The emergence of drug resistance is a major concern for combating against Cutaneous Leishmaniasis, a neglected tropical disease affecting 98 countries including India. Miltefosine is the only oral drug available for the disease and Miltefosine transporter proteins play a pivotal role in the emergence of drug-resistant Leishmania major. The cause of resistance is less accumulation of drug inside the parasite either by less uptake of the drug due to a decrease in the activity of P4ATPase–CDC50 complex or by increased efflux of the drug by P-glycoprotein (P-gp, an ABC transporter). In this paper, we are trying to allosterically modulate the behavior of resistant parasite (L. major) towards its sensitivity for the existing drug (Miltefosine, a phosphatidylcholine analog). We have used computational approaches to deal with the conservedness of the proteins and apparently its three-dimensional structure prediction through ab initio modeling. Long scale membrane-embedded molecular dynamics simulations were carried out to study the structural interaction and stability. Parasite-specific motifs of these proteins were identified based on the machine learning technique, against which a peptide library was designed. The protein–peptide docking shows good binding energy of peptides Pg5F, Pg8F and PC2 with specific binding to the motifs. These peptides were tested both in vitro and in vivo, where Pg5F in combination with PC2 showed 50–60% inhibition in resistant L. major's promastigote and amastigote forms and 80–90% decrease in parasite load in mice. We posit a model system wherein the data provide sufficient impetus for being novel therapeutics in order to counteract the drug resistance phenotype in Leishmania parasites.

Introduction

Leishmaniasis, the second most neglected tropical disease, is caused by the protozoan parasite Leishmania and spreads through Phlebotomus or Lutzyomia sand-fly. India is amongst those 98 countries of the world that have been declared endemic for Cutaneous, Mucocutaneous and Visceral leishmaniasis [1,2]. Of these, the most affected countries of South East Asia Region are Bangladesh, India, Nepal and Bhutan. In India, most affected states are Bihar, Jharkhand, West Bengal and Uttar Pradesh. India is one amongst the six countries which share 90% of the global burden of leishmaniasis. It has been estimated that ∼600 000 to 1 million of new cases and ∼30 000 deaths occur annually. These figures are expected to increase every year [3–5] and the severity of the disease is further enhanced from HIV co-infection [6,7]. Cutaneous Leishmaniasis is the most common form causing an open sore at the site of bite leaving an unpleasant scar. The causative agents are Leishmania major, L. tropica, L. mexicana and L. amazonensis [8].

Lipids play an important role in all types of cells by constituting the main component of all biological membranes. Apart from acting as the structural component of membranes, they help in anchoring proteins and glycoconjugates to the membrane, source and storage of energy, signaling and trafficking of molecules within and between cells [9]. Lipid trafficking within a cell may be mediated either through vesicles or non-vesicular mechanisms. Maximum transport between organelles occurs by vesicular mechanism. But non-vesicular transport also plays an important role in lipid trafficking and involves three different mechanisms: monomeric lipid exchange, lateral diffusion and transbilayer flip–flop. Transbilayer flip–flop mechanism depends on flippases and floppases for the transport. Flippases are a class of primary transporters, mainly P-type, which uses the energy of ATP hydrolysis in transporting their substrates towards the inner leaflet of lipid bilayer (Figure 1a). The type 4 subfamily of P-type ATPase superfamily, i.e. P4ATPase are involved in the flipping of phospholipids. P4ATPase works in the heterodimeric form with a β subunit CDC50. Structurally, P4ATPase have been categorized in four parts, membrane domain (M) consisting of 10 transmembrane helices, nucleotide-binding domain (N), phosphorylating domain (P) and actuator domain (A) while CDC50 consists of two transmembrane helices and a large exo-cytoplasmic loop. This exo-cytoplasmic loop of CDC50 acts as a lid thereby holding the substrate in position so that it is in-fluxed inside the membrane [10,11]. Floppases functions opposite to that of flippases by transporting its substrate from inner to outer leaflet of the lipid bilayer. ABC (ATP-binding cassette) exporters family acts as floppases. A large number of ABC exporters have been identified of which subfamily B member 4 (ABCB4 or multi-drug resistance protein 1, MDR1 or P-glycoprotein, P-gp) acts as phospholipid transporter in L. major. P-gp is a transmembrane protein consisting of two transmembrane domains (TMDs) each with six transmembrane helices and two nucleotide-binding domains (NBDs) arranged in a head to tail fashion. As per the ATP switch model, it is believed that the binding of two ATP molecules leads to the dimerization of NBDs and subsequently change in structural conformation in open-outward form. This rearrangement is followed by the efflux of the substrate. The open-inside form is achieved by the hydrolysis of either or both ATPs [12,13].

The transporter proteins involved and their behavior in the transport of drug, Miltefosine.

Figure 1.
The transporter proteins involved and their behavior in the transport of drug, Miltefosine.

(a) General mechanism of flippase and floppase proteins. (b) The general assumption: where drug is being in-fluxed by P4ATPase–CDC50 complex and effluxed by P-gp. (c) Miltefosine resistant condition: Less amount of drug accumulates inside the parasite and the maximum amount is effluxed out due to decrease in P4ATPase–CDC50 complex activity and an increase in P-gp activity. (d) Peptide–Miltefosine combination treatment condition: The maximum amount of drug retains inside the parasite and the minimum amount is effluxed out on the combinatorial treatment of the drug with small peptides.

Figure 1.
The transporter proteins involved and their behavior in the transport of drug, Miltefosine.

(a) General mechanism of flippase and floppase proteins. (b) The general assumption: where drug is being in-fluxed by P4ATPase–CDC50 complex and effluxed by P-gp. (c) Miltefosine resistant condition: Less amount of drug accumulates inside the parasite and the maximum amount is effluxed out due to decrease in P4ATPase–CDC50 complex activity and an increase in P-gp activity. (d) Peptide–Miltefosine combination treatment condition: The maximum amount of drug retains inside the parasite and the minimum amount is effluxed out on the combinatorial treatment of the drug with small peptides.

Lipids also play a major role in the survival of intracellular pathogens. They use lipids as food, building blocks, in evading host immune response, manipulating lipid composition and its availability in the host, autophagy, protection against host-derived reactive oxygen species, etc. and thereby causing the disease. This is true for the protozoan parasite Leishmania too [9,14]. Many lipid targeting drugs available for leishmaniasis are Miltefosine, edelfosine, amphotericin B, Antimonials, Paromomycin, etc. Of these, Miltefosine (hexadecylphosphocholine, a lipid analog) is the only oral drug available since 2002 [15,16]. The exact mechanism of action of Miltefosine is still unknown but is believed to cause antileishmanial activity by immuno-modulatory effects on the host cell (macrophage). Within the parasite, Miltefosine acts by manipulating lipid metabolism and its composition (as phosphatidylcholine constitutes 30–40% of the phospholipids on the cell membrane) or DNA fragmentation or decrease in cytochrome c oxidase [17]. Due to irregular and inappropriate use of the drug in the endemic regions, the parasite is developing resistance against Miltefosine not only in laboratory conditions [15] but relapse is seen in many patients [18–20]. Other factors responsible for the spread of resistance are the dose and frequency of the drug administered, duration of the infection and its relapse, fitness of the parasite as well as the host and its immune response [21]. Studies on various species of Leishmania have suggested that the resistance is due to less accumulation of the drug Miltefosine either due to decrease in the activity of the protein responsible for the intake of the drug, P4ATPase along with its beta subunit, which may be a result of mutation; or by an increase in efflux of the drug by P-gp due to increase in its activity or overexpression of transporter proteins [16,21,22]. The involvement of P-gp leads to the possibility of the emergence of multi-drug resistance (MDR). Although in case of leishmaniasis no such cases of cross resistance have yet been reported [23,24] but the possibility of its emergence can also not be ruled out.

Since the past two decades, researchers have been focusing on various ways to combat drug resistance and MDR. Various approaches have been considered and tested like modification of drug, alteration of drug targets, combinations of drugs, increment in affinity as well as specificity, etc. However, every effort has been limited to in vitro conditions only mainly because of the side effects caused by inhibiting transporter proteins [25,26]. There have been a few success reports for the reversal of MDR in case of cancer-targeting ABC efflux proteins [27,28]. Recently, a new approach, being used for reversal of resistance is based on the concept of allostericity. The advantage of targeting allosteric sites is that they offer unique binding sites with higher specificity specially in case of evolutionarily conserved proteins, less toxicity and side effects [29]. Allosteric modulators have been designed to cope with neuronal diseases as well as oncogenesis [30–32] but to the best of our knowledge, no successful work has yet been reported in case of leishmaniasis.

Our present work explores the concept of allostericity of Miltefosine transporter proteins in L. major. This work is an extension of our previous work [33] in consideration with conservedness and protein–membrane and protein–protein interactions at the atomistic level, further extended via long scale molecular dynamic (MD) simulations. Based on these, we aim to allosterically modulate Miltefosine transporter proteins (P-gp and P4ATPase–CDC50) with the help of small peptides to reduce the drug efflux and increase its uptake so that more amount of drug retains inside the parasite and hence may cause reversal in resistance (Figure 1b).

Materials and methods

Sequence retrieval and conservedness analysis

Sequences for P-gp, P4ATPase and CDC50 proteins were extracted manually from the National Centre for Biotechnology Information (NCBI) in FASTA format. All the annotated sequences available in the database ranging from archaea to mammalia were retrieved. Multiple sequence alignment (MSA) for each protein was done using Clustal Omega [34]. Phylogenetic Tree was constructed through MrBayes v3.2.6 [35] using Metropolis coupled Markov chain Monte Carlo (MCMCMC) method. The analysis was terminated when the average standard deviation of split frequency reached below 0.01. The constructed trees were visualized using FigTree v1.4.2 [36]. For identifying position specific and correlated conservedness amongst amino acids, statistical coupling analysis was done through SCA 5.0 [37] tool box in MATLAB R2015a, The Mathworks, Inc.

Structure prediction and validation

Three-dimensional (3D) structures of P-gp, P4ATPase and CDC50 proteins were modeled using MODELLER 9.14 [38] as well as Robetta Server [39]. P4ATPase and CDC50 protein–protein complex docking were done using HADDOCK2.2 Web Server [40]. The structures so obtained were viewed and analyzed using PyMOL v1.7.4.5 (Delano Scientific) and validated through RAMPAGE [41] and PDBsum [42].

Molecular dynamics simulation

The stability and sustainability of the modeled structures were checked through MD simulations (without and with membrane bilayer insertion) using DESMOND 3.2 (D.E. Shaw Research) [43]. TIP3P explicit water type was used as solvent model with an orthorhombic box shape. The ions were neutralized with Na+ ions. Nose–Hoover chain thermostat and Martyna–Tobias–Klein barostat methods were implemented for short scale MD simulations of up to 20 ns time scale. Long-range MD simulations of 300 ns and 500 ns with membrane bilayer insertion (DPPC, dipalmitoylphosphotidylcholine) were done using Berendsen thermostat and barostat method. MD simulation was done with NPT ensemble class at 300 K temperature and 1 bar pressure. Post MD simulation, root mean square deviations (RMSD), root mean square fluctuations (RMSF), energies and other parameters of the trajectories were calculated and the corresponding graphs were obtained with simulation event analysis and simulation interaction diagrams using default parameters.

Motif identification and peptide designing

Motifs unique for L. major's P-gp and P4ATPase–CDC50 proteins were identified by means of MEME suite v4.11.2 [44]. A motif identification criterion includes any number of repetition sites of motifs in the given set of input sequences. The number of motifs that should be found was set as 6 for P-gp and 10 for P4ATPase and CDC50 proteins. The motifs with a minimum length of six residues to a maximum length of 20 residues were to be identified. Motifs shortlisting was done based on their less similarity with the host protein, their secondary structure and the amino acids found in SCA sectors. For these sets of shortlisted motifs, peptides were designed based on physicochemical properties using the dead end elimination algorithm, a deterministic search approach used for designing novel sequences. A peptide library consisting of 12–14 amino acid residue lengths was constructed considering overlapping combinations. 3D structure for these peptides was predicted via PEPstr [45]. Peptide filtering criteria include geometric and chemical complementarity, stearic hindrance, structural stability, hydropathy index (accounting for hydrophobic nature of the lipid membrane), net charges, binding affinity, specific interactions, etc. From the constructed peptide library, a total of six peptides were shortlisted for P-gp and two peptides for P4ATPase–CDC50 complex.

Molecular docking

The protein–peptide docking was accomplished by AutoDock Vina [46] and analysis was done using MGL Tools 1.5.6. [47]. Rigid docking with site directed as well as blind docking was performed. The genetic algorithm with an empirical scoring type was selected for docking through Autodock. The Gasteiger charges were computed and applied on both receptor and ligand before preparing the ‘pdbqt’ files. The dimensions of grid box for P-gp were 122 × 98 × 126 with x, y and z center, respectively, as −0.01, −4.217, −9.056 and that for P4ATPase–CDC50 complex were 126 × 126 × 86 with −0.234, 0.136 and −0.151 as x, y and z center, respectively. The protein–peptide conformations with the lowest binding energy and specific binding were selected. The docked complexes were viewed in PyMOL v1.7.4.5 (Delano Scientific) and the residual interactions were mapped using Ligplot+ [48].

Strains and culture conditions

L. major promastigotes (MHOM/IL/67/JERICHOII) were maintained at 27°C in RPMI-1640 media supplemented with 20% heat-inactivated fetal bovine serum (Sigma/Gibco). All experiments were performed with parasite cultures in the late logarithmic phase of growth. DMEM medium supplemented with 10% heat-inactivated fetal bovine serum (Sigma/Gibco) was used for the murine macrophage cell line RAW264.7. The macrophages were maintained at 37°C with 5% CO2 humidity.

Miltefosine resistant strain development

For the development of Miltefosine resistant strains of L. major, parasites were exposed to low dosage of Miltefosine starting from 3 µM concentration. When the parasite got adapted at 3 µM (i.e. growth pattern similar to that of wild type), the concentration was gradually increased to 5, 7, 10, 15, 20, 30, 40 and 50 µM. At each concentration parasite took at least 10–12 passages to adapt and show growth as that of wild type promastigotes. Cell viability of adapted strains (M5 → 5 μM; M7 → 7 μM and so on) at their respective Miltefosine concentrations were checked with and without drug pressure using MTT assay. IC50 for Miltefosine was also calculated for all the adaptive strains.

IC50 calculation

Promastigotes

MTT colorimetric assay was used to calculate the IC50 of Miltefosine for wild type (W) as well as resistant strains (M5, M7, M10, M20, M30, M40 and M50). Promastigotes (1 × 106 cells/ml) were plated in 96-well flat-bottomed microtiter plates in triplicates with desired concentrations of Miltefosine. Cells were incubated for 48 h and centrifuged. Pellet is then resuspended in 180 µl 1×PBS and 20 µl of MTT solution (5 mg/ml) and incubated again in dark for 4 h. Formazan crystals so formed were dissolved in 100 µl DMSO and absorbance in each well was measured at 570 nm (MULTISKAN Sky, Thermo Scientific). Percentage inhibition was then calculated: ((Control – Test)/Control) × 100 and IC50 was then calculated using GraphPad Prism 5 for Windows, GraphPad Software, La Jolla California, U.S.A. (www.graphpad.com).

Amastigotes

In case of amastigotes, the IC50 was calculated on the basis of parasite clearance seen through DAPI staining. In brief, the RAW264.7 cells were seeded at concentration of 5 × 105 cells/ml in eight-well chamber slides. The cells were kept for 3 h so that it would adhere. The macrophages were then infected for 24 h with L. major promastigotes in the stationary phase (wild type and resistant) with the parasite to cell ratio of 10 : 1. The infected cells were then treated with Miltefosine (1.75, 3.5, 7, 14 and 28 μM concentration) for 48 h. After incubation, cells were fixed with 3.7% formaldehyde, permeabalized with 0.1% TritonX and stained with DAPI (1 μg/ml). The number of intracellular amastigotes was counted for 100 randomly selected infected macrophages by visualizing slides at 40× EVOS FL microscope (Life Technology). The inhibition in treated macrophages was expressed in form of percentage as compared with the control and IC50 was then calculated using GraphPad Prism 5 for Windows, GraphPad Software, La Jolla California U.S.A. (www.graphpad.com).

Cytotoxicity assay

Cytotoxic effects of Miltefosine, as well as peptides, were tested on host cells, i.e. RAW264.7 through MTT dye. 1 × 105 cells/ml would be plated in 96-well flat-bottomed microtiter plates in triplicates. After 3 h of incubation for adherence, the media was changed with the treatment of choice and the cells were incubated for 48 h. The rest method for MTT assay was done as mentioned above. To check the toxic effect of Miltefosine on host, the concentrations used were 12, 14, 16 and 18 μM. In case of peptides, the concentrations considered were 50, 100 and 125 μg/ml.

Antileishmanial activity assay

The peptides were tested directly on wild type and resistant promastigotes in logarithmic growth phase in the presence and the absence of drug, i.e. Miltefosine. In case of amastigotes, the antileishmanial activity was observed on the basis of parasite clearance seen through DAPI staining as mentioned above. The infected and untreated macrophages were considered as control. The Miltefosine was administered at IC50 dose and peptides at 50, 100 and 125 μg/ml concentrations.

Flow cytometry

Flow cytometry analysis with Annexin V-FITC and PI (propidium iodide) was used to quantify the percentage of healthy, apoptotic and necrotic cells in wild type and resistant L. major promastigotes. Parasites (1 × 106 cells/ml) were treated with Miltefosine at 50 μM concentration and incubated for 48 h. After centrifugation and washing the cells in PBS, the resultant pellet was resuspended in 300 μl of Annexin binding buffer. Cells were then stained with Annexin V-FITC and PI and incubated in the dark for 15 min. Data were then acquired using FACS Calibur, BD and analyzed using the CELLQUEST PRO software.

SEM analysis

In total 1 × 106 cells/ml of L. major promastigotes (wild type and resistant) were seeded in six well plates in the presence of desired treatment and incubated for 48 h. The cells were then harvested and washed twice with 1×PBS and smeared on poly l-lysine coated silicon wafer. These smeared cells were fixed with 2.5% glutaraldehyde solution for 3–5 h at 27°C followed by PBS wash and dehydration with a series of increasing concentration of acetone from 30%, 50%, 70% to 100%. The sample was then sputter coated with gold for 35 s. FEI Nova NanoSEM 450 was used for field emission scanning electron microscopy. Scanning electron microscopy (SEM) analysis was done at Central Instrumentation Facility, S.P. Pune University.

Cell permeation assay

In total 1 × 106 cells/ml of L. major promastigotes (resistant) were seeded in six well plates in the presence of peptides (Pg5F: 125 μg/ml and PC2: 50 μg/ml) and Miltefosine (13.63 μM) and were incubated for 48 h. The cells were then harvested and washed twice with 1×PBS. The parasites were smeared on a glass slide and fixed and permeabalized with chilled methanol for 1 min. The fixed cells were stained with Giemsa stain for 15 min and mounted with glycerol. The stained slides were then viewed under Evos FL Microscope at 100× magnification. Thio-Pyridine modification of Cysteine residue was done for peptide Pg5F with the removal of 1 Cys residue for increasing efficacy of Pg5F while PC2 was tagged with FITC at N′ terminal. Pg5F: AW{CYS(THIO-PYRIDINE)}DSATNSEI; PC2: FITC-Ahx-LESASVNADATL.

Animal maintenance and drug dosage

Balb/c mice were maintained at the Experimental Animal Facility at NCCS. 6–8 weeks old female mice with 18–20 g weight were used for the study. L. major promastigotes at the stationary phase (maintained in vitro as mentioned above; 2 × 107 cells in 40 μl PBS) were injected subcutaneously in hind limb footpad of mice. The treatment was started on reaching lesion size of 3 mm thickness. The mice were divided into eight groups (four mice per group [49]): infected (I), treated-Miltefosine (T-M), treated-peptide (T-P), treated-peptide + Miltefosine (T-PM) for wild type (W) and resistant (R) strains, respectively. Miltefosine was administered orally at 45 mg/kg body weight for 6 days a week and peptides (Pg5F:125 μg/ml and PC2: 50 μg/ml) were administered intravenously once a week. Mice were anesthetized with ether for intravenous administration of peptides. The duration of treatment was 28 days and then mice were monitored further for 14 days. Lesion size was measured every week post infection using a Vernier caliper. Parasite load in the draining lymph node was estimated post euthanization by limiting dilution assay for each of the treated and untreated groups as per the Taswell's method [50].

Statistical analysis

All the experiments for IC50 and viability calculations were performed in triplicates and those for peptide effect were done in duplicates. Data are represented as mean ± SD from three independent experiments. Unpaired Student t-test and one way ANOVA with non-parametric Tukey's post-test was used for p-value calculations as per their applicability. All the data analysis was performed with the help of GraphPad Prism 5 for Windows, GraphPad Software, La Jolla California U.S.A., (www.graphpad.com).

Animal ethics

All animal experiments were approved by the Institutional Animal Ethics Committee (IAEC) (7/GO/c/99/CPCSEA) of the National Centre for Cell Science (IAEC Project Number-EAF/2016/B-260 and EAF/2019/B-260(I)).

Results

Sequence retrieval and conservedness analysis

In total 96, 207 and 189 sequences were extracted, respectively, for P-gp, P4ATPase and CDC50. The phylogenetic tree, thus constructed, shows that these proteins are closely related within the members of the Trypanosomatidae family. The clades of the Leishmania genus are colored in red (Figure 2a). Statistical coupling analysis (SCA) [37] was performed to find positional conservation and correlated amino acid residues. The sequences were grouped into ‘sectors’ of co-evolving amino acid positions (Figure 2b; Supplementary Figure S1) shown in the form of the heat map, where the red color represents higher conservedness. List of sequences retrieved along with their Accession IDs and organism names as well as the sectors identified for P-gp, P4ATPase and CDC50 are provided as Supplementary Files S1 and S2, respectively. For identifying sectors, positions with less than 20% gap frequency were considered. 179, 128 and 110 conserved positions of amino acid residues were identified, respectively, for P-gp, P4ATPase and CDC50. The statistical coupling analysis clearly depicts that a large number of residues are co-evolved and highly conserved for these Miltefosine transporter proteins.

Conservedness analysis and structure prediction of Miltefosine transporter proteins.

Figure 2.
Conservedness analysis and structure prediction of Miltefosine transporter proteins.

(a) Phylogenetic tree and (b) SCA matrix after spectral cleaning for P-gp, P4ATPase and CDC50 proteins. The taxa's belonging to the Leishmania genus are colored red in the phylogenetic tree. (c) Three-dimensional structures (embedded in DPPC lipid membrane bilayer) and RMSD and RMSF graphs post large scale molecular dynamics simulation of P-gp and P4ATPase–CDC50 complex.

Figure 2.
Conservedness analysis and structure prediction of Miltefosine transporter proteins.

(a) Phylogenetic tree and (b) SCA matrix after spectral cleaning for P-gp, P4ATPase and CDC50 proteins. The taxa's belonging to the Leishmania genus are colored red in the phylogenetic tree. (c) Three-dimensional structures (embedded in DPPC lipid membrane bilayer) and RMSD and RMSF graphs post large scale molecular dynamics simulation of P-gp and P4ATPase–CDC50 complex.

3D structure modeling, validation and MD simulation

3D structures modeled for P-gp (Accession ID: CAJ07806), P4ATPase (Accession ID: CAJ03162), CDC50 (Accession ID: CAJ08536) and P4ATPase–CDC50 complex are shown in Figure 2c and Supplementary Figure S3. Large scale MD simulation with DPPC membrane bilayer insertion of 300 ns for P-gp, 100 ns for P4ATPase, 50 ns for CDC50 and 500 ns for P4ATPase–CDC50 complex, respectively, were performed. The RMSD graphs represent that P-gp structure is stable after 200 ns with 5 Å deviation while P4ATPase–CDC50 complex with 10 Å deviation post 300 ns simulation time. The movie files of the MD simulation trajectories are attached as Supplementary Data.

Motif identification, peptide designing and screening

For motif identification, the clade of leishmanial species as highlighted in the phylogenetic tree, were selected. The identification was based on the criteria mentioned in the methodology section employing machine learning MEME suite [44]. MEME (multiple expectation-maximization for Motif Elicitation) is a unique software tool that uses artificial intelligence techniques to discover motifs shared by a set of protein sequences in a fully automated manner. We have used the MEME suite to analyze biologically relevant set of sequences of Leishmanial species, and to evaluate the sensitivity and accuracy of MEME in identifying structurally important motifs for transporter proteins. The motifs obtained for P-gp, P4ATPase and CDC50 proteins of L. major are represented in logo format as well as mapped on 3D structures in Figure 3a. The 6, 10 and 10 consensus motifs were identified in leishmanial species, respectively, for P-gp, P4ATPase and CDC50 (represented in Alphabetical Logo format). These consensus motifs constitute a total 29, 30 and 19 motifs, respectively, in P-gp, P4ATPase and CDC50 proteins of L. major (mapped on the 3D structure). Out of these obtained set of motifs, nine motifs (Figure 3b) were shortlisted based on their secondary structure, their domain location, least sequence similarity with respective human homologs as well as SCA sectors.

. Motif Identification and shortlisting.

Figure 3
. Motif Identification and shortlisting.

(a) The consensus motifs identified through MEME suite are shown in logo format as well as mapped on 3D structures of P-gp and P4ATPase–CDC50 proteins. (b) Shortlisted motifs for P-gp, P4ATPase and CDC50 along with their sequence, start site and 3D structures.

Figure 3
. Motif Identification and shortlisting.

(a) The consensus motifs identified through MEME suite are shown in logo format as well as mapped on 3D structures of P-gp and P4ATPase–CDC50 proteins. (b) Shortlisted motifs for P-gp, P4ATPase and CDC50 along with their sequence, start site and 3D structures.

Based on the physiochemical properties of the motifs, a library of 60 peptides were designed against P-gp motifs and 43 peptides against P4ATPase–CDC50 complex (Supplementary File S3). The tertiary structure of these peptides was predicted via PEPstr [45] (Supplementary Figures S3–S11). Site directed as well as blind docking of these peptides with P-gp and P4ATPase–CDC50 complex was then performed. Based on minimum binding energy conformation with 0.0 RMSD value for each peptide and their specific binding with proteins, number of interactions, stearic hindrance as well as reverse docking; a total of 6 and 2 peptides in number were shortlisted for P-gp and P4ATPase–CDC50 complex, respectively. The shortlisted peptides with their sequence, binding energies and number of interactions with their respective motifs are denoted in Table 1. The docked poses of protein and peptide complexes and protein–peptide interactions are shown in Supplementary Figures S12 and S13. Furthermore, the protein–peptide complexes were then docked with ATP and Miltefosine (Figure 4; Supplementary Figure S14) and short scale MD Simulations without membrane bilayer of 20 ns was performed to visualize the extent of interactions maintained between protein and Miltefosine in the presence of various peptides (Tables 2 and 3). Their respective RMSDs along with interacting residues plot for Pg5F, Pg8F and PC2 peptides are depicted in Figure 5 and Supplementary Figure S15. It was observed that in the presence of peptide Pg5F, the number of interacting residues between P-gp and Miltefosine reduced comparatively while the interaction of Miltefosine with P4ATPase–CDC50 complex increased in the presence of peptide PC2. The Pgp–ATP–MIL complex showed 31 interactions between protein and Miltefosine which decreased to 21 and 26 in the presence of peptide Pg5F and Pg8F, respectively. On the other hand, during MD simulations, interactions between Miltefosine and P4ATPase–CDC50 complex increased from 29 to 40 in the presence of peptide PC2. The changes were also observed in protein–peptide interactions during short MD simulation of 20 ns. In case of the Pgp–ATP–MIL–Pg5F complex, during MD simulation, interactions increased from 21 to 63 in total with 10 specific to motif f. In case of peptide PC2, the rise was from 22 to 39 interactions with 11 being specific to motif i. N538, D954, E962, D978, R979 and N538, R549, D954, T955, T961, A963, D978, R979, R993 are the residues maintaining more than two types of interactions (H-bonds/ionic/water bridges/hydrophobic) with the peptides Pg5F and Pg8F, respectively, throughout the simulation chemical time. The motif specific interactions and those contacts that were maintained throughout 20 ns chemical time are shown in Figure 6. Moreover, the trajectory and RMSD graphs suggest that those interactions were maintained with a lesser intensity suggesting that Pg5F might be blocking the efflux of Miltefosine. In case of PC2, the interactions between Miltefosine and P4ATPase were maintained for almost 20 ns as visualized in the RMSD plot, suggesting more influx of drug in the presence of the peptide. These short scale MD simulations indicate Pg5F, Pg8F and PC2 as potential peptides that can allosterically modulate functionality of transporter proteins. These peptides were procured from Sigma–Aldrich with >90% purity levels (HPLC and LC–MS graphs shown in Supplementary Figure S16).

Protein–peptide complex docking.

Figure 4.
Protein–peptide complex docking.

Protein complex with peptide, ATP and MIL shown along with surface electrostatic potentials. The inset figure shows the binding region of the peptide with protein. Protein is represented in magenta cartoon form, Miltefosine in orange spheres, ATP in yellow line and stick form and peptide in green line and stick form. The cartoon form in cyan color represents the motif against which peptide has been designed.

Figure 4.
Protein–peptide complex docking.

Protein complex with peptide, ATP and MIL shown along with surface electrostatic potentials. The inset figure shows the binding region of the peptide with protein. Protein is represented in magenta cartoon form, Miltefosine in orange spheres, ATP in yellow line and stick form and peptide in green line and stick form. The cartoon form in cyan color represents the motif against which peptide has been designed.

Protein–peptide complex MD simulation.

Figure 5.
Protein–peptide complex MD simulation.

RMSD and protein–ligand interaction graphs obtained after short scale molecular dynamics simulation of 20 ns. Protein RMSD plot is colored blue while the ligand RMSD plot is in red. In the interactions fractions bar graph different color representations, green: H-bonds, cyan: hydrophobic, pink: ionic and blue: water bridges.

Figure 5.
Protein–peptide complex MD simulation.

RMSD and protein–ligand interaction graphs obtained after short scale molecular dynamics simulation of 20 ns. Protein RMSD plot is colored blue while the ligand RMSD plot is in red. In the interactions fractions bar graph different color representations, green: H-bonds, cyan: hydrophobic, pink: ionic and blue: water bridges.

Motif specific protein–peptide interactions during MD simulation.

Figure 6.
Motif specific protein–peptide interactions during MD simulation.

(a) Pgp–ATP–MIL–Pg5F complex. (b) Pgp–ATP–MIL–Pg8F complex. (c) P4ATPase–CDC50–ATP–MIL–PC2 complex.

Figure 6.
Motif specific protein–peptide interactions during MD simulation.

(a) Pgp–ATP–MIL–Pg5F complex. (b) Pgp–ATP–MIL–Pg8F complex. (c) P4ATPase–CDC50–ATP–MIL–PC2 complex.

Table 1.
Shortlisted peptides along with their sequence ID, sequence, binding energy and specific number of interactions with their respective motifs
IDSequenceBE (Kcal/mol)# Interactions (bonded and non-bonded)
Pg1A DNLDDGVNLVHE −7.7 
Pg6A VLRPKDNLEDGV −7.4 
Pg8A VRPKDNLEDVLNLV −7.4 
Pg5F ACWCDSATNSEI −7.0 
Pg8F VNHACECDSASNSD −7.4 
Pg10F VNHACWCDSITNTE −6.8 
PC1 NVVNVDEDLSINIH −6.8 
PC2 LESASVNADATL −6.7 
IDSequenceBE (Kcal/mol)# Interactions (bonded and non-bonded)
Pg1A DNLDDGVNLVHE −7.7 
Pg6A VLRPKDNLEDGV −7.4 
Pg8A VRPKDNLEDVLNLV −7.4 
Pg5F ACWCDSATNSEI −7.0 
Pg8F VNHACECDSASNSD −7.4 
Pg10F VNHACWCDSITNTE −6.8 
PC1 NVVNVDEDLSINIH −6.8 
PC2 LESASVNADATL −6.7 
Table 2.
Interactions between protein and peptides before and during molecular dynamics simulation
ComplexTotal interactionsSpecific interactions
Pgp–5F 20 1 (V967) 
Pgp–5F–ATP–MIL (pre MD) 21 5 (T961, E962, S965, M975, D978) 
Pgp–5F–ATP–MIL (during MD) 63 10 (I959, V960, T961, E962, L964, S965, S966, N974, M975, D978) 
Pgp–8F 28 5 (T961, E962, S965, V967, R968) 
Pgp–8F–ATP–MIL (pre MD) 24 7 (I959, V960, E962, A963, N974, M975, D978) 
Pgp–8F–ATP–MIL (during MD) 43 8 (I959, V960, T961, E962, A963, N974, M975, D978) 
P4ATPase–CDC50–PC2 21 6* (R138, G139, S140, S143, T144, L145) 
P4ATPase–CDC50–PC2–ATP–MIL (pre MD) 22 6* (R138, G139, S140, S143, T144, L145) 
P4ATPase–CDC50–PC2–ATP–MIL (during MD) 39 11* (V133,S136,G137, R138, G139, S140, A141, G142, S143, L145, C146) 
ComplexTotal interactionsSpecific interactions
Pgp–5F 20 1 (V967) 
Pgp–5F–ATP–MIL (pre MD) 21 5 (T961, E962, S965, M975, D978) 
Pgp–5F–ATP–MIL (during MD) 63 10 (I959, V960, T961, E962, L964, S965, S966, N974, M975, D978) 
Pgp–8F 28 5 (T961, E962, S965, V967, R968) 
Pgp–8F–ATP–MIL (pre MD) 24 7 (I959, V960, E962, A963, N974, M975, D978) 
Pgp–8F–ATP–MIL (during MD) 43 8 (I959, V960, T961, E962, A963, N974, M975, D978) 
P4ATPase–CDC50–PC2 21 6* (R138, G139, S140, S143, T144, L145) 
P4ATPase–CDC50–PC2–ATP–MIL (pre MD) 22 6* (R138, G139, S140, S143, T144, L145) 
P4ATPase–CDC50–PC2–ATP–MIL (during MD) 39 11* (V133,S136,G137, R138, G139, S140, A141, G142, S143, L145, C146) 
*

The residues belong to chain B, i.e. CDC50 protein.

Table 3.
Interactions between protein and Miltefosine before and during molecular dynamics simulation
Interactions with Miltefosine
Pgp–ATP–MIL (during MD) 31 
Pgp–ATP–MIL–5F (pre MD) 
Pgp–ATP–MIL–5F (during MD) 21 
Pgp–ATP–MIL–8F (pre MD) 
Pgp–ATP–MIL–8F (during MD) 26 
P4ATPase–CDC50–ATP–MIL (during MD) 29 
P4ATPase–CDC50–ATP–MIL–PC2 (pre MD) 
P4ATPase–CDC50–ATP–MIL–PC2 (during MD) 40 
Interactions with Miltefosine
Pgp–ATP–MIL (during MD) 31 
Pgp–ATP–MIL–5F (pre MD) 
Pgp–ATP–MIL–5F (during MD) 21 
Pgp–ATP–MIL–8F (pre MD) 
Pgp–ATP–MIL–8F (during MD) 26 
P4ATPase–CDC50–ATP–MIL (during MD) 29 
P4ATPase–CDC50–ATP–MIL–PC2 (pre MD) 
P4ATPase–CDC50–ATP–MIL–PC2 (during MD) 40 

Cytotoxicity and IC50 calculation

The IC50 of Miltefosine for L. major wild type promastigotes was obtained to be 13.63 μM (Figure 7a) when administered for 48 h, which is in accordance with the literature [51]. Similarly, IC50 for various resistant strains viz. M5 to M50 was also calculated and their viability (Figure 7d,e) was seen in the presence and the absence of drug pressure in order to ensure the true adaptiveness of cells at their respective drug concentrations. As expected, the IC50 for M50 (final resistant strain, mentioned further as ‘R’) was much higher (113.33 μM) when compared with that of wild type. In addition to these, the effect of Miltefosine at 13.63 μM concentration shows ∼86% viability in host macrophages i.e. RAW264.7 (Figure 7c). The Giemsa stained images depict that there is no morphological difference between wild type and resistant promastigotes (Figure 9e,h). For wild type amastigotes, the IC50 of Miltefosine was found to be 7.685 μM (Figure 7b) and at this concentration, the resistant strain exhibited <10% inhibition (Figure 7f). The FACS analysis of Miltefosine resistant strain with and without drug pressure exhibits similar behavior as that of wild type. The resistant strain also indicated a significant decrease in apoptotic death in the presence of 50 μM concentration of Miltefosine (Figure 7g,h).

Miltefosine Resistant L. major strain development and IC50 calculations.

Figure 7.
Miltefosine Resistant L. major strain development and IC50 calculations.

(a) Miltefosine IC50 for wild type (W) L. major promastigotes. (b) Miltefosine IC50 for wild type (W) L. major amastigotes. (c) Viability of Macrophage at Miltefosine IC50 value (13.63 μM). (d) Viability graph of different intermediate resistant strains. (e) IC50 graph of different intermediate resistant strains. (f) Effect of Miltefosine on amastigotes (at IC50 value, 7.685 μM) in wild type and resistant strains. (g,h) FACS analysis of wild type and resistant strain in the absence and the presence of Miltefosine at 50 μM concentration. Unpaired Student t-test is used and the p-value is compared with respective groups of ‘W’ with *: p < 0.05, **: p < 0.01, ***: p < 0.001. W: wild type strain; W.M: wild type treated with Miltefosine; R: resistant strain; R.M: resistant treated with Miltefosine.

Figure 7.
Miltefosine Resistant L. major strain development and IC50 calculations.

(a) Miltefosine IC50 for wild type (W) L. major promastigotes. (b) Miltefosine IC50 for wild type (W) L. major amastigotes. (c) Viability of Macrophage at Miltefosine IC50 value (13.63 μM). (d) Viability graph of different intermediate resistant strains. (e) IC50 graph of different intermediate resistant strains. (f) Effect of Miltefosine on amastigotes (at IC50 value, 7.685 μM) in wild type and resistant strains. (g,h) FACS analysis of wild type and resistant strain in the absence and the presence of Miltefosine at 50 μM concentration. Unpaired Student t-test is used and the p-value is compared with respective groups of ‘W’ with *: p < 0.05, **: p < 0.01, ***: p < 0.001. W: wild type strain; W.M: wild type treated with Miltefosine; R: resistant strain; R.M: resistant treated with Miltefosine.

Effect of peptides

In vitro. The shortlisted computationally designed peptides were commercially synthesized and procured from Sigma–Aldrich Inc. These peptides were tested directly on a murine macrophage cell line, RAW264.7, L. major promastigotes and amastigotes (Figure 8). Pg5F and Pg8F peptides were tested at 100 μg/ml and 125 μg/ml concentrations; PC2 at 50 μg/ml and 100 μg/ml and Miltefosine at the IC50 concentrations for promastigotes and amastigotes, respectively. In case of macrophages, all peptide combinations, as well as concentrations, showed the viability of more than 60% except for Pg8F (125 μg/ml) when administered in combination with PC2 (100 μg/ml). It was found to be toxic on macrophages with less than 60% viability. In case of promastigotes, the mixed effect of peptides was observed in inhibition with an increase in concentration and also when administered in combination. It was observed that Pg5F (125 μg/ml) in combination with PC2 (50 μg/ml) and Miltefosine is more effective on resistant promastigotes with nearly 50% inhibition. As for inhibition in amastigotes, it was seen in terms of parasite clearance in infected macrophages upon treatment with peptides and Miltefosine. A pattern of concentration dependent increase in inhibition was observed for Pg5F, Pg8F, PC2 and Pg5F + PC2 except for Pg8F (125 μg/ml) in combination with PC2 (100 μg/ml). This deviation in the pattern of concentration-dependent inhibition ascertained that Pg8F and PC2 combination might be toxic at higher concentrations since the macrophage viability was also below 60%. Pg5F (125 μg/ml) with PC2 (100 μg/ml) showed >60% inhibition on resistant amastigotes in the presence as well as the absence of Miltefosine. Morphological changes on wild type and resistant promastigotes upon administration of Miltefosine at IC50 and peptide Pg5F (125 μg/ml) and PC2 (50 μg/ml) have been seen through Giemsa staining as well as SEM (Figure 9c–j). The peptide treated promastigotes showed similar morphology as that of wild type promastigote in the presence of Miltefosine, reflecting that as per our hypothesis, the presence of peptide is able to retain Miltefosine inside parasite through allosteric modulation of transporter proteins P-gp and P4ATPase–CDC50 complex. Furthermore, FACS analysis showed a significant increase in apoptotic deaths in wild type as well as resistant promastigotes treated with peptide Pg5F (125 μg/ml) and PC2 (50 μg/ml) in the presence of Miltefosine indicating towards the reversal of resistant strain towards sensitive. The cell permeation assay done with fluorescence tagged peptides Pg5F and PC2 further confirms that both the peptides not only interacts with the cell membrane of parasites but penetrates the membrane also (Figure 10). This data corroborates with our in silico simulation studies pin-pointing allosterisitic modulation of the transporter proteins in L. major.

Effect of peptide along with Miltefosine on macrophages, wild type as well as Miltefosine resistant promastigotes and amastigotes.

Figure 8.
Effect of peptide along with Miltefosine on macrophages, wild type as well as Miltefosine resistant promastigotes and amastigotes.

The left panel represents the viability of macrophages upon peptide administration. Centre and right panels show % inhibition in the growth of promastigotes and macrophage infected amastigotes, respectively, on peptide administration. Here term untreated refers to the absence of peptide. One way ANOVA with Tukey's post-test is used and p-values here are represented with respect to each group of ‘untreated’ with *: p < 0.05, **: p < 0.01, ***: p < 0.001. W: wild type strain; W.M: wild type treated with Miltefosine; R: resistant strain; R.M: resistant treated with Miltefosine.

Figure 8.
Effect of peptide along with Miltefosine on macrophages, wild type as well as Miltefosine resistant promastigotes and amastigotes.

The left panel represents the viability of macrophages upon peptide administration. Centre and right panels show % inhibition in the growth of promastigotes and macrophage infected amastigotes, respectively, on peptide administration. Here term untreated refers to the absence of peptide. One way ANOVA with Tukey's post-test is used and p-values here are represented with respect to each group of ‘untreated’ with *: p < 0.05, **: p < 0.01, ***: p < 0.001. W: wild type strain; W.M: wild type treated with Miltefosine; R: resistant strain; R.M: resistant treated with Miltefosine.

Effect of peptides in presence of Miltefosine on L. major promastigotes morphology.

Figure 9.
Effect of peptides in presence of Miltefosine on L. major promastigotes morphology.

(a,b) FACS analysis of wild type and resistant strain in the absence and the presence of Miltefosine (M) at 13.63 μM (IC50 concentration) and peptide (P) Pg5F (125 μg/ml) and PC2 (50 μg/ml) concentrations. (c,d) Images from SEM. (ej) Giemsa stained images at 40× using EVOS FLc Microscope, Life technologies. One way ANOVA with Tukey's post-test is used and p-values here are represented with respect to each group of ‘W’ with *: p < 0.05, **: p < 0.01, ***: p < 0.001. W: wild type strain; W.M: wild type treated with Miltefosine; W.M.P: wild type treated with Miltefosine and peptide: Pg5F + PC2; R: resistant strain; R.M: resistant treated with Miltefosine; R.M.P: resistant treated with Miltefosine and peptide: Pg5F + PC2.

Figure 9.
Effect of peptides in presence of Miltefosine on L. major promastigotes morphology.

(a,b) FACS analysis of wild type and resistant strain in the absence and the presence of Miltefosine (M) at 13.63 μM (IC50 concentration) and peptide (P) Pg5F (125 μg/ml) and PC2 (50 μg/ml) concentrations. (c,d) Images from SEM. (ej) Giemsa stained images at 40× using EVOS FLc Microscope, Life technologies. One way ANOVA with Tukey's post-test is used and p-values here are represented with respect to each group of ‘W’ with *: p < 0.05, **: p < 0.01, ***: p < 0.001. W: wild type strain; W.M: wild type treated with Miltefosine; W.M.P: wild type treated with Miltefosine and peptide: Pg5F + PC2; R: resistant strain; R.M: resistant treated with Miltefosine; R.M.P: resistant treated with Miltefosine and peptide: Pg5F + PC2.

Cell permeation assay of fluorescence tagged peptides Pg5F and PC2.

Figure 10.
Cell permeation assay of fluorescence tagged peptides Pg5F and PC2.

The 100× image of resistant L. major promastigote treated with peptide Pg5F (a Thio-pyridine modification of Cysteine residue); PC2 (b FITC tagged at N′ terminal) and Miltefosine. (c) The overlay image of red channel (Pg5F) and green channel (PC2).

Figure 10.
Cell permeation assay of fluorescence tagged peptides Pg5F and PC2.

The 100× image of resistant L. major promastigote treated with peptide Pg5F (a Thio-pyridine modification of Cysteine residue); PC2 (b FITC tagged at N′ terminal) and Miltefosine. (c) The overlay image of red channel (Pg5F) and green channel (PC2).

In vivo. From the collective analysis of the effect of peptides on macrophages, promastigotes and amastigotes; Pg5F at 125 μg/ml and PC2 at 50 μg/ml were selected to be administered on Balb/c female mice. The peptides were administered intravenously in the tail vein once a week while Miltefosine was given orally 6 days a week. The treatment was started when wound size reached 3 mm thickness, continued for 28 days + 14 days of observation. The foot pad size measured every week post treatment (Supplementary Figure S20; Figure 11a) shows that the mice treated with Miltefosine in the presence of peptide, group RT-PM, showed similar effect as that of mice of WT-M group. Groups WT-P and WT-PM showed slight improvement in footpad size compared with WR-M. In terms of parasite load, a significant decrease was found in peptide treated groups of mice. While WT-M showed 69% decrease in parasite load as compared with WI; 5% and 6% increase in parasite load was seen in RI and RT-M groups, respectively. The peptide decreased the parasite load by 80–90% either in the absence or presence of the drug, Miltefosine, irrespective of being infected with wild type or resistant L. major strain. The wound size and parasite load data are represented in Figure 11.

Effect of peptides on Mice.

Figure 11.
Effect of peptides on Mice.

(a) Wound size measured weekly post treatment (b) Parasite load of the treated groups. WI: wild type infected; WT-M: wild type infected and treated with Miltefosine; WT-P: wild type infected and treated with peptides; WT-PM: wild type infected and treated with peptides and Miltefosine; RI: resistant infected; RT-M: resistant infected and treated with Miltefosine; RT-P: resistant infected and treated with peptides; RT-PM: resistant infected and treated with peptides and Miltefosine. One way ANOVA with Tukey's post-test is used and p-values here are represented with respect to WI; ***: p < 0.001.

Figure 11.
Effect of peptides on Mice.

(a) Wound size measured weekly post treatment (b) Parasite load of the treated groups. WI: wild type infected; WT-M: wild type infected and treated with Miltefosine; WT-P: wild type infected and treated with peptides; WT-PM: wild type infected and treated with peptides and Miltefosine; RI: resistant infected; RT-M: resistant infected and treated with Miltefosine; RT-P: resistant infected and treated with peptides; RT-PM: resistant infected and treated with peptides and Miltefosine. One way ANOVA with Tukey's post-test is used and p-values here are represented with respect to WI; ***: p < 0.001.

Discussion

P-gp and P4ATPase–CDC50 proteins are highly conserved at the amino acid sequence level as well as at their positional level. The initial clues of topology are being provided by the conservation analysis which helped in the identification of the functionally relevant conformational changes. These proteins are an intrinsic part of the mechanism resulting in the development of resistance in L. major against Miltefosine, a phosphatidylcholine analog. Utmost care have been taken in identifying the functionally and structurally important motifs as silent polymorphism can lead to conformational changes affecting the substrate specificity and thereby its insertion into the membrane. The mechanistic view adopted in relation to structural rearrangement results from the intra- and inter-molecular forces that govern the substrate influx and efflux. This has also caused profound effect on protein–membrane and protein–protein interactions. Through our study, many a detailed aspects have been elucidated in the structural information of these transporter proteins which encodes for functional differences accounting selectivity, sensitivity, specificity, response to substrates and the designed set of peptides serving as modulators. The mechanistic models of transport, as discussed in this paper, and their varied associated conformational changes over a range of time period has allowed the allostericity being imparted at the atomistic level. Allosteric conformational transitions as observed through MD simulations have helped in laying a framework to design and interpret the experimental results. In combination with the experimental approaches, in silico methodology adopted in this paper helped form a critical insight into the dynamicity and structural elements being determined individually. Furthermore, the approach laid also envisages the study in combination with the peptides binding specificities and their associations with the lipid bilayer system.

Based on the conservedness analysis and extensive MD simulations, the parasite-specific sequential motifs, capable of initiating conformational changes if interfered, were identified. Against these motifs, a library of 103 computationally designed synthetic peptides was created. These peptides were then screened against their respective proteins, i.e. P-gp and P4ATPase–CDC50 complex, for their specific binding and minimum energy conformations. Six peptides (Pg1A, Pg6A, Pg8A, Pg5F, Pg8F and Pg10F) were shortlisted against P-gp and two peptides (PC1 and PC2) against P4ATPase–CDC50 complex. The short scale (20 ns) MD simulation of the protein–peptide complex with ATP and Miltefosine revealed that in the presence of peptide Pg5F and Pg8F, P-gp–Miltefosine interactions have weakened suggesting that these peptides might have caused allosteric changes and thereby decrease in drug efflux. P-gp, an ABC transporter with 2 TMDs and 2 NBDs, drive the transport processes through binding and hydrolysis of ATP accounting for intra- and inter-domain interactions. MD simulations have provided an insight of maintaining the closed form of NBD dimer through ATP binding. In addition, the interaction between P-gp and Pg5F is subtle at the hinge of TMD1 and NBD1, which determines the overall conformational changes. Similarly, the exo-cytoplasmic loop of CDC50 in P4ATPase–CDC50 complex is being targeted by PC2 peptide. These observations obtained from our in silico analysis were in accordance with the results obtained from in vitro and in vivo testing. In in vitro, the peptides in the absence and the presence of Miltefosine were tested directly on wild type and resistant promastigotes as well as macrophage infected amastigotes. The peptides were found to be more effective on the amastigote stage compared with the promastigote stage. In amastigotes, concentration-dependent inhibition was seen on wild type as well as resistant strains with Pg5F in combination with PC2 and Miltefosine being the most effective. This combination was found to lower the wound size to some extent and drastic decrease in parasite load (80–90% decrease) in Balb/c mice infected with either wild type or resistant strains. The results in mice were comparable to those of wild type infected mice when treated with Miltefosine (∼70% decrease in parasite load). In addition, these peptides were tested directly on macrophages also for their toxic effect and Pg8F with PC2 was found to be toxic at higher concentrations. These results suggest that the peptides in combination with Miltefosine were able to reverse the resistant strain into sensitive ones by allosterically modulating Miltefosine transporter proteins. The effect of peptides seen on wild type amastigotes and wild type infected Balb/c indicates the probability of these peptides behaving as synthetic AMPs.

In nutshell, homeostasis and metabolism of the movement of substances across the membrane of these transport proteins are essential with respect to the allostericity and substrate-binding effects on the translocation process. The nature of results obtained with the aforementioned strategies has made it possible to rationally optimize an existing set of peptides/ligands for transporter proteins. Furthermore, in future, it is also possible to develop new set of compounds that may shift the conformational equilibrium of transporters facilitating functional studies and leading to drug discovery. To the best of our knowledge, in this case, the approach and development of peptides targeting the L. major Miltefosine transporters is novel.

Competing Interests

The authors declare that there are no competing interests associated with the manuscript.

Funding

Intramural Funding was used and R.K. was also supported from Department of Biotechnology (DBT/JRF/14/AL/22/3544).

Authors Contribution

R.K. performed in silico, in vitro and in vivo experimentation. P.I. performed in vitro assays related to Miltefosine resistant strain development and also facilitated in vivo testing of peptides. S.S. designed and conceptualized the whole idea, project implementation to its successful execution. All authors participated in manuscript writing.

Acknowledgements

R.K. would like to acknowledge her SRF (Senior Research Fellowship) from the Department of Biotechnology (DBT/JRF/14/AL/22/3544), Ministry of Science and Technology, Government of India. Authors would also like to thank the Director, National Centre for Cell Science (NCCS) Pune, for supporting the Bioinformatics and High-Performance Computing Facility (BHPCF) at NCCS Pune, India. Authors also extend their thanks to FACS Facility, Experimental Animal Facility (EAF) at NCCS, Pune and Central Instrumentation Facility at S.P. Pune University, Pune, India.

Abbreviations

     
  • ABC

    ATP-binding cassette

  •  
  • AMPs

    antimicrobial peptides

  •  
  • CDC50

    cell division control protein 50

  •  
  • DPPC

    dipalmitoylphosphotidylcholine

  •  
  • MCMCMC

    Metropolis coupled Markov chain Monte Carlo

  •  
  • MD

    molecular dynamics

  •  
  • MDR

    multi-drug resistance

  •  
  • NBDs

    nucleotide-binding domains

  •  
  • P4ATPase

    Type 4 P-type ATPase

  •  
  • P-gp

    P-glycoprotein

  •  
  • RMSD

    root mean square deviation

  •  
  • RMSF

    root mean square fluctuation

  •  
  • SCA

    statistical coupling analysis

  •  
  • TMDs

    transmembrane domains

References

References
2
Read
,
A.
,
Hurwitz
,
I.
and
Durvasula
,
R
. (
2013
) Leishmaniasis: an update on a neglected tropical disease. In
Dynamic Models of Infectious Diseases
(Rao, V. and Durvasula, R., eds.),
Springer
,
New York, NY
3
Herwaldt
,
B.L.
(
1999
)
Leishmaniasis
.
Lancet
354
,
1191
1199
5
Leishmaniasis
. [cited 1 Sep 2016]. Available: http://www.who.int/mediacentre/factsheets/fs375/en/
6
Lindoso
,
J.A.L.
,
Cunha
,
M.A.
,
Queiroz
,
I.T.
and
Moreira
,
C.H.V.
(
2016
)
Leishmaniasis–HIV coinfection: current challenges
.
HIV AIDS (Auckl)
8
,
147
156
7
Alvar
,
J.
,
Cañavate
,
C.
,
Gutiérrez-Solar
,
B.
,
Jiménez
,
M.
,
Laguna
,
F.
,
López-Vélez
,
R.
et al (
1997
)
Leishmania and human immunodeficiency virus coinfection: the first 10 years
.
Clin. Microbiol. Rev.
10
,
298
319
8
Singh
,
S.
(
2006
)
New developments in diagnosis of leishmaniasis
.
Indian J. Med. Res.
123
,
311
330
PMID:
[PubMed]
9
Ramakrishnan
,
S.
,
Serricchio
,
M.
,
Striepen
,
B.
and
Bütikofer
,
P.
(
2013
)
Lipid synthesis in protozoan parasites: a comparison between kinetoplastids and apicomplexans
.
Prog. Lipid Res.
52
,
488
512
10
López-Marqués
,
R.L.
,
Poulsen
,
L.R.
,
Bailly
,
A.
,
Geisler
,
M.
,
Pomorski
,
T.G.
and
Palmgren
,
M.G.
(
2015
)
Structure and mechanism of ATP-dependent phospholipid transporters
.
Biochim. Biophys. Acta
1850
,
461
475
11
van der Mark
,
V.A.
,
Oude Elferink
,
R.P.J.
and
Paulusma
,
C.C.
(
2013
)
P4 ATPases: flippases in health and disease
.
Int. J. Mol. Sci.
14
,
7897
7922
12
Dawson
,
R.J.P.
and
Locher
,
K.P.
(
2006
)
Structure of a bacterial multidrug ABC transporter
.
Nature
443
,
180
185
13
Seeger
,
M.A.
and
van Veen
,
H.W.
(
2009
)
Molecular basis of multidrug transport by ABC transporters
.
Biochim. Biophys. Acta
1794
,
725
737
14
Zhang
,
K.
and
Beverley
,
S.M.
(
2010
)
Phospholipid and sphingolipid metabolism in leishmania
.
Mol. Biochem. Parasitol.
170
,
55
64
15
Singh
,
N.
,
Kumar
,
M.
and
Singh
,
R.K.
(
2012
)
Leishmaniasis: current status of available drugs and new potential drug targets
.
Asian Pac. J. Trop. Med.
5
,
485
497
16
Ghorbani
,
M.
and
Farhoudi
,
R.
(
2018
)
Leishmaniasis in humans: drug or vaccine therapy?
Drug Des. Devel. Ther.
12
,
25
40
17
Dorlo
,
T.P.C.
,
Balasegaram
,
M.
,
Beijnen
,
J.H.
and
de vries
,
P.J.
(
2012
)
Miltefosine: a review of its pharmacology and therapeutic efficacy in the treatment of leishmaniasis
.
J. Antimicrob. Chemother.
67
,
2576
2597
18
Tiwary
,
P.
,
Kumar
,
D.
and
Sundar
,
S.
(
2018
)
Identification and functional validation of a biomarker for the diagnosis of miltefosine relapse during visceral leishmaniasis
.
Am. J. Trop. Med. Hyg.
98
,
492
496
19
Veronica
,
J.
,
Chandrasekaran
,
S.
,
Dayakar
,
A.
,
Devender
,
M.
,
Prajapati
,
V.K.
,
Sundar
,
S.
et al (
2019
)
Iron superoxide dismutase contributes to miltefosine resistance in leishmania donovani
.
FEBS J.
286
,
3488
3503
20
Rijal
,
S.
,
Ostyn
,
B.
,
Uranw
,
S.
,
Rai
,
K.
,
Bhattarai
,
N.R.
,
Dorlo
,
T.P.C.
et al (
2013
)
Increasing failure of miltefosine in the treatment of kala-azar in Nepal and the potential role of parasite drug resistance, reinfection, or noncompliance
.
Clin. Infect Dis.
56
,
1530
1538
21
Croft
,
S.L.
,
Sundar
,
S.
and
Fairlamb
,
A.H.
(
2006
)
Drug resistance in leishmaniasis
.
Clin. Microbiol. Rev.
19
,
111
126
22
Pérez-Victoria
,
F.J.
,
Sánchez-Cañete
,
M.P.
,
Seifert
,
K.
,
Croft
,
S.L.
,
Sundar
,
S.
,
Castanys
,
S.
et al (
2006
)
Mechanisms of experimental resistance of leishmania to miltefosine: implications for clinical use
.
Drug Resist. Updat.
9
,
26
39
23
Van den Kerkhof
,
M.
,
Mabille
,
D.
,
Chatelain
,
E.
,
Mowbray
,
C.E.
,
Braillard
,
S.
,
Hendrickx
,
S.
et al (
2018
)
In vitro and in vivo pharmacodynamics of three novel antileishmanial lead series
.
Int. J. Parasitol. Drugs Drug Resist.
8
,
81
86
24
Mishra
,
J.
and
Singh
,
S.
(
2013
)
Miltefosine resistance in leishmania donovani involves suppression of oxidative stress-induced programmed cell death
.
Exp. Parasitol.
135
,
397
406
25
Higgins
,
C.F.
(
2007
)
Multiple molecular mechanisms for multidrug resistance transporters
.
Nature
446
,
749
757
26
Tanwar
,
J.
,
Das
,
S.
,
Fatima
,
Z.
,
Hameed
,
S.
,
Tanwar
,
J.
,
Das
,
S.
et al (
2014
)
Multidrug resistance: an emerging crisis, multidrug resistance: an emerging crisis
.
Interdiscip. Perspect. Infect. Dis.
2014
,
e541340
27
Wang
,
B.
,
Zhao
,
B.
,
Chen
,
Z.S.
,
Pang
,
L.P.
,
Di
,
Z..Y.
,
Guo
,
Q.
et al (
2018
)
Exploration of 1,3-triazole-pyrimidine hybrids as potent reversal agents against ABCB1-mediated multidrug resistance
.
Eur. J. Med. Chem.
143
,
1535
1542
28
Hsiao
,
S.H.
,
Murakami
,
M.
,
Yeh
,
N.
,
Li
,
Y.Q.
,
Hung
,
T.H.
,
Wu
,
Y.S.
et al (
2018
)
The positive inotropic agent DPI-201106 selectively reverses ABCB1-mediated multidrug resistance in cancer cell lines
.
Cancer Lett.
434
,
81
90
29
Astl
,
L.
and
Verkhivker
,
G.M.
(
2019
)
Data-driven computational analysis of allosteric proteins by exploring protein dynamics, residue coevolution and residue interaction networks
.
Biochim. Biophys. Acta Gen. Subj.
30
Guglielmo
,
S.
,
Contino
,
M.
,
Lazzarato
,
L.
,
Perrone
,
M.G.
,
Blangetti
,
M.
,
Fruttero
,
R.
et al (
2016
)
A potent and selective P-gp modulator for altering multidrug resistance due to pump overexpression
.
ChemMedChem
11
,
374
376
31
Beshore
,
D.C.
,
Di Marco
,
C.N.
,
Chang
,
R.K.
,
Greshock
,
T.J.
,
Ma
,
L.
,
Wittmann
,
M.
et al (
2018
)
MK-7622: a first-in-class M1 positive allosteric modulator development candidate
.
ACS Med. Chem. Lett.
7
,
652
656
32
Poslusney
,
M.S.
,
Salovich
,
J.M.
,
Wood
,
M.R.
,
Melancon
,
B.J.
,
Bollinger
,
K.A.
,
Luscombe
,
V.B.
et al (
2019
)
Novel M 4 positive allosteric modulators derived from questioning the role and impact of a presumed intramolecular hydrogen-bonding motif in β-amino carboxamide-harboring ligands
.
Bioorganic. Med. Chem. Lett.
29
,
362
366
33
Singh
,
S.
and
Mandlik
,
V.
(
2015
)
Structure based investigation on the binding interaction of transport proteins in leishmaniasis: insights from molecular simulation
.
Mol. BioSyst.
11
,
1251
1259
34
Sievers
,
F.
,
Wilm
,
A.
,
Dineen
,
D.
,
Gibson
,
T.J.
,
Karplus
,
K.
,
Li
,
W.
et al (
2011
)
Fast, scalable generation of high-quality protein multiple sequence alignments using clustal omega
.
Mol. Syst. Biol.
7
,
539
35
Ronquist
,
F.
,
Teslenko
,
M.
,
Van Der Mark
,
P.
,
Ayres
,
D.L.
,
Darling
,
A.
,
Höhna
,
S.
et al (
2012
)
Mrbayes 3.2: efficient Bayesian phylogenetic inference and model choice across a large model space
.
Syst. Biol.
61
,
539
542
37
Halabi
,
N.
,
Rivoire
,
O.
,
Leibler
,
S.
and
Ranganathan
,
R.
(
2009
)
Protein sectors: evolutionary units of three-dimensional structure
.
Cell
138
,
774
786
38
Sali
,
A.
and
Blundell
,
T.L.
(
1993
)
Comparative protein modelling by satisfaction of spatial restraints
.
J. Mol. Biol.
234
,
779
815
39
Kim
,
D.E.
,
Chivian
,
D.
and
Baker
,
D.
(
2004
)
Protein structure prediction and analysis using the Robetta server
.
Nucleic Acids Res.
32
,
W526
W531
40
van Zundert
,
G.C.P.
,
Rodrigues
,
J.P.G.L.M.
,
Trellet
,
M.
,
Schmitz
,
C.
,
Kastritis
,
P.L.
,
Karaca
,
E.
et al (
2016
)
The HADDOCK2.2 Web server: user-friendly integrative modeling of biomolecular complexes
.
J. Mol. Biol.
428
,
720
725
.
41
Lovell
,
S.C.
,
Davis
,
I.W.
,
Adrendall
,
W.B.
,
de Bakker
,
P.I.W.
,
Word
,
J.M.
,
Prisant
,
M.G.
et al (
2003
)
Structure validation by C alpha geometry: phi, psi and Cbeta deviation
.
Proteins
50
,
437
450
42
Thornton
,
J.M.
(
2018
)
PDBsum : structural summaries of PDB entries
.
Protein Sci.
27
,
129
134
43
Bowers
,
K.J
,
Chow
,
E.
,
Xu
,
H.
,
Dror
,
R.O
,
Eastwood
,
M.P
,
Gregersen
,
B.A.
et al. (
2006
)
Scalable algorithms for molecular dynamics simulations on commodity clusters
. In
Proceedings of the 2006 ACM/IEEE Conference on Supercomputing,
SC'06
,
Association for Computing Machinery
,
New York, NY, United States, Tampa, Florida
44
Bailey
,
T.L.
,
Boden
,
M.
,
Buske
,
F.A.
,
Frith
,
M.
,
Grant
,
C.E.
,
Clementi
,
L.
et al (
2009
)
MEME suite: tools for motif discovery and searching
.
Nucleic Acids Res.
37
,
202
208
45
Kaur
,
H.
,
Garg
,
A.
and
Raghava
,
G.P.S.
(
2007
)
PEPstr: a de novo method for tertiary structure prediction of small bioactive peptides
.
Protein Pept. Lett.
14
,
626
631
46
Trott
,
O.
and
Olson
,
A.J.
(
2010
)
Autodock vina: improving the speed and accuracy of docking with a new scoring function, efficient optimization, and multithreading
.
J. Comput. Chem.
31
,
455
461
47
Morris
,
G.
and
Huey
,
R.
(
2009
)
Autodock4 and AutoDockTools4: automated docking with selective receptor flexibility
.
J. Comput. Chem.
30
,
2785
2791
48
Laskowski
,
R.A.
and
Swindells
,
M.B.
(
2011
)
Ligplot+: multiple ligand-protein interaction diagrams for drug discovery
.
J. Chem. Inf. Model.
51
,
2778
2786
49
Arifin
,
W.N.
and
Zahiruddin
,
W.M.
(
2017
)
Sample size calculation in animal studies using resource equation approach
.
Malays. J. Med. Sci.
24
,
101
105
50
Taswell
,
C.
(
1981
)
Limiting dilution assays for the determination of immunocompetent cell frequencies. I. Data analysis
.
J Immunol.
126
,
1614
1619
PMID:
[PubMed]
51
Escobar
,
P.
,
Matu
,
S.
,
Marques
,
C.
and
Croft
,
S.L.
(
2002
)
Sensitivities of leishmania species to hexadecylphosphocholine (miltefosine), ET-18-OCH3 (edelfosine) and amphotericin B
.
Acta Trop.
81
,
151
157