The gp63 Gene Cluster Is Highly Polymorphic in Natural Leishmania (Viannia) braziliensis Populations, but Functional Sites Are Conserved

GP63 or leishmanolysin is the major surface protease of Leishmania spp. involved in parasite virulence and host cell interaction. As such, GP63 is a potential target of eventual vaccines against these protozoa. In the current study we evaluate the polymorphism of gp63 in Leishmania (Viannia) braziliensis isolated from two sets of American tegumentary leishmaniasis (ATL) cases from Corte de Pedra, Brazil, including 35 cases diagnosed between 1994 and 2001 and 6 cases diagnosed between 2008 and 2011. Parasites were obtained from lesions by needle aspiration and cultivation. Genomic DNA was extracted, and 405 bp fragments, including sequences encoding the putative macrophage interacting sites, were amplified from gp63 genes of all isolates. DNA amplicons were cloned into plasmid vectors and ten clones per L. (V.) braziliensis isolate were sequenced. Alignment of cloned sequences showed extensive polymorphism among gp63 genes within, and between parasite isolates. Overall, 45 different polymorphic alleles were detected in all samples, which could be segregated into two clusters. Cluster one included 25, and cluster two included 20 such genotypes. The predicted peptides showed overall conservation below 50%. In marked contrast, the conservation at segments with putative functional domains approached 90% (Fisher’s exact test p<0.0001). These findings show that gp63 is very polymorphic even among parasites from a same endemic focus, but the functional domains interacting with the mammalian host environment are conserved.


Introduction
The Leishmania species parasite cause a variety of clinical syndromes in individuals living in tropical and subtropical areas of the globe [1]. An estimated 1.5 to 2 million people develop symptomatic leishmaniasis per year, and approximately 12 million people are infected worldwide [2]. American Tegumentary Leishmaniasis (ATL) refers to cutaneous forms of leishmaniasis in the New World, caused by species of the subgenera Leishmania Viannia and Leishmania Leishmania [1,3]. L. (V.) braziliensis is the main cause of ATL in South America, including Brazil [4,5].
The standard treatment for ATL in Brazil is pentavalent antimony [Sb(V)] at a daily dose of 15-20 mg/Kg for 20 to 30 days [5]. However, up to 50% of patients fail Sb(V) therapy [6][7][8][9]. This scenario is even worse in areas of high L. (V.) braziliensis endemicity such as Corte de Pedra in the state of Bahia, Brazil. The most common form of disease due to L. (V.) braziliensis, and the most responsive to therapy, is cutaneous leishmaniasis (CL). More difficult to treat forms of ATL in the region include the classically recognized mucosal leishmaniasis (ML), and emerging clinical forms such as disseminated leishmaniasis (DL) and atypical cutaneous leishmaniasis (ACL). Patients with these unusual forms tend to respond poorly to Sb(V), with failure occurring in up to 60-90% of these individuals [10][11][12][13][14][15][16]. Such experience has led investigators to realize the need for new treatment modalities for ATL [9,16,17].
Leishmanolysin, or glycoprotein 63 (GP63), is a major surface protease (MSP) of Leishmania [18][19][20][21], capable of hydrolyzing a variety of substrates in the parasite's immediate environment within the host [18]. GP63 genes are expressed in promastigotes and amastigotes, and their products are involved in the adhesion to and internalization of the parasite by the host macrophages [19,[22][23][24][25]. Furthermore, GP63 is in part responsible for Leishmania spp. to migrate through extracellular matrix, to avoid lysis by inactivating components of the complement system, and to hydrolyze intracellular macrophage targets [26,27]. Its increased expression has been correlated with increased virulence of L. (V.) braziliensis [28].
Given its role in pathogenesis, GP63 might prove a good target for treatment of or prophylactic immunization against leishmaniasis [21,29]. Its use might be complicated by the variability in genes encoding these molecules [30,31]. However, peptides synthesized according to short regions in GP63, conserved across different species of Leishmania, have been shown to inhibit internalization of the parasite by macrophages [32]. Effective inhibitory oligopeptides surround the sequences SRYD, involved in binding macrophage surface receptors, and HExxH, an essential region for metalloprotease activity [32]. The goal of the current study was to determine whether these short regions are conserved in gp63 genes from a panel of L. (V.) braziliensis isolates from a variety of patients whose diagnosis is temporally distributed over time. The study utilized parasites isolated from ATL patients of Corte de Pedra, Brazil. The results provide the basis for considering these peptides for therapeutic use in management of leishmaniasis. Furthermore, polymorphic gp63 alleles could potentially serve as molecular markers of functionally distinct L. (V.) braziliensis isolates.

Study area
Corte de Pedra is composed of 20 municipalities in a rural area located in the southeastern region of the state of Bahia, in the northeast of Brazil within geographic coordinates (latitude/ longitude) 14°/39°, 13°/39°, 14°/40°, 13°/40°. Lutzomyia (Nyssomyia) whitmany and L. (N.) intermedia sandflies are the most important vectors of L. (V.) braziliensis in this endemic area [33]. The residents work mostly in agriculture, a vocation that often takes them into primary ATL patients' disease definitions CL consists of an ulcerated skin lesion at a single body site with no more than two secondary or satellite lesions, without clinical evidence of mucosal involvement. ML was defined as the presence of an inflamed or ulcerated mucosal lesion(s) at a site that is noncontiguous with any cutaneous lesion. Subjects with ML might have concomitant lesions of CL, but not always. DL was defined as 10 or more skin lesions of mixed types (acneiform, papular, nodular, and/or ulcerated) located on two or more body parts (head, trunk, arms and legs).

Parasites
The L. (V.) braziliensisis isolates analyzed in the present study were obtained by culture of aspirates from the borders of skin or mucosal lesions. Aspirate material was immediately suspended in biphasic liver infusion tryptose-Novy, McNeal and Nicolle medium (LIT/NNN) and incubated at 26°C for 1 to 2 weeks. The suspension was transferred to Schneider's medium complemented with 10% heat-inactivated fetal calf serum and 2 mM L-glutamine, and incubated at 26°C for up to an additional 2 weeks. Parasites were frozen without further subculture in 10% dimethyl sulfoxide (DMSO) 90% growth medium and maintained in liquid nitrogen until used. L. (V.) braziliensis genomic DNA extraction and parasite species determination by PCR Genomic DNA was extracted from approximately 10 6 promastigotes of each isolate. Briefly, parasites were pelleted and suspended in 150μL of TELT buffer (50mM Tris-HCl pH 8.0, 62.5mM EDTA pH 9.0, 2.5M LiCl, 4% v/v Triton x 100) for 5 min at room temperature, followed by phenol-chloroform extraction (150μL) to remove protein and lipids. Nucleic acids were precipitated with ethanol (300μL), followed by an ethanol rinse (1,000μL). The pellets were suspended in 100μL of TE buffer (Tris-HCl 10mM, EDTA 1mM pH 8.0). Samples were stored at either -20°C or -70°C until used. The determination of the infecting Leishmania species was performed by real-time qPCR assay, using primers based on sequences of KDNA1, KDNA3 and MAG1 as previously described [34].

PCR amplification and cloning of selected targets
The gp63 used as reference for the alignment encodes a 557 amino acids peptide: "Leishmania braziliensis MHOM/BR/75/M2904 GP63 leishmanolysin (LBRM_10_0540)", accessed at www. ncbi.nlm.nih.gov/nuccore/XM_001562773.2. We focused the regions that encode the putative functional sites (macrophage binding, protease activity) of GP63 within the host [32]. A fragment of gp63 genes containing these sites was amplified with the forward 5:ATGTCCCGCGAC CGCAGCAG and reverse 5:TCACACCGCCGCTGTGTCGG primers in a 50μL reaction volume, in a 96-well thermal cycler Veriti 1 from Applied Biosystems.
Amplified products were separated in 1.3% agarose gels, stained with ethidium bromide and visualized with a UV trans-illuminator-digital imager (UVP Labworks Laboratory Imaging and Analysis System Inc., CA, EUA) in order to confirm the amplicons were the expected 405 bp size.
Amplified gp63 fragments were cloned using the Original TA Cloning Kit pCR 2.1 VEC-TOR (Invitrogen), according to manufacturer's instructions. The recombinant PCR 2.1 plasmids were transformed into competent DH5α Escherichia coli [35]. Ten colonies with inserts were selected per studied L. (V.) braziliensis isolate.

Sequence analysis
Plasmid inserts were sequenced by the Sanger method at Macrogen Inc. (Seoul, South Korea), with M13 sequencing primers flanking PCR 2.1 cloning sites. Sequencing was bidirectional and only sequences with 100% identity among overlapping insert strands were recorded. Insert sequences were aligned to the Leishmania braziliensis MHOM/BR/75/M2904 GP63 leishmanolysin (LBRM_10_0540) gene sequence with MEGA 5.0 software. ClustaW algorithm was used in the process and no manual adjustments to maximize sequence alignments were necessary.
Upon alignment we searched for events of SNP and indels among the cloned 405 bp gp63 fragments. Polymorphism among sequences of gp63 fragments and their predicted peptides was evaluated manually and confirmed using the Dna Sequence Polymorphism software, version 5. 10.01 [36,37]. We defined a polymorphism as a single bp difference between isolates; a polymorphic allele as a linear gp63 DNA sequence detected in more than one clone per parasite isolate, and in more than one isolate of L. (V.) braziliensis in our study sample.
Classification of gp63 alleles employed Neighbor Joining algorithm. Consistency of nodes within dendrogram was tested by bootstrap.

Statistical Analysis
Differences in the distribution frequencies of each polymorphism detected among gp63 fragments were analyzed by Fisher's exact test. Results with p 0.05 were considered statistically significant.

Ethics statement
This study was approved by the institutional review board of the Federal University of Bahia Medical College, under document number CAAE: 37297614.0.0000.5577.

Identification of polymorphisms in the vicinity of the putative macrophage binding region of gp63 among human isolates of L. (V.) braziliensis
Forty-five alleles of gp63 could be distinguished among 410 clones of the gene fragment evaluated in thirty-five isolates of L. (V.) braziliensis collected between 1992 and 2001, and six isolates of the parasite drawn between 2008 and 2011 from ATL patients of Corte de Pedra. The median number of alleles per parasite isolate was 5, ranging from 2 to 9. One hundred eightysix polymorphic positions were found in the 405 bp fragment, considering all forty-five alleles of the gene (Fig 1). This results in an average of one polymorphic position every 2.2 base-pairs. The frequency of polymorphic positions varied from eight in allele 28 to one hundred one in allele 8 (Fig 1), with a median of 93 polymorphisms per allele (Table 1). 169 polymorphisms consisted of nucleotide substitutions. Many variations occurred in the first two nucleotides of codons, predicting amino acid replacements in 73 (54%) of the translated proteins. Silent changes were also observed in only 52 (31%) nucleotide positions. Nucleotide positions 579, 594, 795 and 906 had either amino acid replacement or silent polymorphisms in different gp63 alleles (Fig 1).

Classification of gp63 alleles
Neighbor-Joining classification revealed two distinct clusters of alleles, which could be further subdivided into four smaller clades (Fig 2; clades A to D). We defined as clade each discrete group of gp63 alleles immediately lower hierarchically to the cluster nodes in the dendrogram. Cluster 1 included 25 gp63 alleles, all belonging to clade A. Cluster 2 included 20 alleles distributed across clades B, C and D.  The predicted peptides reveal that segments of gp63 encoding functional peptides are highly conserved in the natural population of L.

(V.) braziliensis
The predicted peptide encoded by the gp63 fragment is 135 amino acids long. The number of polymorphic amino acid positions varied from seven in allele 28 to fifty-two in alleles 38 and 40, with a median of 43 amino acid changes per allele (Fig 3, Table 1). In 54% of these polymorphic positions there were changes in classes of predicted amino acids between alleles (Fig 3). The 45 gp63 translated alleles presented an overall conservation of 46%. Despite the overall low conservation of the predicted GP63 fragment studied, the two previously described functional regions, defined by the primary sequences SRYD (macrophage receptor binding) and HEVAH (metalloprotease activity) [32] exhibited approximately 70% identity across the 45 gp63 alleles. There were two other highly conserved short regions that flanked the SRYD and HEVAH sequences (PAVGNIPA and KAREQYGC). Altogether, these four stretches of amino acids encompassed 27 of the 135 evaluated residues (Fig 3, shaded positions). The data showed approximately 89% identity between these peptides across the predicted polypeptides of all gp63 alleles sequenced. This is in marked contrast with the overall identity of 46% of predicted protein residues in the full peptide fragments (Fisher's exact p <0.0001; Fig 3).

Discussion
Polymorphic alleles of mammalian hosts have been a focus of investigations into the diverse outcome of infection with different Leishmania species. Our prior work has underscored the fact that polymorphic isolates of L. (V.) braziliensis are independently associated with different clinical outcomes of infection [43,44]. These analyses have focused on anonymous markers that are not associated with any functional significance, in part due to a lack of knowledge of genomic markers in genes of functional significance. In the current study, we investigated the variability in the coding regions of gp63 genes, focusing on peptides with known functional significance in the host. We were able to analyze the variability between gp63 genes among clinical isolates from a population of individuals naturally infected with L. (V.) braziliensis. We found that gp63 genes are highly polymorphic, but that sequences encoding the functional peptides involved in macrophage binding, or in protease activity were remarkably conserved among parasites from one of the regions with highest endemicity of ATL, in northeast Brazil.
GP63 is encoded by tandemly repeated gene clusters in all species in which the gene organization has been investigated [28]. Study of gp63 gene cluster in L. (V.) braziliensis revealed approximately 37 genes in the cluster, with 8 distinct classes [31]. Our experimental approach, in which we selected 10 clones from each isolate for sequence analysis, could not distinguish the full spectrum of polymorphic gp63 genes in single isolates. Nonetheless the polymorphism of the total population was much higher than expected from a single parasite clone, and the number of polymorphic alleles greatly exceeded the expected 8 reported. As such, this analysis leads to a reasonable conclusion that the polymorphism of gp63 genes in the entire population exceeds that expected in individual L. (V.) braziliensis isolates. Further documentation of isolates containing different gp63 genes will require the generation of allele-specific markers from our sequence information, and re-examination of these specific polymorphisms in the collection of clinical isolates.
Polymorphism concentrated in stretches of the gene that encode segments of GP63 that do not directly participate in the interaction between parasite and host cell. We speculate that population-wise the extensive variability of GP63 may be in part driven by selective pressure caused by the host immune responses, during infection with Leishmania spp. Studies suggest that GP63 polymorphisms are more abundant in segments of the protein that serve as epitopes for T and B cells [45,46]. The existence of several dozen, and potentially hundreds of gp63 alleles within a single L. (V.) braziliensis population might perhaps allow the parasite overcome host herd immunity, and maintain endemicity at each successive transmission season of ATL.
We detected four different clades of alleles, and a median of approximately five distinct alleles per parasite isolate. This suggests that gp63 polymorphism in the natural population of L. (V.) braziliensis from Corte de Pedra may be warranted by multiple loci and / or gene copies distributed in different chromosomes. This hypothesis is consistent with observations in previous studies of both physical mapping and whole genome sequencing of L. (V.) braziliensis strains [47,48]. It is important to note that we analyzed gp63 amplified from the genomes of non-cloned parasites by specific PCR primers. As such, the findings reported herein may consist in an underestimation of the actual complexity of these loci, in as much as they may also be influenced by isolates comprised of multiple strains of L. (V.) braziliensis. Future studies should employ deep sequencing approaches to address these limitations. Besides, since we did not evaluate gene expression then we cannot conclude on the functionality of detected alleles.
GP63 and the complement receptor CR3 on host-cells seem to interact in part via the leishmanolysin segment containing SRYD amino acids [49]. SRYD is highly conserved among Leishmania spp., and monoclonal antibodies against this oligo-peptide inhibit internalization . Nevertheless biological testing with inhibition assays is necessary to ascertain that these segments are really functional for the interplay between host and parasite cells.
Experimental studies do not take into account the complexity of natural pathogen populations. As in the reports cited above, immunization and challenge are usually carried out with a single pathogenic strain and its components. This approach is sound, but does not take into account the variability that some antigens used as immunogens may present within and / or between human disease transmission foci.
As we demonstrate in this molecular epidemiology study, GP63 is one such example of a highly variable molecule. Thus experimental results may not easily translate into its successful use as an immune prophylaxis reagent in affected regions. Its ample variability reported for Corte de Pedra likely reflects the realm found in other affected regions as well.
The polymorphisms in gp63 genes reported herein might prove particularly interesting as molecular markers of different parasite isolates. As we previously showed, distinct L. (V.) braziliensis clades are associated with different clinical outcomes of infection [43,44]. If future investigations are able to discover that polymorphisms in this highly important functional protein associate with distinct outcomes of infection, this could have implications for GP63 function in different disease forms. Focusing on polymorphic markers avoids the need to define all gp63 genes present in isolates, but rather highlights only those genes that are different between isolates. This approach might be expanded to study polymorphisms in other known Leishmania spp. proteins important in pathogenesis.