The sequence of LPG2 protein (341 amino acids) of Leishmania donovani was downloaded for structural modeling from NCBI. Multiple alignments of the related sequences were performed using the online available ClustalW program accessible through the European Bioinformatics Institute (Thompson et al., 1994; http://www.ebi.ac.uk/Tools/ clustalw2/index.html). No X-ray crystallographic or NMR structure of this protein of any Leishmania species has yet been determined. Tertiary structures of LPG2 protein of different Leishmania strains were modeled on the basis of different template structures from PHYRE, I-Tasser. Structure validation was performed using ANOLEA, Profiles-3D, WHATIF, and Model-3D, molecular modeling tool (Profiles- 3D) of discovery studio.

Different servers i.e. TMHMM, SOSUI, HMMTOP and TMpred servers were accessed to validate the TM region of LPG2 protein (Krogh et al., 2001; Hirokawa et al., 1998, TusnaÂdy et al., 1998; Hofmann et al., 1993). TMHMM, a new membrane protein topology prediction method, is based on a hidden Markov model.

Pocket-Finder is a pocket detection algorithm based on Ligsite written by Hendlich et al (1997). Pocket-Finder works by scanning a probe radius 1.6A° along all gridlines of grid resolution 0.9 A° surrounding the protein. The probe also scans cubic diagonals. Grid points are defined to be part of a site when the probe is within range of protein atoms followed by free space followed by protein atoms. Grid points are only retained if they are defined to be part of a site at least five times (Hendlich et al., 1997).

To know novel functions of LPG2 protein of different Leishmania strains were searched at BIDD (Cai et al., 2003; http://jing.cz3.nus.edu.sg/cgi-bin/svmprot.cgi). The web-based software, SVMProt, support vector machine(SVM) classifies a protein into functional families from its primary sequence based on physico-chemical properties of amino acids (Cai et al., 2003). Novel protein function assignment of different proteins of SARS virus and Japanese encephalitis virus has already been reported (Cai et al., 2005; Sahoo et al., 2008).

Functional sites in eukaryotic proteins which fit to the description “linear motif” are currently specified as patterns using regular expression rules. ELM server provides core functionality including filtering by cell compartment, phylogeny, globular domain clash (using the SMART/ Pfam databases) and structure (Puntervoll et al., 2003). Individual functions assigned to different sequence segments combine to create a complex function for the whole protein.

Predict Protein provides PROSITE sequence motifs, lowcomplexity regions (SEG), nuclear localization signals, regions lacking regular structure (NORS) and predictions of secondary structure, solvent accessibility, globular regions, transmembrane helices, coiled-coil regions, structural switch regions, disulfide-bonds, sub-cellular localization, and functional annotations (Puntervoll et al., 2003; Rost et al., 2004; Bairoch et al., 1997; Ceroni et al., 2004).

Different servers have been accessed for accurate prediction analysis of transmembrane region e.g. TMHMM, HMMTOP, SOSUI and TMpred. Trans-membrane prediction analysis found that LPG2 protein of L. infantum and L. donovani are having same number of transmembrane regions (ten) and involvement of particular amino acids in the TM regions is similar. The LPG2 sequences of L. infantum and L. donovani are very similar to each other except at two positions 220 and 221 where threonine (T) is replaced by isoleucine (I) and methionine (M) respectively in L. infantum. Even if the number of transmembrane region are same (ten) in L. mexicana, L. donovani and L. infanum, it is found that 3^rd and 4^th transmembrane regions are different (3^rd TM region comprises of aa 72-94 and 4^th TM comprises 98-120 amino acids) in L. mexicana whereas in L. infantum and L. donovani it is coded by 77-99 and 101- 123 amino acids. LPG2 of L. braziliensis and L. major both comprise of nine TM regions but the transmembrane regions are quite different, only in few cases these are similar. Involvement of different amino acids in formation of nine different trans-membrane regions of different Leishmania strains is shown in (Figure.1).

Multiple alignment of amino acid sequences of LPG2 protein of different Leishmania strains shows that they are very close to each other ranging from 78-99% (Table 1). Multiple alignment of LPG2 protein shows that Ldv and L.infantum are having 99% identity (Figure. 2). From phylogram, LPG2 protein of L. braziliensis is found to be far from other Leishmania strains (Figure.3). Many amino acid changes are found to be present towards the carboxy terminus in the annotated amino acid sequences.

Five different models of LPG2 protein of different Leishmaniastrains were screened for profiles-3d score of DS(Accelrys). The best model of different Leishmania strains were screened by profiles-3d score and the best was selected for further analysis. About 14 - 27 helices have been predicted for LPG2 protein of different models of various Leishmania strains. The best model of LPG2 protein of L. infantum contains 22 helices, whereas 19 helices are there in case of L. donovani and L. major. The models for LPG2 protein of different Leishmania strains are shown in Figure 4. The profiles-3D scores of best predicted models of LPG2 protein of various Leishmania strains (Table 2), shows that highest score (111.29) has been found in case of LPG2 protein model of L. major and lowest score (91.23) has been found in case of L. braziliensis. Invalid regions have been detected in different models. Highest numbers of invalid regions have been found in one model of L. mexicana whereas lowest number of invalid regions is found in other models of L. mexicana, L. major and L. donovani. On further side chain and loop modeling refinements of the predicted models of LPG2 protein, the profiles-3D score is found to be decreased. In Ramachandran plots (Procheck), 88- 91% residues belong to core region, 6-8 % residue in allowed region, 1-2.5% in generously allowed regions and 1- 3% in disallowed regions (Table 3). From Ramachandran plot, it is known that maximum residues in LPG2 protein are responsible for formation of helices. Also transmembrane region prediction analysis detected nine or ten transmembrane regions to be present in LPG2 protein of all these strains. Hence there is involvement of helices in formation of LPG2 protein.

The models of different Leishmania strains were based on template PDB coordinates of 2i68, 1ee4, 1pw4, 1ej1, 3b5d and 1xm9. The PDB ‘2i68’ codes for transmembrane domain of the multi drug resistance antiporter from E. coli Emr E. This protein has antiporter activity and belongs to a family of plasma membrane proteins and proteins integral to membrane. The PDB ‘1pw4’ codes for protein with transporter activity and belongs to MFS general substrate transporter fold and is involved in glycerol metabolic process and glycerol 3-phosphate transport. It is a protein of plasma membrane and protein integral to membrane. The PDB ‘1xm9’ codes for structure of armadillo repeat domain of plakophilin 1, belongs to all alpha proteins with alpha alpha superhelix fold which has ARM repeat forming plakophilin 1 domain . This protein is involved in molecular functions like signal transduction, cell adhesion and protein binding. The PDB ‘3b5d’ codes for x-ray crystallographic structure of Emr E multidrug transporter in complex with TPP, involved in antiporter like transporter function. It is a plasma membrane component. Few drugs have been targeted to this protein like dapsone and tamoxifen. The PDB coordinates of ‘1ee4’ code for crystal structure of yeast karyopherin (importin) in a complex with a c-myc NLS peptide. This karyophilin alpha protein have armadillo domain with ARM repeat. This protein is classified as transporter protein which is a transmembrane protein with transporter activity involved in import into nucleus, protein targeting to membrane and intracellular transport. This is a leucine rich alpha helical protein. The PDB coordinates of ‘1ej1’ code for crystal structure of mRNA 5’cap binding protein (eIF 4E) bound to 7-methyl-GDP. This protein is having RNA binding and translation factor activity and belongs to class of alpha and beta proteins. From all these template analyses, it is found that except ‘1ej1’, all other templates are transporter–like protein. LPG2 protein also belongs to this group of protein.

It is found that the best model of LPG2 protein of all these strains consisted of only one chain (Table 2). In all the models 68-73 percent is helical. The best model of LPG2 protein of L. donovani is having highest number (24) of helices where minimum five and maximum sixteen residues take part in formation of a helix. In L. donovani three 3₁₀ helices are present. In L. mexicana 28 residues has been found to be forming a helix and helices in this strain accounts for 73% of total residues.

From the comparative analysis of LPG2 protein functional assignment of five Leishmania strains shows that it belongs to transporter group of proteins (Table 4). LPG2 protein also belonged to trans-membrane region protein. LPG2 protein of L. mexicana strain M379 belonging to copper binding (58.6%), magnesium binding (58.6%), metal-binding, iron-binding (73.8%) and sodium-binding (78.4 %) protein function families. It also belongs to type II (general) secretory pathway (IISP) family and major facilitator family (MFS) (58.6%). LPG2 protein of L. braziliensis also belongs to incompletely characterized transport systems - putative uncharacterized transport proteins (73.8%), G protein coupled receptors (58.6%) and major facilitator family (MFS) (58.6%). From NCBI, it is also known that this protein is a nucleotide sugar transporter protein which is also likely to be post-translationally modified and belongs to chaperone or intracellular trafficking group of proteins.

Various post-translational modification sites for LPG2 protein of various Leishmania strains were identified using protein predict program. Two aspargine glycosylation sites (at 2^nd amino acid [NHTR] and at 335^th amino acid [NDTS]) have been found in L. infantum and L. donovani whereas only one glycosylation site at 2^nd amino acid [NHTR] was found in other three Leishmania strains (Table 5). Two protein kinase C activation sites were predicted in three Leishmania strains and one PKC site was predicted in other two strains. Four casein kinase II phosphorylation (CKP) sites are present in two Leishmania strains whereas in other strains one or two CKPs are present. One to five numbers of N-myristoylation sites are present in different Leishmania strains. Only one disulfide bond between 20^th and 244^th amino acid residues was detected in this protein of different Leishmania strains (Table 5).

ELM server detected short functional sites in LPG2 protein. In L donovani, L. major and L. infantum, phosphothreonine motif binding a subset of FHA domains having a preference for an acidic amino acid at the pT+3 position (Nucleus, Replication fork) (LIG_FHA_2) has been predicted at 318-324[GKTTAES] position. In L. braziliensis the predicted site and amino acids involved were different (Site 306-312 and sequence is SATDAEN) (Table 6). Since phosphorthreonine motif of LPG2 protein of L. braziliensis is different, according to this motif, LPG2 protein of all other four Leishmania strains are homologous to each other which is also confirmed by Clustal W and phylogenetic analysis. From analogy point of view, all the five strains are equal as the phosphothreonine motif is found in all five Leishmania strains.

The MAP kinase (MAPK) cascades convey a signal in form of phosphorylation events. MAPKs are phosphorylated by MAP kinase kinases (MAPKKs), phosphorylate various targets, such as transcription factors and MAPKactivated protein kinases (MAPKAPKs), and are dephosphorylated and inactivated by several MAPK-phosphatases (MKPs) (Sturgill et al., 1991; Ahn et al., 1992; Nishida et al., 1993; Marshall et al., 1995). In LPG2 protein of three Leishmania strains, MAPK interacting molecules (e.g. MAPKKs, substrates, phosphatases) carrying docking motif helping to regulate specific interaction in the MAPK cascade, have been detected in 75- 82 residues (Table 6).

Amino acid sequence “GRLVLA” (223^rd - 228^th) of L. mexicana, was found to have an RXXL motif that binds to the Cdh1 and Cdc20 components of APC/C thereby targeting the protein for destruction in a cell cycle dependent manner (Table 6).

Cell cycle depends upon the well orchestrated activation and deactivation of cyclin-dependent kinases which phosphorylate a number of substrates required for entry into the next phase of the cell cycle (Takeda et al., 2001). Three substrate recognition sites have been identified in LPG2 protein of Leishmania strains that interact with cyclin and thereby increase phosphorylation by cyclin / cdk complexes which are required for cell cycle events. Predicted proteins also have the MOD_CDK sites which are used by cyclin inhibitors (Table 6).

USP7 plays an important role in regulating cell proliferation and apoptosis through p53 and Mdm2 interactions (Saridakis et al., 2005). Motif containing amino acid sequence[AKASS-304^th -308^th] of LPG2 protein of four Leishmania strains is the USP7 NTD domain binding motif variant based on the MDM2 and P53 interactions, this site is absent in Leishmania braziliensis (Table 6).

WW domain is one of the domains mediating cellular processes which require physical interactions between proteins (Bedford et al., 1998). Two to four numbers of class IV WW domains interaction motifs (phosphorylation-dependent interaction motifs found in both nuclear and cytosolic proteins) are present in LPG2 protein of Leishmania strains (Table 6).

CK1 is a ‘‘phosphate-directed’’ protein kinase which is able to phosphorylate with high efficiency Ser/Thr residues specified by a prephosphorylated side chain (either pS or pT) at position n–3 (or less effectively n–4) (Meggio et al., 1979; Donella-Deana et al.,1985; Flotow et al., 1990). This observation led to the concept of ‘‘hierarchical phosphorylation’’ (Roach et al., 1991) and the term ‘‘primed phosphorylation’’ to indicate the ability to phosphorylate residue(s) specified by another phosphorylated residue at a predetermined (critical) position. This feature is shared by a small number of acidophilic Ser/Thr kinases, notably CK1 and CK2, glycogen synthase kinase 3 (GSK3), and the Golgi apparatus casein kinase (G-CK) (Pinna et al., 1996). Casein kinase I have a wide variety of substrates. CK1 phosphorylation motifs (7-11 numbers) (throughout the LPG2 sequence) have been identified in LPG2 protein of Leishmania strains. Reports on various CK1 inhibitors are known which can be applied to LPG2 protein by docking to find novel drug candidates for leishmanisis treatment (Table 6) (Rena et al., 2004). Protein kinase CK2 is a pleiotropic and ubiquitous serine or threonine kinase, which is highly conserved during evolution (Faust et al., 2000). Protein kinase CK2 can phosphorylate many protein substrates in addition to casein. CK2 phosphorylation motifs have been identified in LPG2 protein of Leishmania strains, one at the beginning of amino terminus and others towards carboxy terminus.

Glycogen synthase kinase 3 (GSK3) is a well conserved serine/threonine kinase that is implicated in different cellular processes controlling cell proliferation and programmed cell death (Frame et al., 2001). In resting cells, GSK3 is a constitutively active kinase that phosphorylates a wide range of protein substrates to directly inhibit their biochemical activities, interfere with their sub-cellular localization, or promote their degradation (Ali et al., 2001). GSK3 phosphorylation recognition sites are found in LPG2 protein throughout the sequence of Leishmania. On comparison of GSK3 sites, amino acids involved in formation of this recognition site are same but in few cases it is different.

N-glycosylation motifs have been detected in different strains at the beginning of N-terminal and towards carboxy terminus. In LPG2 protein of L. braziliensis, three Nglycosylation motifs are present. Among the different types of glycosylation, the N-linked attachment of sugars to the polypeptide backbone is by far the most abundant modification (Vijay et al., 1998).

Two proline-directed kinase (e.g. MAPK) phosphorylation sites (124-130 and 246-252) are found in four Leishmania strains whereas in L. braziliensis, four PDK phosphorylation sites are present. One way that they achieve this is through direct interactions with substrate residues flanking the phosphorylation (P) site (Pinna et al., 1996; Lu et al., 2002; Songyang et al., 1996). The preference for proline at the P+1 position may be linked to downstream signaling mechanisms mediated by Pin1 proline isomerization (Zhou et al., 1999).

Tyrosine-based sorting signal sequences have been found throughout the LPG2 protein sequence of Leishmania which is responsible for the interaction with mu subunit of AP (adaptor protein) complex. Multiple sorting steps within eukaryotic cells are mediated by tyrosine-based sorting motifs. These motifs are recognized by the medium-chain subunits of heterotetrameric adaptor complexes (Stephens et al., 1998).

Some proteins re-exported from the nucleus contain a leucine-rich nuclear export signal (NES) binding to the CRM1 exportin protein. CRM1 mediates the nuclear export of proteins exposing leucine-rich nuclear-export signals (NESs) (Kutay et al., 2005). Two such motifs [36^th - 47^th and 103^rd -118^th amino acids] are detected in LPG2 protein of Leishmania strains except L. major where only one site [36^th - 47^th amino acids] is present. Hence the LPG2 protein manufactured in nucleus is transported to cytosol or Golgi due to presence of these motifs in LPG2 protein itself.

Interestingly, the sequences “KTTTES” and “KAQTPS” of LPG2 protein of L. major and L. infantum respectively matches a 14-3-3 interaction consensus site derived from natural interactors. Dimerization of 14-3-3 is essential for a good interaction (Cahill et al., 2001; Tzivion et al., 1998). P53 tumor suppressor protein, which plays a major role in maintaining genomic stability, has only one 14-3-3 binding site (Waterman et al., 1998). Hence LPG2 protein of Leishmania may have genomic stability role, which is required to be proved experimentally.

Only novel motif [FKSE] recognized for modification by SUMO-1 has been identified in LPG2 protein of L. major. The small, ubiquitin-related protein SUMO-1 is highly conserved from yeast to humans (Muller et al., 2001) and has been associated with subnuclear localization of many cellular proteins. Like ubiquitination, sumoylation leads to attachment of SUMO-1 to target proteins through the γ-NH2 group of lysine residues, using a cascade of E1, E2, and E3 enzymes. SUMO-1 is an important determinant of protein localization, required for the speckled nuclear distribution of the proteins PML, TEL, and HIPK2.

Likewise major TRAF2-binding consensus motif found in members of the tumor necrosis factor receptor (TNFR) superfamily, which initiate intracellular signaling by recruiting the C-domain of the TNFR-associated factors (TRAFs) through their cytoplasmic tails, is also present in LPG2 protein of only one strain i.e. L. braziliensis.

In this analysis we have identified many motifs in LPG2 protein of Leishmania, which were previously reported in other proteins of different organisms, i.e. yeast to mammals. It is learned from this computational analysis that in LPG2 protein of L. braziliensis several distinct motifs are present which are not found in other Leishmania strains. The graphical view of all the motifs that present in LPG2 protein of all Leishmania Strains are shown in (Figure. 5).

Potential ligand binding sites in LPG2 protein of all Leishmania strains (L. donovvani, L. major, L. infantum, L. mexicana and L. braziliensis) have been found by using pocket finder program (Table 7) (Ruppert et al., 1997). Different LBSs of LPG2 protein of Leishmania are shown in (Figure. 6). First LBS of all the strains involve more than 35 amino acids. At most thirty eight amino acids (L. mexicana) involved in formation of binding sites have been found and in few sites a single amino acid is changed or lacking in formation of ligand binding sites, have been detected. Lowest number (four) of amino acids involved in formation of LBSs has been found in L. infantum (7^th). It is known from this analysis that some LBSs are similar in all the strains but other LBSs are specific to each strain. All LBSs of L. donovani and L. mexicana are very similar, except at 1^st and 6^th where one amino acid is absent in L. donovani (1^st site: G-188^th and 6^th site: V-328^th).

All these structural, functional and proteomic analysis about LPG2 protein of Leishmania species will lead to identification of novel lead compounds to eliminate Leishmania infection in Bihar state, India.