The past, present, and future of Leishmania genomics and transcriptomics.

It has been nearly 10 years since the completion of the first entire genome sequence of a Leishmania parasite. Genomic and transcriptomic analyses have advanced our understanding of the biology of Leishmania, and shed new light on the complex interactions occurring within the parasite-host-vector triangle. Here, we review these advances and examine potential avenues for translation of these discoveries into treatment and control programs. In addition, we argue for a strong need to explore how disease in dogs relates to that in humans, and how an improved understanding in line with the 'One Health' concept may open new avenues for the control of these devastating diseases.

Burden of leishmaniasis and the need for a ‘One Health’ initiative

Leishmaniases are a group of diseases caused by digenetic protozoa of the genus Leishmania, which are transmitted by phlebotominae sand flies (Table 1). Based on recent estimates, up to 0.4 million and 1.2 million cases of visceral (VL) and cutaneous leishmaniasis (CL), respectively, occur each year in 98 countries and three territories where these diseases are endemic [1]. Despite their widespread distribution, over 90% of global VL cases occur in only six countries (India, Bangladesh, Sudan, South Sudan, Ethiopia, and Brazil), while most cases (70–75%) of CL occur in ten countries (Afghanistan, Algeria, Colombia, Brazil, Iran, Syria, Ethiopia, North Sudan, Costa Rica, and Peru) [1]. In most cases, leishmaniases are zoonoses, affecting the poor in rural and natural areas, where a plethora of domestic and wild reservoir hosts and sand fly vectors maintain the infection [2]. For instance, 13 out of the 21 human-infective Leishmania have also been reported in domestic dogs, the latter having a major role in maintaining and transmitting the infection to other receptive hosts via the sand fly vectors [3] (Table 1). In accordance with the concept of ‘One Health’, defined as ‘a movement to forge co-equal, all inclusive collaborations between physicians, […], veterinarians and other scientific-health and environmentally related disciplines […] to improve and defend the health and well-being of all species’ (http://www.onehealthinitiative.com), successful control strategies against human leishmaniases must include preventative measures focussed on the human and animal hosts and arthropod vectors, as well as on the environments where the latter perpetuate [3]. To achieve these goals, a thorough understanding of the host–pathogen–vector triangle, and particularly of their intimate interactions at the molecular level, is imperative. Recent advances in -omics technologies, including genomics and transcriptomics, together with the considerable decrease in the cost of these techniques, provide exciting opportunities to reveal details of the intimate relations between Leishmania parasites, human and animal hosts, and sand fly vectors. In this review, we provide an overview of a range of milestone studies that have used genomics and transcriptomics techniques to improve current understanding of the biology of Leishmania, as well as of the molecular interactions between this parasite and its vertebrate and arthropod hosts. In addition, given the intimate relations between human and canine leishmaniases in endemic areas, and in line with the ‘One Health’ movement, we argue that current and future efforts should be directed towards integrating -omics technologies (i.e., genomics, transcriptomics, proteomics, metabolomics, and interactomics) to achieve a better understanding of the similarities and differences between human and canine infections, with the ultimate aim of developing new diagnostics, and treatment and control strategies against this devastating group of diseases.

Table 1

Principal causative agents of human leishmaniases

Leishmania species	Principal tropisma	Geographical distributionb	Notes on the infection in dogsc
Leishmania aethiopica	C	Old World: Ethiopia, Kenya
Leishmania amazonensis	C	New World: Argentina, Bolivia, Brazil, Colombia, Ecuador, French Guiana, Peru, Suriname, Venezuela	VL cases in Brazil
Leishmania archibaldid	V	Old World: Ethiopia, Kenya, Lebanon, Sudan	VL cases in Sudan
Leishmania braziliensis	C, MC	New World: Argentina, Belize, Bolivia, Brazil, Colombia, Costa Rica, Ecuador, Guatemala, French Guiana, Honduras, Mexico, Nicaragua, Panama, Paraguay, Peru, Venezuela	CL cases in Argentina, Bolivia, Brazil Colombia, Peru, and Venezuela
Leishmania colombiensis	C	New World: Colombia, Panama, Venezuela	VL in a dog in Venezuela
Leishmania donovani	V	Old World: Bangladesh, Bhutan, China, Cyprus, Djibouti, Ethiopia, India, Iraq, Israel, Kenya, Nepal, Saudi Arabia, Somalia, Sri Lanka, Sudan, Ukraine, Uganda, Yemen	Dogs are commonly infected in some countries (e.g., Sudan), but their role as reservoirs is unknown
Leishmania garnhamid	C	New World: Costa Rica, Venezuela
Leishmania guyanensis	C	New World: Argentina, Bolivia, Brazil, Colombia, Ecuador, French Guiana, Guyana, Peru, Suriname, Venezuela	CL cases in Colombia
Leishmania infantum	V, C	Old World: Afghanistan, Albania, Algeria, Armenia, Azerbaijan, Bosnia and Herzegovina, Bulgaria, Central African Republic, China, Cyprus, Croatia, Egypt, France, Gambia, Georgia, Greece, Iraq, Iran, Israel, Italy, Libyan Arab Jamahiriya, Jordan, Kazakhstan, Kirgizstan, Lebanon, Macedonia, Malta, Morocco, Mauritania, Monaco, Montenegro, Oman, Pakistan, Palestine, Portugal, Syria, Romania, Senegal, Saudi Arabia, Slovenia, Spain, Sudan, Tunisia, Turkmenistan, Turkey, Ukraine, Uzbekistan, Yemen. NEW WORLD: Argentina, Bolivia, Brazil, Colombia, Costa Rica, El Salvador, Guatemala, Honduras, Mexico, Nicaragua, Paraguay, Venezuela	VL cases usually found in areas where human cases are reported. Autochthonous cases reported in dogs in the USA (no human cases reported so far)
Leishmania killickid	C	Old World: Algeria, Libyan Arab Jamahiriya, Tunisia
Leishmania lainsoni	C	New World: Bolivia, Brazil, French Guiana, Peru, Suriname
Leishmania lindenbergi	C	New World: Brazil
Leishmania major	C	Old World: Afghanistan, Algeria, Azerbaijan, Burkina Faso, Cameron, Chad, Egypt, Ethiopia, Georgia, Ghana, Guinea, Guinea-Bissau, India, Iraq, Israel, Libyan Arab Jamahiriya, Jordan, Kazakhstan, Kenya, Kuwait, Mali, Morocco, Mauritania, Mongolia, Niger, Nigeria, Oman, Pakistan, Palestine, Saudi Arabia, Syria, Iran, Senegal, Sudan, Tunisia, Turkmenistan, Uzbekistan, Yemen	CL in Egypt and Saudi Arabia
Leishmania mexicana	C	New World: Belize, Colombia, Costa Rica, Ecuador, Guatemala, Mexico, United States	CL in Ecuador and USA
Leishmania naiffi	C	New World: Brazil, French Guiana,
Leishmania panamensis	C, MC	New World: Colombia, Costa Rica, Ecuador, Guatemala, Honduras, Nicaragua, Panama	CL in Ecuador and Colombia
Leishmania peruviana	C	New World: Peru	CL in Peru
Leishmania pifanoid	C	New World: Venezuela	CL in Ecuador
Leishmania shawi	C	New World: Brazil
Leishmania tropica	C	Old World: Afghanistan, Azerbaijan, Egypt, Ethiopia, Greece, India, Iraq, Israel, Iran, Jordan, Kenya, Morocco, Namibia, Pakistan, Palestine, Saudi Arabia, Syria, Turkmenistan, Turkey, Uzbekistan, Yemen	CL cases in India, Iran, Israel, Morocco, and Syria
Leishmania venezuelensis	C	New World: Venezuela

Abbreviations: C, dermotropic; MC, mucotropic; V, viscerotropic.

Based on [63,64].

Based on [54,65,66]. In addition, Leishmania arabica has been reported in dogs in Saudi Arabia [67]. Moreover, other Leishmania species (e.g., Leishmania equatorensis and Leishmania utingensis) [68,69] have been described from wildlife and/or sand flies, but have not yet been detected in humans or dogs.

Species status is under discussion [63,70].

The fight against leishmaniasis: how can -omics help?

The control of leishmaniases generally relies on the early diagnosis and treatment of human cases, vector control, and, in some cases, management of reservoir hosts (i.e., treatment and/or elimination) [3]. However, the control of leishmaniases, as with any vector-borne disease, is not trivial due to challenges relating to intervention programs, mostly in developing countries, where the burden of disease is heavier (due to a combination of factors including, but not limited to, a lack of political will, of human resources, and of infrastructure). In addition, our limited knowledge of the host–pathogen–vector triangle, particularly of their intimate interactions at the molecular level, impairs the development of more affordable and effective control tools, such as antivector vaccines and more effective chemotherapeutics.

-Omics technologies are increasingly being applied to investigations of determinants of disease phenotype [4], mode of action of current drugs [5], and parasite biology [6]. These studies have improved our understanding of the pathogenesis of disease in humans and possible mechanisms of resistance to antileishmanial drugs. Without a doubt, -omics approaches are likely to reveal details of the intimate relations between hosts, parasites, and vectors. This refined knowledge will foster the development of new control tools (e.g., antivector vaccines) that could assist the fight against leishmaniases. The determination of the whole genome sequences of a range of Leishmania parasites causing both VL and CL represents the first step towards these goals, providing the scientific community with a solid infrastructure for postgenomic investigations of the parasite biology, pathogenicity, and mastery mechanisms of manipulation of both insect and vertebrate hosts.

The Leishmania genomes: a ‘toolbox’ to understand host–parasite interactions

Efforts to determine the whole genome sequence of key Leishmania species infecting humans were consolidated in 1994 in Rio de Janeiro (Brazil), with the establishment of the Leishmania Genome Network (LGN) initiative. Not only did this network represent the researchers’ first move to expand existing knowledge of the fundamental molecular biology of this parasite, with a view towards promoting the discovery of novel treatment and control strategies, but it also saw the support of the FIOCRUZ and UNICEF/UNDP/World Bank/WHO Special Programme for Research and Training in Tropical Diseases [7]. In 2005, these efforts proved successful, with the publication of the first complete genome sequence of Leishmania major (causing CL) [8], soon followed by those of Leishmania infantum (causing VL) and Leishmania braziliensis (causing mucocutaneous leishmaniasis; MCL) [9]. In recent years, the advent of high-throughput sequencing technologies (Box 1) has assisted relentless progress in the genomics of human leishmaniases, with the completion of the whole genome sequences of Leishmania mexicana (CL; [10]), Leishmania donovani (VL; [11]) and Leishmania amazonensis (CL; [12]) (Box 1). The availability of these genome sequences has provided unprecedented opportunities to perform detailed comparative analyses of Leishmania species associated with different human diseases at a scale previously unimaginable [9,12].

The genomes of Leishmania vary from 29 Mb (L. amazonensis; [12]) to 33 Mb in size (L. major, L. infantum and L. braziliensis; [9]) and are organised into a variable number of chromosomes (i.e., 34 in L. amazonensis and L. mexicana, 35 in L. brasiliensis, and 36 in L. major, L. donovani, and L. infantum) [12]. Despite the striking variability in pathogenicity and tissue tropism of different Leishmania species, their genomes are remarkably similar, displaying a high degree of conservation in gene content and architecture (synteny) [9,12]. The genomes of Leishmania spp. are characterised by a high gene density, the presence of long arrays of polycistronic gene clusters, and the almost complete absence of introns [7]. However, careful examination of protein-coding genes in Leishmania allowed the identification of a relatively small number of species-specific genes, the majority of which encode predicted proteins of unknown function [10]. Only a few of these genes could be associated to specific tissue tropism. For instance, LinJ.28.0340, a gene specific to L. infantum and occurring as a pseudogene in L. major, L. braziliensis, and L. mexicana [10], has been implicated in the ability of the latter to spread and survive in visceral organs of the vertebrate hosts [13]. Indeed, when a L. donovani gene orthologue of LinJ.28.0340 was introduced into transgenic L. major, the latter displayed a significantly increased capacity to survive in visceral organs of BALB/C mice [13]. Conversely, the spleens and livers of mice infected with the LinJ.28.0340/L. donovani null mutant were characterised by significantly reduced parasite burdens compared with those infected with the wild type L. donovani counterpart, thus providing solid evidence for a role of this gene in the visceralisation of the infection [13]. Among other genes thought to have key roles in the ability of species within the L. donovani complex to colonize visceral organs, those belonging to the A2 gene family are also present as pseudogenes in L. major [14]. These genes were first identified in L. donovani and shown to be exclusively expressed by the amastigote stage (cf. [14]) (Figure 1); subsequently, these genes were demonstrated to be essential for the survival of L. donovani in visceral organs, while transgenic L. major expressing A2 genes displayed increased survival in the spleens of infected mice [14]. Despite the evidence for a role of A2 genes in the pathogenesis of VL, molecules encoding A2 proteins have also been identified in Leishmania species responsible for CL, such as L. amazonensis and L. mexicana [10,12]. While the presence of these proteins in Leishmania parasites with skin tropism has been attributed to functional divergence between Old World and New World species [14], their role in the pathogenesis of CL is yet to be ascertained. Interestingly, a recent comparative analysis of the genomes and transcriptomes of two phenotypically distinct substrains of L. donovani (i.e., one causing VL and one responsible for a large number of cases of CL in Sri Lanka) revealed an increased copy number of A2 genes in L. donovani causing VL, which was also associated with significant upregulation in A2 mRNA transcription and protein expression in strains causing VL [15]. In the same study, Zhang and colleagues [15] identified the presence of several nonsynonymous SNPs in genes from the L. donovani CL strain. Among these was a molecule encoding a ras-like small GTPase-RagC protein; insertion of the corresponding orthologous gene from the L. donovani VL isolate into the CL counterpart resulted in a significant increase in parasite burdens in the spleen of infected mice [15]. These data provided evidence of the impact of SNPs on gene function and phenotype, thus refining current understanding of their potential impact on the pathogenicity of different strains of Leishmania.

Figure 1

Life cycle of Leishmania spp. and examples of molecules putatively involved in parasite infectivity and visceralisation of infection. Phlebotominae sand flies release Leishmania infective stages (i.e., metacyclic promastigotes) to the mammalian hosts during blood feeding (1); the parasites invade macrophages and granulocytes (2 and 3) and develop to amastigotes inside the phagolysosome (4); the amastigote stages replicate within the phagolysosome by simple division (5); then, amastigote-containing macrophages are ingested by susceptible sand flies during the blood meal (6); the parasites are released from the infected macrophages within the sand fly midgut (7), where they transform into procyclic promastigotes and divide. Then, the parasites migrate towards the stomodeal valve (anterior midgut) and transform into different promastigote subtypes that ultimately form metacyclic promastigotes (8). These infective stages are then released into a new mammalian host during a subsequent blood meal (9) [15,25,36,71]. Abbreviation: Cox2, prostaglandin-endoperoxide synthase 2.

While comparative analyses of the whole genome sequences of Leishmania species causing CL and VL represent a solid basis for in-depth investigations of the intimate mechanisms of host–parasite interactions that result in different courses of infection, studies of the regulation of parasite gene expression throughout its life cycle in both the vertebrate hosts and the sand fly vectors are likely to contribute to a better understanding of the pathogenesis of disease. Clearly, the availability of an array of genomes, together with an explosion in microarray and high-throughput transcriptomic sequencing technologies, have facilitated such studies (e.g., [16-19]). However, the same organisation into polycistronic transcription units that makes the genomes of CL- and VL-causing Leishmania so strikingly similar [7] has been deemed responsible for the lack of extensive gene expression regulation at the transcriptional level [20]. Indeed, most Leishmania genes have been shown to be constitutively expressed throughout the transition from promastigote to amastigote stage [21], with post-transcriptional events, including mechanisms that control the abundance of mRNAs, translation rates and post-translational protein stability, hypothesised to have key roles in the regulation of protein abundance [21]. However, the marked variation in chromosome and gene copy numbers among strains of L. infantum, L. mexicana, L. braziliensis, and L. major unveiled, for the first time, a degree of aneuploidy in the genomes of these parasites [10]. Accordingly, unstable ploidy among strains of L. infantum, as well as variable chromosomal contents among cells, revealed that the Leishmania genome is characterised by ‘mosaic aneuploidy’ [11,22]. Therefore, ‘genome plasticity’ and ‘gene dosage’, rather than differential expression of single genes and gene products, are increasingly being considered as two of the keys to the different tissue tropism of Leishmania spp. [22].

Transcriptomics unveils Leishmania-mediated regulation of host gene expression

Several transcriptomic studies have investigated Leishmania-induced regulation of gene expression in infected tissues with the aim to link such responses to disease outcome. As an example, for VL-causing Leishmania, Beattie and colleagues [23] used whole-genome array technologies to compare the gene expression profiles of liver-resident macrophages (Kupffer cells) from mice infected by L. donovani to those of uninfected macrophages exposed to inflammatory stimuli. The authors showed significant upregulation of genes within the retinoid X receptor α pathway (i.e., Rxra), linked to lipid metabolism, in uninfected macrophages exposed to inflammation compared with the infected counterpart [23]; pharmacological perturbation of the activity of this pathway in Kupffer cells resulted in an increased resistance of these cells to Leishmania infection, which led to speculation that either this pathway has a role in the usage of lipids and cholesterol by the parasite, or that Leishmania lipids regulate the activation of innate immune responses that follow the infection [23]. For CL-causing Leishmania, Maretti-Mira and colleagues [24] utilised high-throughput RNA-Seq technologies to characterise and compare the transcriptomes of tissue fragments obtained from human subjects with CL and MCL caused by L. braziliensis [24]. The outcomes from this study highlighted significant upregulation of genes involved in biological pathways linked to the recruitment and activation of immune cells (including lymphocytes, granulocytes, natural killer cells, and antigen-presenting cells) and to regulation of inflammatory responses in tissues from subjects with CL [24]. This suggested that the inability of the host to mount effective immune responses against the parasite at the site of cutaneous infection is linked to the progression of disease [24]. In an effort to characterise differences in macrophage gene expression that might contribute to the ability of different Leishmania spp. to cause localised (CL) or systemic infections (VL), Gregory and colleagues [25] used a DNA microarray approach to perform comparative analyses of the transcriptomes of murine macrophages infected by L. major and L. donovani. Interestingly, both parasites induced a similar differential regulation of relatively small numbers of macrophage genes, with most of these genes unsurprisingly linked to the development of immune responses [25]. The only noticeable difference in gene expression profiling between L. major- and L. donovani-infected macrophages was a remarkable increase in levels of transcription of mRNAs encoding prostaglandin-endoperoxide synthase (Cox2) in the latter, which led to speculation that this pathway is involved in the pathogenesis of VL [25] (Figure 1).

Clearly, the availability of high-throughput transcriptomic technologies has resulted in rapid expansion of the already substantial plethora of knowledge of the molecular interactions occurring between Leishmania and the human host; nevertheless, significant variation in host responses to infection has been described in several studies (cf. [26]), although a review of this variation is beyond the scope of the present article. However, these technologies have also enabled progress towards the exploration of the molecular relationships between the parasite and the sand fly vector and the patterns of Leishmania development into its infective, nondividing metacyclic form [27].

Transcriptomics in Leishmania–sand fly interactions

Transmission of Leishmania from an infected to a susceptible host requires development of the parasites in the midgut of a competent sand fly vector. Macrophages containing Leishmania amastigotes are ingested by sand fly vectors via a blood meal and, once released in the insect midgut, develop through several developmental stages into infective, metacyclic promastigotes [26] (Figure 1). The reproductive mode of Leishmania parasites has traditionally been considered clonal, based on strong linkage disequilibrium (cf. [28]); however, several studies have provided solid evidence of the occurrence of genetic exchange between species and/or strains of Leishmania (i.e., L. major and L. infantum) during growth and development in the sand fly vector, with successful transmission of the hybrid progeny to a susceptible vertebrate host [28-32]. The range of vertebrate and invertebrate host species that Leishmania can infect, as well as the multiple forms of disease that it causes, have been partly attributed to the ability of this parasite to undergo genetic exchange in the sand fly vector (cf. [29]). Clearly, the molecular interactions that occur at the parasite–sand fly interface are key processes that determine the successful development and transmission of Leishmania; therefore, a detailed understanding of these mechanisms has become a priority. Previous studies had used Sanger sequencing of cDNA libraries from the midgut of sand fly vectors of both CL- and VL-causing Leishmania (i.e., Phlebotomus papatasi, vector of L. major and Lutzomyia longipalpis, vector of L. infantum; [33,34]) to identify molecules putatively involved in the development of the parasites in their insect vectors. While sand fly infections by L. major and L. infantum were consistently associated with downregulation of molecules encoding microvilli-like proteins and chymotrypsin and upregulation of trypsin-encoding transcripts, the transcription profiles of peritrophin-like molecules were inconsistent between P. papatasi and Lu. longipalpis [33,34]. Peritrophins are the protein component of the peritrophic matrix (PM), an extracellular chitin-containing structure that encapsulates the blood meal following its ingestion by the sand fly [35]. The formation of the PM (immediately following the blood meal) has long been considered advantageous for Leishmania, because the parasites are thought to be protected from the action of the sand fly proteolytic enzymes during the vulnerable time of development to promastigotes [35]. Several key investigations have contributed to further understanding of the relations between Leishmania promastigotes and the sand fly PM (e.g., [36]). In particular, while previous studies hypothesised a role of Leishmania chitinases in the disintegration of the sand fly PM (cf. [36]), current evidence supports the theory that the breakdown of the PM is independent from the activity of Leishmania enzymes and that parasite promastigotes escape the PM by migrating through a posterior opening that forms irrespective of the infection status of the sand fly [36]. In the same study, Sadlova and Volf [36] showed that the anterior plug of the PM serves as a ‘barrier’ for parasite migration to the thoracic midgut, until its degradation from sand fly proteolytic enzymes is complete [36]. The elucidation of patterns of sand fly gene expression during the disintegration of the PM in the presence (or not) of Leishmania parasites, and during migration of the latter from the abdominal to the thoracic midgut, may help to either confirm or confute this point.

Together with studies of the midgut of sand flies, other investigations used transcriptomic technologies to shed light on the molecular mechanisms that govern the development of Leishmania parasites into their infective metacyclic stage [37]. While little information is available on sand fly molecular pathways acting as trigger of Leishmania metacyclogenesis, recent studies highlighted the role of key genes and gene products in the differentiation of promastigote stages into metacyclic forms in the sand fly vector. Among these molecules, a hydrophilic acylated surface protein (HASPB) and a small hydrophilic endoplasmic reticulum (ER)-associated protein (SHERP) showed increased expression in the metacyclic stages [38]; in addition, creation of HASPB and SHERP null mutants in L. major resulted in the accumulation of non-infective parasite stages in the midgut of the sand fly vector, thus providing evidence for the essentiality of these molecules for parasite development [38]. Investigations of patterns of gene transcription during Leishmania metacyclogenesis in vitro have led to the identification of genes and gene products potentially related to parasite infectivity (e.g. [37,39-41]). For instance, recent functional studies of essential molecules in L. major metacyclic promastigotes highlighted major roles of mitogen-activated protein kinases (i.e., MAPK4) and metallopeptidases of the M24A family in the establishment of intracellular macrophage infections [40] and proliferation in infected cells [41]. These data provided a solid basis for the exploration of the role of these molecules as novel targets for intervention strategies.

In recent years, the search for new and effective preventative measures against Leishmania transmission has also involved the characterisation of key components of the saliva of the sand fly vectors (e.g., [27,42-46]). The interest of the scientific community in salivary gland transcriptomes (‘sialotranscriptomes’) is mainly derived from knowledge that selected saliva proteins have crucial roles in facilitating the successful establishment of Leishmania parasites in vertebrate hosts, including the regulation of the immune response at the site of bite [27,46,47]. Therefore, sialotranscriptomes of several competent sand fly vector species are now available (e.g., [43-45,48-51]), which, in some cases, have led to the selection of key sand fly molecules that are involved in the blood-feeding process and that may assist the immunoevasive strategies of Leishmania [47,52] (Box 2). For instance, a potent vasodilator (maxadilan) abundantly detected in the saliva of Lu. longipalpis [53] has not been identified in transcriptomic data sets from the salivary glands of Lutzomyia ayacuchensis [44]. Similarly, a maxadilan homologue identified in Lutzomyia intermedia showed only 34% identity to maxadilan from Lu. longipalpis [50]. It is worth noting that both Lu. intermedia and Lu. ayacuchensis are vectors of dermotropic Leishmania species, whereas Lu. longipalpis, whose saliva contains large amounts of maxadilan, is the main vector of the viscerotropic L. infantum in the New World [54] (Figure 1). While these observations suggested a role of maxadilan in visceralisation of L. infantum infection [55], the absence of maxadilan homologues from the saliva of sand fly vectors of VL in the Old World raises questions about the role/s of other salivary components in disease progression. Indeed, other enzymes, such as hyaluronidases and apyrases, have been identified using transcriptomic and proteomic technologies from several sand fly vectors of VL in both the Old and New Worlds [45]. These enzymes have been shown to positively contribute to the spread of Leishmania parasites by promoting the enlargement of the feeding lesion and the diffusion of other salivary active compounds (hyaluronidases) and preventing haemostasis (apyrases) [45].

Besides containing components essential to the infection process, the saliva of sand flies contains molecules that can elicit specific immune responses that are indicative of host exposure to sand fly bites (e.g., [56,57]). In particular, three proteins from the saliva of P. perniciosus (i.e., two yellow proteins and an apyrase), expressed in recombinant form, were shown to be useful in determining the intensity of exposure to sand fly bites in experimentally bitten mice and dogs [57]. While cross-reactivity between anti-P. perniciosus antibodies and those from closely related sand fly species was not assessed [57], the authors hypothesised that this may occur. Both yellow proteins and apyrases have been detected in the saliva of a range of sand fly species. However, subtle differences in sequence may result in varying immunogenic properties; future investigations using transcriptomic and proteomic technologies may assist elucidating this point via, for instance, the generation of whole transcript and/or protein data sets from sand fly vector species, with the ultimate aim of identifying suitable targets for the development of commercial diagnostic tools to assess the risk of human and canine transmission in both endemic and nonendemic areas, and the evaluation of the effectiveness of antivector campaigns [56]; this improved knowledge could also aid current efforts aimed at developing recombinant vaccines containing immunogenic components from both the parasite and the sand fly vectors. In addition, thus far, no data are available on the effects of Leishmania infections on the global transcriptional profiles of sand fly vectors. Future studies could, for instance, utilise RNA-Seq technologies to investigate differences in gene expression profiling of Leishmania-infected and uninfected sand flies. Exploring and identifying molecular pathways involved in the parasite–vector–host interactions may lead to the identification of new molecular pathways implicated in the infection process, which would be instrumental for refining current control strategies against sand flies. Undoubtedly, some challenges exist in performing large-scale transcriptomic studies of species for which a reference genome is unavailable; among these challenges, the de novo assembly of full-length transcripts in absence of reference sequences is one of the most significant [58]. Nevertheless, other resources, such as the genomes and transcriptomes of selected mosquito species [59] that are phylogenetically related to sand fly vectors of Leishmania [60], could be exploited for the accurate reconstruction of (at least) a proportion of full-length sand fly transcripts, thus reducing overall project costs and limiting potential biases introduced by de novo assembly.

Concluding remarks and research needs

Over the past decade, advances in genomics and transcriptomics technologies have contributed to considerably enhance our knowledge of the set of molecular interactions that occur within the host–parasite–vector triangle. However, some gaps still exist in our understanding of the similarities and/or differences between human leishmaniases and the disease in animal reservoir hosts. Dogs, for instance, represent the most important host reservoir for L. infantum (causing VL) [3,61]. Therefore, differences and similarities between human and canine infections should be comprehensively analysed. However, most studies of Leishmania immunobiology and genetics, as well as of host–parasite interactions, utilise murine models of infection as ‘mirrors’ of human disease [61]. Given that transmission of key Leishmania species (e.g., L. infantum) to humans strictly relies on the circulation of the parasite among canine populations, elucidating whether dog leishmaniasis serves as a model for human infections should become a priority. This could provide avenues for studies aimed, for instance, at evaluating the ‘translatability’ of novel treatment and vaccine strategies from humans to dogs and vice versa. The availability of in vivo canine models of leishmaniasis [62], together with advances in genomics and/or transcriptomics, proteomics, and metabolomics technologies, may assist this quest. For instance, RNA-Seq and high-throughput proteomics platforms provide a golden opportunity to monitor changes in host gene transcription and protein expression throughout the course of canine and human infections, thus enabling one to draw parallels between them. Similarly, large-scale analyses of metabolites produced during the course of infection, both by the parasite and the vertebrate host, may represent a gold mine for the identification of novel diagnostic biomarkers, as well as of potential new Leishmania ‘Achilles’ heels’ that could assist current programs aimed at breaking the transmission cycle of human and canine leishmaniases.