)). In contrast, in the MMAG complex, the occupancy rate was consistently lower for all replicates (Fig. S15B). However, it was observed that the H-bond interactions of GOXL in MMG and MMAG complexes were slightly different, yet all these residues were in near vicinity (Figs. S15A(d) & B(d)). The observation indicated stable interactions in MMG compared to MMAG. Likewise, the occupancies of ACOA in MMA (shown in Fig. 7f) and in MMAG complex (shown in Fig. 7h) were depicted. The ACOA showed persistent interactions in MMA and MMAG simulated complexes with the docked poses. The LYS106 interacted with three different side groups of ACOA (in MMA) with H-bond occupancies of 35.7%, 61.2%, 20.4% respectively. Two H-bonds between LYS156 and ACOA were found with 31.3 and 35.1% occupancy rates. Yet another interaction with occupancy of 23.8% was TYR158 with ACOA substrate (Fig. 7f). The residues LYS106 and LYS506 consistently showed interactions with ACOA in MMA complex in all replicates (Fig. S16A(a-c)). Interestingly, the other interacting residues, though not common in replicates, were also positively charged and located in the vicinity of the ACOA binding site (Fig. S16A(d)). On the other hand, the ACOA in MMAG complex, the LYS106 formed the hydrogen bonds with three side groups of ACOA with occupancies of 81.7%, 49.8% and 27.3%. The residue ASN110 showed an occupancy of 75.4% and 23.4% with two side groups of ACOA molecule. An additional hydrogen bond with TYR158 showed an occupancy of 42.2% (shown in Fig. 7h). Similarly, in the MMAG complex, ACOA also interacted with positively charged residues located near the binding site, as observed in the MMA complex, but with higher occupancy (Fig. S16B(a-d)). Furthermore, the 2D interactions of the substrate in the CaMLS1 were depicted by taking the final frame of MD trajectory. In MMG complex (Fig. 7i), the GOXL substrate formed four hydrogen bonds with TRP285 and ARG284. Additionally, one electrostatic interaction (with Mg 2+ ) and five hydrophobic interactions were observed. In MMA complex (Fig. 7j), the substrate ACOA formed eight hydrogen bonds with the residues LYS106, LYS156, TYR158, HIS377 and CYS449 with eleven hydrophobic interactions. In MMAG complex, the two substrate interactions were also analysed separately, as shown in Fig. 7k-l. The GOXL substrate in MMAG showed one hydrogen bond with ARG306, one electrostatic interaction with Mg 2+ , and five hydrophobic contacts with CaMLS1 residues. On the other hand, ACOA in MMAG exhibited five hydrogen bonds with LYS106, ASN110 and LYS156, and seven hydrophobic interactions with CaMLS1 residues in complex. Interestingly, all these hydrogen bonds showed a good occupancy rate (Fig. 7 (e-h)). Overall, the hydrogen bond occupancy was correlated with hydrogen bond plots and 2D interaction studies in relation to the stability of the system in the presence of substrates. The GOXL substrate in MMG showed strong interactions when compared to MMAG, while ACOA substrate showed uniform type of interacting residues which were identified in docking studies. To illustrate how the substrate binding influences the conformational distributions, free energy landscapes (FEL) for MMAG, MMG and MMA complexes were determined and shown in Fig. S17. The FEL plot of MMAG (Fig. S17a) revealed two clusters, each corresponding to multiple metastable conformations associated with distinct free energy minima. Similarly, the FEL plot of MMG (Fig. S17b) indicated a stable global minimum with a single cluster, and the corresponding minimum energy conformation was extracted at the 86.39 ns time scale. For MMA (Fig. S17c), two clusters were identified, representing several metastable states. The corresponding conformations of MMA and MMAG were extracted from clusters at 94.60 ns, and 47.32 ns respectively. The extracted structures were validated through ERRAT, Verify3D, and Ramachandran’s plot analysis (Table S1). The complexes exhibited high structural quality and good stereochemical integrity with 80% of residues in the favoured region. DCCM analysis To investigate the correlated and uncorrelated conformational motions of CaMLS1 complexes, dynamic cross-correlation matrices (DCCM) analysis was performed. The plots for CaMLS1 complexes were shown in Fig. 8 (a-g). From the DCCM plot, the regions with positive correlation (colored in blue to light blue shades) indicated residue pairs which move in the same direction (upper triangle in each panel), while regions of negative correlation (colored from orange to red) indicated residue pairs which move in opposite directions (lower triangle in each panel), while the diagonal line corresponded to self-interaction of residues. As shown in Fig. 8, all panels exhibited a noticeable change in the pattern of negative correlations, as well as positive correlations. Among the seven systems, CaMLS1 has more negative correlation motions (predominantly red in the lower diagonal Fig. 8a), followed by MMA, MMG, CaMLS1-Mg-1, and CaMLS1-Mg-2. This indicates that dynamism of the system increases MMA to CaMLS1. Furthermore, to analyse the dynamics and conformation rearrangements, the map is divided into different regions for analysis. The correlation between N-terminal and C-terminal (long dash), TIM-barrel and C-terminal (solid line), TIM-barrel and N-terminal (dashed line) were boxed individually as shown in Fig. 8 (a-g). Further, the correlations between N-terminal and N-terminal (dotted box), TIM-barrel and TIM-barrel (dots and dashes), C-terminal and C-terminal (dot dot dash lines) were also shown along the diagonal. The positive (upper triangle) and negative (lower triangle) correlations for each domain pair was summed up individually using an in-house R script and was shown in Fig. S18. For a given pair of domains, the net correlation was considered after taking both the positive and negative correlations. The following is detailed analysis based on the net correlations. Positive and negative correlations in systems with cofactor alone The correlation between the N-terminal and C-terminal residues was consistently slightly negative across all four systems (CaMLS1, CaMLS1-Mg, CaMLS1-Mg-1 and CaMLS1-Mg-2), with values ranging from -0.02 to -0.08. The residues from N-terminal and TIM-barrel showed consistently a correlation close to zero in all four systems (ranging from -0.04 to +0.04). The residues from C-terminal and TIM-barrel showed slightly negative correlation in all four systems in the range of -0.03 to -0.07. The correlation between the residues of N-terminal and N-terminal showed predominantly positive correlation with varying degrees of net correlation. The CaMLS1showed the highest of all four systems with a net correlation of +0.27 while the CaMLS1-Mg-2 showed the lowest net correlation (+0.17). Likewise, the correlation between the residues of C-terminal and C-terminal showed predominantly positive correlation with a less degree of net correlation compared to CaMLS1 (Fig. S18d). On the other hand, the correlation between the residues of TIM-barrel and TIM-barrel showed no variation in net correlation for all four systems (Fig. S18 (a-d)). Positive and negative correlations in systems with substrates The correlations between the residues of N-terminal and C-terminal, N-terminal and TIM-barrel and C-terminal and TIM-barrel showed almost no significant correlation for MMAG, MMG and MMA complexes (Fig. S18 (e-f)). Similar to the net correlation observed with cofactors, the correlation between the residues of N-terminal and N-terminal showed predominantly positive correlation with varying degrees of net correlation. The MMAG showed the lowest of all three systems with a net correlation of +0.19 while the MMG and MMAG showed the net correlations of +0.28 and +0.29 respectively. Likewise, the correlation between the residues of C-terminal and C-terminal showed predominantly positive correlation for MMA (+0.30) when compared to other two complexes (Fig. S18 (e-f)). On the other hand, the correlation between the residues of TIM-barrel and TIM-barrel showed no variation in net correlation for all three systems (Fig. S18 (e-f)). MM-PBSA analysis To evaluate the binding energetics of the protein-substrate complex, the polar and non-polar components of binding free energy (BE) were calculated for 100 ns time scale with 1 ns interval. Binding free energies of GOXL and ACOA substrates in the three complexes (MMAG, MMG, and MMA) were calculated using MM/PBSA method (in table S3) and represented in Fig. S19. The binding energies of GOXL substrate in MMAG and MMG complexes were found to be -6 and 3 kJ/mol respectively. Likewise, the ACOA in MMAG and MMA complexes exhibited binding energy of -921and -881 kJ/mol (Table S3). From the results, it is interpreted that both substrates showed better binding energy in MMAG complex. Furthermore, per-residue energy analysis was performed to obtain the hotspot residues contributing to the total binding energy involved in binding of the substrate (Fig. S20). The analysis revealed a pattern where positively charged residues contributed high negative energy (Fig. S20), while negatively charged residues contributing to high positive energy (data not shown). Interestingly, a change in substrate energy was also observed when bound alone (MMG and MMA) and when combined (MMAG). From Fig. S20a, the GOXL substrate in MMA had an energy of -0.88 kJ/mol, while the same GOXL in MMAG complex (Fig. S20c) exhibited a higher energy of -17.13 kJ/mol. The ACOA substrate in MMA contributed an energy of -510.29 kJ/mol (Fig. S20b), while in MMAG complex (Fig. S20d), the energy was -492.15 kJ/mol. The common residues of GOXL contributing to the binding energy were identified in both MMG and MMAG complexes as ARG55, ARG62, ARG174, ARG176, LYS223, LYS284, LYS293 and LYS362. Likewise, for ACOA in MMA and MMAG complexes, six residues (ARG105, LYS132, LYS156, LYS345, LYS386, ARG445, LYS506, and LYS510) were found to be in common and contributed to the binding energy. These results suggest that the binding interactions are energetically favourable and involve certain structural changes that stabilize the complexes. Discussion The glyoxylate pathway is a known mechanism adapted by the pathogens to survive in nutrient-limited conditions. C albicans ’ MLS1 plays a prominent role in adaptation, virulence and pathogenesis in the glyoxylate pathway and hence considered as a potential drug target. Interestingly, the CaMLS1 bears no similarity with any human enzymes [18, 69]. Hence, targeting the CaMLS1, with no homologues present, is a safe way of treating candidiasis infections. Further, the CaMLS1 requires Mg 2+ cofactors and GOXL and ACOA substrates for its activity [29]. A clear molecular level understanding of the CaMLS1 and the cofactors/substrates interactions is essential in order to disrupt the glyoxylate pathway. With no crystal structure for CaMLS1 is reported so far, this in silico inferences will bridge the existing gap to design inhibitors for glyoxylate pathway. Identification of molecular weight, pI and other physico-chemical properties through any biochemical techniques (such as 2-D gel electrophoresis and mass spectrometry) for an enzyme is tedious and time-consuming. In contrast, ProtParam tool is a valuable source for determining the physico-chemical properties of an enzyme, as it provides information about the molecular weight, pI, amino acid composition, stability, and solubility which could be important for the biological function. MLS1, in general, exists in two isoforms, isoform A and isoform G, in different species. Isoform A is frequently found in fungi while isoform G is reported in bacterial species. Structurally, both isoforms A and G are known to possess N-terminal, TIM-barrel, and C-terminal domains. However, the isoform G is known to have an extra domain, the α/β domain, at the interface of N-terminal and TIM-barrel domains. The multiple sequence alignment, shown in Fig. S2, clearly indicates the lack of α/β domain in isoform A. Also, this lack of α/β domain resulted in a low alignment rate between Candida and other bacterial species. Structural analysis of CaMLS1 The model of CaMLS1 was subjected to the secondary structure analysis using PDBSUM and compared with the results obtained from PSIPRED and GOR IV. PSIPRED and GOR IV tools have different predictions while the PDBSUM supported the PSIPRED predictions (Fig. 1b). Further, the CaMLS1 structure was subjected to domain analysis using InterPro. The InterPro predicted three domains for CaMLS1; the N-terminal domain (13-76); the TIM-barrel domain (170-416) and C-terminal domain (423-539). However, the TIM-barrel predicted by InterPro appeared to be incomplete. On the other hand, the template (PDB ID: 3CUX) used for modeling the CaMLS1 is reported to contain the same domains. A structural and sequence alignment of CaMLS1 model with the template (3CUX) indicated the presence of the domains as: N-terminal (15-67); TIM-barrel (100-394), and C-terminal (429-532). A close comparison between these domain analyses indicated a deviation in TIM-barrel. Interestingly the InterPro predicted TIM-barrel has missed out a few helices and sheets which were included in the alignment with the template. Hence, the TIM-barrel domain was taken as predicted by the template alignment i.e., 100-394 (Fig. 1c). Further, with this assignment, the TIM-barrel exhibited a best fit with an RMSD of 0.55Å among all three domains (Fig. S4b). Influence of Mg 2+ in CaMLS1 systems Metal ions play a key role in the activity of enzymes. The RMSF profiles of CaMLS1 and CaMLS1-Mg (black and purple in Fig. S11a) showed a considerably decreased fluctuation for CaMLS1-Mg. A close look at the TIM-barrel region, which included the Mg 2+ binding sites, also revealed the same observation (Fig. S11b). This indicated an Mg-induced conformational change in CaMLS1. Also, an overlay of CaMLS1 and CaMLS1-Mg (both FEL generated) resulted in an RMSD of 3.9 Å, which further supported a significant Mg-induced conformational change. Interestingly, Okada et al reported that Mg 2+ dependent activity for MLS1 in C. tropicalis [29] . Since CaMLS1 shared a 94% similarity with MLS1 of C. tropicalis and possessed conserved cofactor binding site residues, it is reasonable to expect a similar Mg 2+ dependent activity in C. albicans . A keen observation of the simulation data, as explained above, revealed an Mg-induced conformational change in CaMLS1, which indicated a biologically relevant model of CaMLS1. Substrate induced changes in CaMLS1 systems MLS1, in addition to Mg 2+ , also requires two substrates – the GOXL and ACOA for its enzymatic activity [29]. A comparison of FEL profiles indicated the presence of two clusters, each for MMAG and MMA while only one cluster was seen for MMG (Fig. S17). This can be understood as ACOA (present in both MMAG and MMA) having greater influence on CaMLS1 compared to GOXL. This can also be observed in the hydrogen bond pattern (Fig. 7). The average H-bonds for ACOA in MMA is ~ 4 while it is ~ 2 for GOXL in MMG complex. During the simulation, the CaMLS1-Mg showed a deviation of 0.45 Å whereas the MMAG showed a considerable deviation with an RMSD of 0.34 Å at 100 ns (green in Fig. 6a). Interestingly, the RMSD for MMA and MMG were seen at 0.36 and 0.40 Å (orange and blue in Fig. 6a) indicating a larger deviation of MMAG from MLS1-Mg compared with MMA and MMG. This pattern suggests an increased deviation in the CaMLS1 and CaMLS1 with Mg²⁺, while the substrate-bound systems maintain relatively lower fluctuations. The dynamics of linker regions can be seen by looking at the gaps between TIM-barrel and C-terminal domain. It was observed that MMG and MMA complexes had more negative correlation in the linker region (lower triangle for linker regions in Fig. 8f and g) as compared to MMAG (Fig. 8e) (highlighted by a red solid circle). Hence, a lowered negative correlation observed for MMAG can be depicted as less dynamic and hence more stability for MMAG. On the other hand, the binding profiles also inferred a similar finding. The substrate GOXL showed a weaker binding affinity, while the ACOA showed stronger binding affinity (Fig. S19 and Table S3). This was also seen in MMAG complex where GOXL showed a weaker binding affinity than ACOA (Fig. S19 (c&d). Likewise, GOXL formed fewer hydrogen bonds in both the MMG and MMAG complexes (Figs. 7a&c and S15). Further, the dominant standard deviation values (in Table S3) of GOXL in MMG and MMAG could be the reason for poor binding and instability in interactions, as reflected in the H-bonds. This is in correlation with the finding of Okada et al where a binding affinity of 1mM for GOXL and 80 µM for ACOA was reported for MLS1 in C. tropicalis [29]. Key residues of CaMLS1 Through the MD studies, the potential residues for CaMLS1 were identified based on H-bond occupancy (Fig. 7 (e-h)) as well as per-residue decomposition analysis (Fig.S20 (a-d)). The residues ARG174, LYS223, ARG284, LYS293, TRP285, and LYS362 were found to be favorable in the GOXL binding region, whereas the residues ARG105, LYS106, ASN110, LYS132, LYS156, LYS345, ARG445, LYS506 and LYS510 were favored in binding of ACOA substrate. It can be observed that most of these are positively charged. Therefore, the inhibitor molecules should be designed to have negative charged moiety, in order to exhibit greater affinity for binding to the CaMLS1 protein. Interestingly, the key residues identified in this study, which are involved in cofactor and substrate binding, were found to be highly conserved across Candida species (Fig. S1). This observation indicates that the insights of the study help in understanding of the malate synthase across Candida sp. highlighting their conserved structural and functional features that are critical for malate synthase activity. Therefore, designing an inhibitor for CaMLS1 could also be utilized for other Candida sp. Role of cofactors and substrates in surface alterations of the CaMLS1 From the FEL structures, it was observed that the CaMLS1 complexes showed noticeable surface alterations leading to tunnels directed towards active site regions. Tunnel dynamics facilitated by loop regions In some enzymes, the catalytic site is located deep within the protein structure, while in others it is exposed on the surface. For those with buried active sites, the transport of ions and substrates to the catalytic site occurs through pores/channels/tunnels, which is essential for catalytic activity [70]. In CaMLS1, the cofactor and substrate binding sites for GOXL are buried inside. The investigation was focused on how the presence of Mg²⁺ and substrates can influence the opening and closing of tunnels to the active site for facilitating the catalytic activity. This was done through a detailed analysis of the FEL structures of CaMLS1 and its various complexes, using both visual inspection (Fig. 9) and tunnel detection via CAVER web tool [62] (Fig. S21). In all, two tunnels were observed giving access to the catalytic site. The residues forming the tunnels – A & B are listed in Fig. 9c. The tunnel A was observed at the interface of the TIM-barrel and the C-terminal domain, which directed towards the catalytic site (Fig. 9 a) for CaMLS1. The tunnel entry regions were primarily loops which connected alpha helices. These loop regions were color-coded (as indicated in Fig. 9c) to highlight their pattern across all complexes. To explore this further, the tunnels in the complexes of CaMLS1 with cofactors/substrates were examined. Surprisingly, the tunnel A appeared to be closed in all complexes there by restricting access to the catalytic site. On the other hand, it was observed that tunnel B was open in CaMLS1-Mg, which was closed in CaMLS1. This made the analysis particularly intriguing. Similar to tunnel A, the loop regions responsible for mediating the tunnel B were distinctly colored (Fig. 9b). Tunnel B was formed between three domains of CaMLS1. Notably, these loops appeared to be dynamic across all complexes examined. Additionally, the MMAG complex exhibited closure of two tunnels (tunnels A and B), whereas the MMG and MMA complexes opened the tunnel B. Based on these observations, we hypothesize that the presence of Mg²⁺ results in the closure of the tunnel A. This is supported by Fig. S21b, where a change in the tunnel direction is observed. In contrast, when Mg²⁺ is accompanied by one substrate (GOXL or ACOA), the tunnel B remains wide open. This suggested that both the Mg 2+ and the two substrates (GOXL and ACOA) play a crucial role in enzyme activity, consistent with findings from Okada et al [29]. Cofactors/Substrates can alter the proximity of the active site residues. A structural alignment of CaMLS1 with other complexes revealed a similar trend to that observed in the tunnel analysis. Specifically, a 3D structural alignment between CaMLS1 and other complexes showed a decreasing RMSD trend, from 3.87 Å (CaMLS1) to 2.69 Å (MMAG). Further analysis was carried out on the residues at the catalytic site to assess conformational changes upon binding of cofactors/substrates. Lohman et al reported that ARG276 at the catalytic site forms a salt bridge with ASP352 and GLU356, with a movement of approximately 7Å upon substrate binding in bacterial MLS1 [46]. A similar trend was observed in CaMLS1, CaMLS1-Mg and substrate-bound complexes. The distance between the catalytic site residue ARG284 and GLU365 was measured across all systems, as shown in Fig. 10A. The distance between ARG284 and GLU365 was found to be decreased in MMAG (3.1 Å) compared to CaMLS1 (9.5 Å) (Fig. 10A (e&a)). Likewise, the residues TRP285 and MET338 were found to be at distances of 9.5 Å (CaMLS1) and 7.8 Å (MMAG) (Fig. 10B (a&e)). A similar observation was reported for bacterial MLS1 when the interaction between TRP277 and MET330 was decreased by 1.8 Å. The observations made in this study through computational approaches could give rise to a better understanding of the CaMLS1 enzyme. Though the results were correlated with experimental observations of MLS1 in other species, in vitro studies designed on CaMLS1 could provide more support to this enzyme’s activity in the presence of cofactor and substrate molecules. Conclusion CaMLS1, a non-homologous enzyme of C. albicans , gets activated in the nutrient limiting environment contributing to cause candidiasis infection. In this study, the structural dynamics of C. albicans Malate synthase was investigated in the presence of cofactor (Mg 2+ ) as well as substrates (GOXL and ACOA). Homology modeling approach established a 3D model for CaMLS1 protein comprising TIM-barrel, N-terminal and C-terminal domains. MD simulation of CaMLS1 system revealed that TIM-barrel found to be dynamic when compared with other domains. A decrease in fluctuation was observed at the interacting residues in the TIM-barrel region upon binding with cofactors. MD studies of CaMLS1-substrate complex revealed that presence of both substrates (GOXL and ACOA) showed a significant change in RMSD profile of CaMLS1 backbone as well as TIM-barrel domains when compared to other systems. Analyzing the FEL structures of each system revealed the closure of two tunnels in the presence of both substrates. The above structural and dynamic details of C. albicans CaMLS1 systems provided an insight into the role of substrates in protein conformation, which could aid in designing the small molecules to target the Candidiasis infection. Statements and Declarations Ethical approval and consent to participate Not applicable Consent for publication Not applicable Availability of data and material The data will be provided upon request Competing interests No conflict of interest Funding This research did not receive any specific grant from funding agencies Contributions LLK: Conceptualization, Methodology, Data curation, Visualization, Investigation, Software, Validation, Writing- Original draft preparation, Formal analysis AM: Conceptualization, Supervision, Resources, Writing- Reviewing and Editing, Project administration Acknowledgements LLK is thankful to UGC for providing a non-NET fellowship. We would also like to acknowledge the DST-FIST for funding the acquisition of Schrödinger tool, and to the Department of Biotechnology, Govt. of India, New Delhi, for providing the computational facility at the Department of Bioinformatics, Pondicherry University. References 1. Casadevall A (2018) Fungal Diseases in the 21st Century: The Near and Far Horizons. Pathogens & immunity 3:183–196. https://doi.org/10.20411/pai.v3i2.2492. Denning DW (2024) Global incidence and mortality of severe fungal disease. The Lancet Infectious Diseases 24:e428–e438. https://doi.org/10.1016/S1473-3099(23)00692-83. Oliver JC, Laghi L, Parolin C, Foschi C, Marangoni A, Liberatore A, Dias ALT, Cricca M, Vitali B (2020) Metabolic profiling of Candida clinical isolates of different species and infection sources. Sci Rep 10:16716. https://doi.org/10.1038/s41598-020-73889-14. Lamoth F, Lockhart SR, Berkow EL, Calandra T (2018) Changes in the epidemiological landscape of invasive candidiasis. Journal of Antimicrobial Chemotherapy 73:i4–i13. https://doi.org/10.1093/jac/dkx4445. (2022) WHO Fungal Priority Pathogens List to Guide Research, Development and Public Health Action, 1st ed. World Health Organization, Geneva6. Lopes JP, Lionakis MS (2022) Pathogenesis and virulence of Candida albicans. Virulence 13:89–121. https://doi.org/10.1080/21505594.2021.20199507. Tsai C-S, Lee SS-J, Chen W-C, Tseng C-H, Lee N-Y, Chen P-L, Li M-C, Syue L-S, Lo C-L, Ko W-C, Hung Y-P (2023) COVID-19-associated candidiasis and the emerging concern of Candida auris infections. Journal of Microbiology, Immunology and Infection 56:672–679. https://doi.org/10.1016/j.jmii.2022.12.0028. Shetti A, Gupta I, Charantimath SM (2011) Oral Candidiasis: Aiding in the Diagnosis of HIV—A Case Report. Case Reports in Dentistry 2011:929616. https://doi.org/10.1155/2011/9296169. Erfaninejad M, Zarei Mahmoudabadi A, Maraghi E, Hashemzadeh M, Fatahinia M (2022) Epidemiology, prevalence, and associated factors of oral candidiasis in HIV patients from southwest Iran in post-highly active antiretroviral therapy era. Front Microbiol 13:983348. https://doi.org/10.3389/fmicb.2022.98334810. Mohammadi F, Javaheri MR, Nekoeian S, Dehghan P (2016) Identification of Candida species in the oral cavity of diabetic patients. Current medical mycology 2:1–7. https://doi.org/10.18869/acadpub.cmm.2.2.411. Brown AJP, Brown GD, Netea MG, Gow NAR (2014) Metabolism impacts upon Candida immunogenicity and pathogenicity at multiple levels. Trends in Microbiology 22:614–622. https://doi.org/10.1016/j.tim.2014.07.00112. Barelle CJ, Priest CL, MacCallum DM, Gow NAR, Odds FC, Brown AJP (2006) Niche-specific regulation of central metabolic pathways in a fungal pathogen. Cell Microbiol 8:961–971. https://doi.org/10.1111/j.1462-5822.2005.00676.x13. Rohmer L, Hocquet D, Miller SI (2011) Are pathogenic bacteria just looking for food? Metabolism and microbial pathogenesis. Trends in Microbiology 19:341–348. https://doi.org/10.1016/j.tim.2011.04.00314. Gilbert AS, Wheeler RT, May RC (2015) Fungal Pathogens: Survival and Replication within Macrophages. Cold Spring Harb Perspect Med 5:a019661. https://doi.org/10.1101/cshperspect.a01966115. Lorenz MC, Bender JA, Fink GR (2004) Transcriptional response of Candida albicans upon internalization by macrophages. Eukaryotic Cell 3:1076–1087. https://doi.org/10.1128/EC.3.5.1076-1087.200416. Chew SY, Chee WJY, Than LTL (2019) The glyoxylate cycle and alternative carbon metabolism as metabolic adaptation strategies of Candida glabrata: perspectives from Candida albicans and Saccharomyces cerevisiae. J Biomed Sci 26:52. https://doi.org/10.1186/s12929-019-0546-517. Shareck J, Belhumeur P (2011) Modulation of morphogenesis in Candida albicans by various small molecules. Eukaryotic cell 10:1004–1012. https://doi.org/10.1128/EC.05030-1118. Lorenz MC, Fink GR (2001) The glyoxylate cycle is required for fungal virulence. Nature 412:83–86. https://doi.org/10.1038/3508359419. Wayne LG, Lin KY (1982) Glyoxylate metabolism and adaptation of Mycobacterium tuberculosis to survival under anaerobic conditions. Infect Immun 37:1042–1049. https://doi.org/10.1128/iai.37.3.1042-1049.198220. Rude TH, Toffaletti DL, Cox GM, Perfect JR (2002) Relationship of the glyoxylate pathway to the pathogenesis of Cryptococcus neoformans. Infection and immunity 70:5684–5694. https://doi.org/10.1128/IAI.70.10.5684-5694.200221. Kornberg HL, Beevers H (1957) A mechanism of conversion of fat to carbohydrate in castor beans. Nature 180:35–36. https://doi.org/10.1038/180035a022. Nakazawa M, Minami T, Teramura K, Kumamoto S, Hanato S, Takenaka S, Ueda M, Inui H, Nakano Y, Miyatake K (2005) Molecular characterization of a bifunctional glyoxylate cycle enzyme, malate synthase/isocitrate lyase, in Euglena gracilis. Comparative Biochemistry and Physiology Part B: Biochemistry and Molecular Biology 141:445–452. https://doi.org/10.1016/j.cbpc.2005.05.00623. Yang P, Liu W, Chen Y, Gong AD (2022) Engineering the glyoxylate cycle for chemical bioproduction. Frontiers in Bioengineering and Biotechnology 10:1–16. https://doi.org/10.3389/fbioe.2022.106665124. Kondrashov FA, Koonin EV, Morgunov IG, Finogenova TV, Kondrashova MN (2006) Evolution of glyoxylate cycle enzymes in Metazoa: evidence of multiple horizontal transfer events and pseudogene formation. Biology direct 1:31. https://doi.org/10.1186/1745-6150-1-3125. Anstrom DM, Remington SJ (2006) The product complex of M. tuberculosis malate synthase revisited. Protein Science 15:2002–2007. https://doi.org/10.1110/ps.06230020626. Permyakov EA (2021) Metal Binding Proteins. Encyclopedia 1:261–292. https://doi.org/10.3390/encyclopedia101002427. Blanco, A., & Blanco G (2017) Enzymes are biological catalysts. Medical Biochemistry 153–175. https://doi.org/10.1016/B978-0-12-803550-4/00008-228. Gerwien F, Skrahina V, Kasper L, Hube B, Brunke S (2018) Metals in fungal virulence. FEMS Microbiology Reviews 42:. https://doi.org/10.1093/femsre/fux05029. Okada H, Ueda M, Tanaka A (1986) Purification of peroxisomal malate synthase from alkane-grown Candida tropicalis and some properties of the purified enzyme. Arch Microbiol 144:137–141. https://doi.org/10.1007/BF0041472330. Costa FG, Da Silva Neto BR, Gonçalves RL, Da Silva RA, De Oliveira CMA, Kato L, Dos Santos Freitas C, Giannini MJSM, De Fátima Da Silva J, De Almeida Soares CM, Pereira M (2015) Alkaloids as inhibitors of malate synthase from paracoccidioides spp.: Receptor-ligand interaction-based virtual screening and molecular docking studies, antifungal activity, and the adhesion process. Antimicrobial Agents and Chemotherapy 59:5581–5594. https://doi.org/10.1128/AAC.04711-1431. Wilkins MR, Gasteiger E, Bairoch A, Sanchez JC, Williams KL, Appel RD, Hochstrasser DF (1999) Protein identification and analysis tools in the ExPASy server. Methods in molecular biology (Clifton, NJ) 112:531–552. https://doi.org/10.1385/1-59259-584-7:53132. Paysan-Lafosse T, Blum M, Chuguransky S, Grego T, Pinto BL, Salazar GA, Bileschi ML, Bork P, Bridge A, Colwell L, Gough J, Haft DH, Letunić I, Marchler-Bauer A, Mi H, Natale DA, Orengo CA, Pandurangan AP, Rivoire C, Sigrist CJA, Sillitoe I, Thanki N, Thomas PD, Tosatto SCE, Wu CH, Bateman A (2023) InterPro in 2022. Nucleic acids research 51:D418–D427. https://doi.org/10.1093/nar/gkac99333. Johnson M, Zaretskaya I, Raytselis Y, Merezhuk Y, McGinnis S, Madden TL (2008) NCBI BLAST: a better web interface. Nucleic acids research 36:5–9. https://doi.org/10.1093/nar/gkn20134. Thompson JD, Higgins DG, Gibson TJ (1994) CLUSTAL W: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, position-specific gap penalties and weight matrix choice. Nucleic acids research 22:4673–4680. https://doi.org/10.1093/nar/22.22.467335. (2024) QIAGEN CLC Genomics Workbench 24.036. Minh BQ, Schmidt HA, Chernomor O, Schrempf D, Woodhams MD, Von Haeseler A, Lanfear R, Teeling E (2020) IQ-TREE 2: New Models and Efficient Methods for Phylogenetic Inference in the Genomic Era. Molecular Biology and Evolution 37:1530–1534. https://doi.org/10.1093/molbev/msaa01537. Letunic I, Bork P (2024) Interactive Tree of Life (iTOL) v6: recent updates to the phylogenetic tree display and annotation tool. Nucleic Acids Research 5:W78–W8238. Garnier J, Gibrat JF, Robson B (1996) GOR method for predicting protein secondary structure from amino acid sequence. Methods in Enzymology 266:540–553. https://doi.org/10.1016/s0076-6879(96)66034-039. Buchan DWA, Jones DT (2019) The PSIPRED Protein Analysis Workbench: 20 years on. Nucleic Acids Research 47:W402–W407. https://doi.org/10.1093/nar/gkz29740. Kim DE, Chivian D, Baker D (2004) Protein structure prediction and analysis using the Robetta server. Nucleic Acids Research 32:526–531. https://doi.org/10.1093/nar/gkh46841. Fiser A, Sali A (2003) ModLoop: automated modeling of loops in protein structures. Bioinformatics 19:2500–2501. https://doi.org/10.1093/bioinformatics/btg36242. Colovos C, Yeates TO (1993) Verification of protein structures: Patterns of nonbonded atomic interactions. Protein Science 2:1511–1519. https://doi.org/10.1002/pro.556002091643. Laskowski RA, MacArthur MW, Moss DS, Thornton JM (1993) PROCHECK: a program to check the stereochemical quality of protein structures. J Appl Crystallogr 26:283–291. https://doi.org/10.1107/S002188989200994444. Laskowski RA, Jabłońska J, Pravda L, Vařeková RS, Thornton JM (2018) PDBsum: Structural summaries of PDB entries. Protein science : a publication of the Protein Society 27:129–134. https://doi.org/10.1002/pro.328945. Hastings J, Owen G, Dekker A, Ennis M, Kale N, Muthukrishnan V, Turner S, Swainston N, Mendes P, Steinbeck C (2016) ChEBI in 2016: Improved services and an expanding collection of metabolites. Nucleic acids research 44:D1214–D1219. https://doi.org/10.1093/nar/gkv103146. Lohman JR, Olson AC, Remington SJ (2008) Atomic resolution structures of Escherichia coli and Bacillus anthracis malate synthase A: Comparison with isoform G and implications for structure‐based drug discovery. Protein Science 17:1935–1945. https://doi.org/10.1110/ps.036269.10847. Tian W, Chen C, Lei X, Zhao J, Liang J (2018) CASTp 3.0: computed atlas of surface topography of proteins. Nucleic Acids Research 46:W363–W367. https://doi.org/10.1093/nar/gky47348. Schrödinger, LLC (2024) Schrödinger Release 2024-4: Maestro49. Sastry GM, Adzhigirey M, Day T, Annabhimoju R, Sherman W (2013) Protein and ligand preparation: parameters, protocols, and influence on virtual screening enrichments. Journal of computer-aided molecular design 27:221–234. https://doi.org/10.1007/s10822-013-9644-850. Harder E, Damm W, Maple J, Wu C, Reboul M, Xiang JY, Wang L, Lupyan D, Dahlgren MK, Knight JL, Kaus JW, Cerutti DS, Krilov G, Jorgensen WL, Abel R, Friesner RA (2016) OPLS3: A Force Field Providing Broad Coverage of Drug-like Small Molecules and Proteins. Journal of chemical theory and computation 12:281–296. https://doi.org/10.1021/acs.jctc.5b0086451. Yang Y, Yao K, Repasky MP, Leswing K, Abel R, Shoichet BK, Jerome SV (2021) Efficient Exploration of Chemical Space with Docking and Deep Learning. J Chem Theory Comput 17:7106–7119. https://doi.org/10.1021/acs.jctc.1c0081052. Laskowski, R A., Swindells, M B. (2011) LigPlot+: Multiple ligand-protein interaction diagrams for drug discovery. Journal of Chemical Information and Modeling 51:2778–2786. https://doi.org/10.1021/ci200227u53. Abraham MJ, Murtola T, Schulz R, Páll S, Smith JC, Hess B, Lindahl E (2015) GROMACS: High performance molecular simulations through multi-level parallelism from laptops to supercomputers. SoftwareX 1–2:19–25. https://doi.org/10.1016/j.softx.2015.06.00154. Lin Z, Van Gunsteren WF (2013) Refinement of the application of the GROMOS 54A7 force field to β-peptides. Journal of Computational Chemistry 34:2796–2805. https://doi.org/10.1002/jcc.2345955. Malde AK, Zuo L, Breeze M, Stroet M, Poger D, Nair PC, Oostenbrink C, Mark AE (2011) An Automated Force Field Topology Builder (ATB) and Repository: Version 1.0. Journal of Chemical Theory and Computation 7:4026–4037. https://doi.org/10.1021/ct200196m56. Bussi G, Donadio D, Parrinello M (2007) Canonical sampling through velocity rescaling. The Journal of Chemical Physics 126:014101. https://doi.org/10.1063/1.240842057. Nosé S, Klein ML (1983) Constant pressure molecular dynamics for molecular systems. Molecular Physics 50:1055–1076. https://doi.org/10.1080/0026897830010285158. Kawata M, Nagashima U (2001) Particle mesh Ewald method for three-dimensional systems with two-dimensional periodicity. Chemical Physics Letters 340:165–172. https://doi.org/10.1016/S0009-2614(01)00393-159. Ricardo O. S. S (2014) readHBmap: A python script for calculating the H-bond occupancies60. Pettersen EF, Goddard TD, Huang CC, Couch GS, Greenblatt DM, Meng EC, Ferrin TE (2004) UCSF Chimera—A visualization system for exploratory research and analysis. Journal of Computational Chemistry 25:1605–1612. https://doi.org/10.1002/jcc.2008461. Schrödinger LLC (2015) The PyMOL Molecular Graphics System, Version 2.462. Marques SM, Borko S, Vavra O, Dvorsky J, Kohout P, Kabourek P, Hejtmanek L, Damborsky J, Bednar D (2025) Caver Web 2.0: analysis of tunnels and ligand transport in dynamic ensembles of proteins. Nucleic Acids Research gkaf399. https://doi.org/10.1093/nar/gkaf39963. Maisuradze GG, Leitner DM (2007) Free energy landscape of a biomolecule in dihedral principal component space: Sampling convergence and correspondence between structures and minima. Proteins 67:569–578. https://doi.org/10.1002/prot.2134464. Humphrey W, Dalke A, Schulten K (1996) VMD: visual molecular dynamics. Journal of molecular graphics 14:33–38. https://doi.org/10.1016/0263-7855(96)00018-565. Grant BJ, Rodrigues APC, ElSawy KM, McCammon JA, Caves LSD (2006) Bio3d: an R package for the comparative analysis of protein structures. Bioinformatics (Oxford, England) 22:2695–2696. https://doi.org/10.1093/bioinformatics/btl46166. RStudio Team (2023) R: A language and environment for statistical computing67. Kumari R, Kumar R, Lynn A (2014) g_mmpbsa–a GROMACS tool for high-throughput MM-PBSA calculations. Journal of chemical information and modeling 54:1951–1962. https://doi.org/10.1021/ci500020m68. Kyte J, Doolittle RF (1982) A simple method for displaying the hydropathic character of a protein. Journal of Molecular Biology 157:105–132. https://doi.org/10.1016/0022-2836(82)90515-069. Ansari MA, Fatima Z, Ahmad K, Hameed S (2018) Monoterpenoid perillyl alcohol impairs metabolic flexibility of Candida albicans by inhibiting glyoxylate cycle. Biochemical and Biophysical Research Communications 495:560–566. https://doi.org/10.1016/j.bbrc.2017.11.06470. Kingsley LJ, Lill MA (2015) Substrate tunnels in enzymes: Structure–function relationships and computational methodology. Proteins 83:599–611. https://doi.org/10.1002/prot.24772 Figure Legends Fig. 1 Phylogenetic analysis, secondary and tertiary structure prediction of Candida albicans Malate synthase (CaMLS1). (a) Phylogenetic analysis of Malate synthase showing evolutional relationship among pathogenic organisms. The clades represented in red were pathogenic fungal species, while clades in purple were pathogenic bacterial species. The blue circle in the phylogenetic tree represented bootstrap values. (b) Comparative analysis of secondary structure composition of CaMLS1 predicted by PSIPRED (in orange), GOR (in blue) and PDBSum result of CaMLS1 modeled (in green). (c) Modeled 3D structure of CaMLS1. The colour scheme was indicated in the graphical representation of CaMLS1. The color scheme used in the graphical representation of CaMLS1 were as follows. Red: N-terminal; Yellow: TIM-barrel; Cyan: C-terminal . (d) The superimposition of modeled CaMLS1 structure (orange) with the template structure (PDB ID: 3CUX) (blue) resulted in an RMSD of 0.65 Å, as calculated by PyMOL. Fig. 2 Post MD analysis of CaMLS1 protein . (a) Representative RMSD of CaMLS1 protein backbone. (b) RMSD plots of different domains of CaMLS1: N-terminal (red); TIM-barrel (orange); C-terminal (cyan). (c) RMSF graph of Cα residues for CaMLS1. The schematic representation of individual domains was also shown for their easy identification. The major peaks in RMSF plot were numbered consecutively. (d) The B-factor representation was shown for CaMLS1 protein; the highly flexible regions were shown with thick surfaces in B-factor representation. The regions with major peaks (represented with numbers in panel c) were seen with greater thickness in B-factor representation (indicated by numbered arrows). (e) Free energy landscape (FEL) and corresponding minima structure of CaMLS1 protein. The FEL structure colored according to the secondary structure elements (helices in orange, sheets in purple and coils in gray). Fig. 3 Incorporation of Mg 2+ ions into CaMLS1 and MD analysis. (a) Flowchart representation of CaMLS1 docked with Mg 2+ metal ion with three different cases: CaMLS1 with two Mg 2+ ions (CaMLS1-Mg shown in purple); CaMLS1 with one Mg 2+ ion at the catalytic site (CaMLS1-Mg-1 shown in Cyan); CaMLS1 with one Mg 2+ ion on the surface (CaMLS1-Mg-2 shown in Green). All CaMLS1 models were shown in cartoon representation. A close-up view of interactions of Mg 2+ at each binding site was shown in the inset. A graphical representation of CaMLS1 showing the location of Mg 2+ ion in each of these three cases was shown at the bottom of panel a. (b) RMSD profiles of protein backbone of CaMLS1-Mg (purple), CaMLS1-Mg-1 (cyan), and CaMLS1-Mg-2 (green). (c) RMSD profile of TIM-barrel in CaMLS1(black), CaMLS1-Mg (purple), CaMLS1-Mg-1 (cyan), and CaMLS1-Mg-2 (green). Fig. 4 RMS fluctuations of CaMLS1 in presence of Mg 2+ . The RMSF plot of Cα residues for (a) CaMLS1-Mg, (b) CaMLS1-Mg-1, and (c) CaMLS1-Mg-2. The B-factor representations were shown for (d) CaMLS1-Mg, (e) CaMLS1-Mg-1, and (f) CaMLS1-Mg-2; The degree of fluctuations was reflected in thickness in B-factor representation. The regions with major peaks in RMSF plots (represented with numbers) were seen with greater thickness in B-factor representation (indicated by arrows). Fig. 5 Substrates docking studies and 3D & 2D interactions. (A) (a) 3D representation of CaMLS1 docked with GOXL and ACOA. Close up views of interacting residues of CaMLS1 with (b) GOXL and (c) ACOA substrates (in green sticks) and Mg 2+ (in green sphere). 2D views of (d) CaMLS1-GOXL and (e) CaMLS1-ACOA interactions at the active site. (B) (a) 3D representation of CaMLS1 docked with Glyoxylate (GOXL). (b) Close up view of interacting residues of MLS1 with GOXL substrate (in blue sticks) and Mg 2+ (in green sphere) (c) 2D view of CaMLS1-GOXL interactions. (C) (a) 3D representation of CaMLS1 docked with Acetyl-CoA (ACOA). (b) Close up view of interacting residues of CaMLS1 with ACOA substrate (in orange sticks). (c) 2D view of CaMLS1-ACOA interactions in the active site. The H-bond interactions with CaMLS1 residues were shown with a purple arrow, while the amino acids were colored according to their properties. Fig. 6 MD simulation studies of CaMLS1 with substrates. (a) RMSD profile of protein backbone of CaMLS1-Mg-GOXL (MMG) (in blue), CaMLS1-Mg-ACOA (MMA) (in orange), and CaMLS1-Mg-GOXL-ACOA (MMAG) (in dark green). (b) RMSD of substrates (GOXL (blue), ACOA (orange)) in MMG and MMA complexes. Shown also is the RMSD profile of GOXL (green), and ACOA (dark green) in MMAG complex. (c) RMSD plot of TIM-barrel domain in MMG (in blue), MMA (in orange), and MMAG (in dark green). The RMSF plots of Cα residues for (d) MMAG, (e) MMG and (f) MMA. The B-factor representations for (g) MMAG, (h) MMG, and (i) MMA; The degree of mobility was reflected in thickness in B-factor representation. It is noticeable that the regions with major peaks (represented with numbers) in RMSF plots were seen with greater thickness in B-factor representation (indicated by arrows). Fig. 7 H-bond analysis, percentage of occupancy plot and 2D interactions. The hydrogen bond plot between CaMLS1 and substrates. (a) GOXL in MMG complex, (b) ACOA in MMA complex, (c) GOXL in MMAG complex and (d) ACOA in MMAG complex formed during 100 ns of MD simulation. Bar plots showing the hydrogen bonds occupancy (>20 %) formed between substrates and CaMLS1 for (e) MMG, (f) MMA, (g) GOXL in MMAG complex and (h) ACOA in MMAG complex. H-bonds with occupancy ≥20% were considered for these plots. The 2D interaction patterns (H-bonds and hydrophobic interactions) of substrates with CaMLS1 for (i) MMG, (j) MMA, (k) GOXL in MMAG complex and (l) ACOA in MMAG complex at 100 ns time stamp. Fig. 8 Analysis of correlated and uncorrelated motions. A comparative dynamical cross correlation matrix (DCCM) of Cα atoms of (a) CaMLS1, (b) CaMLS1-Mg, (c) CaMLS1-Mg-1, (d) CaMLS1-Mg-2, (e) MMAG, (f) MMG, and (g) MMA complexes. The schematic representation of individual domains was shown for quick identification. The scale to read the degree correlation was also included. In each correlation map, the upper triangle represented positive correlations (shades of blue), the lower triangle represented negative correlations (shades of red) and no correlations in yellow. The boxes represented the CaMLS1 domain regions. For proper understanding of different regions, the correlation between N-terminal and C-terminal (long dash), TIM-barrel and C-terminal (solid line), TIM-barrel and N-terminal (dashed line) were boxed individually. The correlations between N-terminal and N-terminal (dotted box), TIM-barrel and TIM-barrel (dots and dashes), C-terminal and C-terminal (dot dot dash lines), were also shown along the diagonal. For correlation analysis, the linker region between the TIM-barrel and C-terminal is shown with a solid red circle. Fig. 9 Tunnel dynamics of CaMLS1 in presence of cofactor and substrates. (a) Surface view representation of CaMLS1 complexes focussing on the tunnel A, and (b) tunnel B. Zoomed view of CaMLS1 in each complex focussing the tunnel at same site. (c) CaMLS1 system displaying the tunnel A and tunnel B through rotation along the vertical axis. The regions coloured on the protein surface are loop regions in the CaMLS1 protein. Fig. 10 Distance between the active site residues. Distance between (A) ARG284 (in red sticks) and GLU365 (in blue sticks) and (B) TRP285 (in red sticks) and MET338 (in yellow sticks) was measured. The distances for both (A) and (B) were represented for (a) CaMLS1, (b) CaMLS1-Mg, (c) MMG, (d) MMA, and (e) MMAG for both panels A and B. Information & Authors Information Version history V1 Version 1 28 June 2025 Copyright This work is licensed under a Non Exclusive No Reuse License. Keywords candida albicans candidiasis cofactor and substrates glyoxylate pathway homology modeling malate synthase molecular dynamics simulations Authors Affiliations Lukkani Laxman Kumar 0000-0003-1168-2708 Pondicherry University Department of Bioinformatics View all articles by this author Ayaluru Murali 0000-0001-6406-6840 [email protected] Pondicherry University Department of Bioinformatics View all articles by this author Metrics & Citations Metrics Article Usage 316 views 202 downloads .FvxKWukQNSOunydq8rnd { width: 100px; } Citations Download citation Lukkani Laxman Kumar, Ayaluru Murali. Computational insights into Candida albicans Malate synthase: Impact of cofactor and substrates on enzyme conformation and tunnel formation. Authorea . 28 June 2025. DOI: https://doi.org/10.22541/au.175115182.24846590/v1 If you have the appropriate software installed, you can download article citation data to the citation manager of your choice. Simply select your manager software from the list below and click Download. For more information or tips please see 'Downloading to a citation manager' in the Help menu . Format Please select one from the list RIS (ProCite, Reference Manager) EndNote BibTex Medlars RefWorks Direct import Tips for downloading citations document.getElementById('citMgrHelpLink').addEventListener('click', function() { popupHelp(this.href); return false; }); $(".js__slcInclude").on("change", function(e){ if ($(this).val() == 'refworks') $('#direct').prop("checked", false); $('#direct').prop("disabled", ($(this).val() == 'refworks')); }); View Options View options PDF View PDF Figures Tables Media Share Share Share article link Copy Link Copied! Copying failed. Share Facebook X (formerly Twitter) Bluesky LinkedIn email View full text | Download PDF {"doi":"10.22541/au.175115182.24846590/v1","type":"Article"} Now Reading: Share Figures Tables Close figure viewer Back to article Figure title goes here Change zoom level Go to figure location within the article Download figure Toggle share panel Toggle share panel Share Toggle information panel Toggle information panel Go to previous graphic Go to next graphic Go to previous table Go to next table All figures All tables View all material View all material xrefBack.goTo xrefBack.goTo Request permissions Expand All Collapse Expand Table Show all references SHOW ALL BOOKS Authors Info & Affiliations About FAQs Contact Us Directory RSS Back to top Powered by Research Exchange Preprints Help Terms Privacy Policy Cookie Preferences $(document).ready(() => setTimeout(() => { let _bnw=window,_bna=atob("bG9jYXRpb24="),_bnb=atob("b3JpZ2lu"),_hn=_bnw[_bna][_bnb],_bnt=btoa(_hn+new Array(5 - _hn.length % 4).join(" ")); $.get("/resource/lodash?t="+_bnt); },4000)); (function(){function c(){var b=a.contentDocument||a.contentWindow.document;if(b){var d=b.createElement('script');d.innerHTML="window.__CF$cv$params={r:'9feff4ce4c63df94',t:'MTc3OTMyODQ5OA=='};var a=document.createElement('script');a.src='/cdn-cgi/challenge-platform/scripts/jsd/main.js';document.getElementsByTagName('head')[0].appendChild(a);";b.getElementsByTagName('head')[0].appendChild(d)}}if(document.body){var a=document.createElement('iframe');a.height=1;a.width=1;a.style.position='absolute';a.style.top=0;a.style.left=0;a.style.border='none';a.style.visibility='hidden';document.body.appendChild(a);if('loading'!==document.readyState)c();else if(window.addEventListener)document.addEventListener('DOMContentLoaded',c);else{var e=document.onreadystatechange||function(){};document.onreadystatechange=function(b){e(b);'loading'!==document.readyState&&(document.onreadystatechange=e,c())}}}})();