Unrevealing the Mystery of Latent Leishmaniasis: What Cells Can Host Leishmania?

Leishmania spp. (Kinetoplastida) are unicellular parasites causing leishmaniases, neglected tropical diseases of medical and veterinary importance. In the vertebrate host, Leishmania parasites multiply intracellularly in professional phagocytes, such as monocytes and macrophages. However, their close relative with intracellular development—Trypanosoma cruzi—can unlock even non-professional phagocytes. Since Leishmania and T. cruzi have similar organelle equipment, is it possible that Leishmania can invade and even proliferate in cells other than the professional phagocytes? Additionally, could these cells play a role in the long-term persistence of Leishmania in the host, even in cured individuals? In this review, we provide (i) an overview of non-canonical Leishmania host cells and (ii) an insight into the strategies that Leishmania may use to enter them. Many studies point to fibroblasts as already established host cells that are important in latent leishmaniasis and disease epidemiology, as they support Leishmania transformation into amastigotes and even their multiplication. To invade them, Leishmania causes damage to their plasma membrane and exploits the subsequent repair mechanism via lysosome-triggered endocytosis. Unrevealing the interactions between Leishmania and its non-canonical host cells may shed light on the persistence of these parasites in vertebrate hosts, a way to control latent leishmaniasis.


Background
Leishmania (Kinetoplastida: Trypanosomatidae) can be considered a highly successful unicellular parasite that specialises in the phagocytic cells of the mammalian immune system. Infection is initiated by vector-delivered flagellated promastigotes that are engulfed by the mammalian host cells within the skin tissue, where they transform and replicate as round intracellular amastigotes, typically within monocytes and macrophages. The infection is then maintained by replicating amastigotes capable of infecting other host cells. The established amastigote infection manifests as leishmaniasis, a spectrum of diseases with clinical outcomes dependent on multiple factors, including parasite species and virulence, host genetic background, and immune status [1].
Clinical symptoms can usually be detected once the parasite is established in the mammalian host cells; however, some of infected individuals remain asymptomatic with subclinical infection [2]. Moreover, Leishmania parasites can persist within host tissues even after treatment or self-healing [2][3][4]. In all these scenarios, the latent persistence of Leishmania parasites poses a threat to the patient, as they may reactivate during immunosuppression [2]. Therefore, the key questions are how do Leishmania parasites persist in mammalian tissues and where do they replicate?
According to textbook knowledge, Leishmania is an intracellular parasite that can only enter professional phagocytic cells. Lacking the tools, such as an apical complex, that help Surprisingly, Leishmania may also be able to enter host cells actively. The flagellar motility of promastigotes enables them to actively participate in phagocytic uptake by the macrophage (Figure 1B) [29]. This is most likely made possible by the polarised phagocytosis of Leishmania promastigotes induced by the interaction between their flagellar tip and the invaded cell, leading to the formation of pseudopodia. Pseudopodia begin at the tip of the parasite's flagellum and extend towards its cell body [29]. The incessant activity of the promastigote flagellum leads to the reorientation of the parasite within the macrophage, with the flagellum pointing towards the host cell periphery. The oscillations of Leishmania parasites may even cause local damage to the host cell plasma membrane, leading to the Ca 2+ -dependent recruitment of host lysosomes to the site of the parasite invasion and their subsequent exocytosis, which is involved in the host cell plasma membrane repair process, as also observed in Trypanosoma cruzi invading HeLa cells ( Figure 1D) [5,29,38].
The exocytosis of lysosomes during the T. cruzi internalisation occurs by their fusion with the host plasma membrane adjacent to the parasite, inducing the release of acid sphingomyelinase (ASM) and the production of ceramide by the hydrolysis of membrane sphingomyelin. The production of ceramide in the outer leaflet of the membrane induces endocytosis of the injured membrane, which the parasite uses for its own internalisation into the newly formed lysosomal endosomes [5,38]. In contrast to T. cruzi, Leishmaniamediated lysosome exocytosis has been reported to occur after the parasite uptake by the The most widely accepted models of Leishmania promastigote entry into the host cell are actin-dependent phagocytosis (magenta) and caveolin-mediated endocytosis (green). In the newly proposed model of lysosome-triggered endocytosis (blue), Leishmania cause injury (red) to the host cell plasma membrane and exploit the subsequent repair mechanism based on the lysosome exocytosis, which facilitates the endocytosis of the damaged plasma membrane together with the Leishmania promastigotes. (B) Leishmania donovani metacyclic promastigotes preferentially enter primary bone marrow-derived murine macrophages via the flagellar tip, presumably in a receptor-ligand mediated pathway. During the internalisation, lysosomes fuse with the forming phagosome. Prior to complete engulfment, the motile promastigote inside the incomplete parasitophorous vacuole reorients the flagellar tip towards the macrophage plasma membrane, in some cases even protruding out of the host cell. During this phase, the flagellar primary bone marrow-derived murine macrophages via the flagellar tip, presumably in a receptorligand mediated pathway. During the internalisation, lysosomes fuse with the forming phagosome. Prior to complete engulfment, the motile promastigote inside the incomplete parasitophorous vacuole reorients the flagellar tip towards the macrophage plasma membrane, in some cases even protruding out of the host cell. During this phase, the flagellar motility causes damage to the plasma membrane leading to lysosome exocytosis, followed by increased endocytosis. Complete internalisation is accompanied by the loss of promastigote motility and the phagolysosome is located close to the host cell nucleus. Lysosome exocytosis during the later phase of promastigote internalisation appears to promote host cell survival rather than the host cell invasion process itself (as proposed in [29]). (C) Leishmania amazonensis metacyclic promastigotes enter murine fibroblasts (mouse embryonic cell line) via a non-phagocytic pathway dependent on lysosome exocytosis. Prior to internalisation, promastigotes induce host cell membrane injury by an unknown mechanism, probably involving flagellar motility and/or Leishmania-derived pore-forming cytolysins. Membrane damage and the associated increase of intracellular Ca 2+ lead to lysosome exocytosis, followed by increased endocytosis. After internalisation, Leishmania is enclosed in ceramide-rich endocytic vacuoles. In contrast to macrophages, the lysosome exocytosis facilitates the host cell invasion process itself (as proposed in [8]). (D) Trypanosoma cruzi trypomastigotes enter epithelial HeLa cells via a non-phagocytic pathway dependent on lysosome exocytosis. Prior to internalisation, trypomastigotes cause injury to the host cell membrane by an unknown mechanism, probably involving parasite motility and/or Trypanosoma-derived pore-forming toxins. Membrane damage and the associated increase in intracellular Ca 2+ lead to lysosome exocytosis. Acid shingomyelinase released from lysosomes hydrolyses sphingomyelin on the outer membrane leaflet to ceramide, leading to increased ceramide-driven endocytosis of the injured plasma membrane together with trypomastigotes. After internalisation, T. cruzi is found in ceramide-rich endocytic vacuoles. In some cases, engulfed trypomastigotes move towards the host cell plasma membrane, protruding their flagella out of the host cell. During this event, the flagellar motility causes additional damage to the plasma membrane, leading to increased endocytosis. Lysosome exocytosis facilitates the host cell invasion process itself and supports host cell survival by restoring membrane integrity (as proposed in [38]).

Other Potential Host Cells of Leishmania
There is increasing evidence that Leishmania can invade or even survive and multiply in other cells that are not considered to be professional phagocytes. To enter these cells, Leishmania can use both passive as well as active strategies ( Figure 1A).
Although the phagocytic process is more commonly associated with professional phagocytes, it has been shown that almost all cells in the human body are capable of phagocytosis (including skin fibroblasts) and caveolin-dependent endocytosis (including adipocytes, endothelial, and muscle cells) [18,40,41]. Entry into host cells by a process similar to classical phagocytosis has been documented, for example, in Chinese hamster ovary cell lines co-incubated with L. amazonensis amastigotes [42].
However, Leishmania may also be actively involved in the entry process into the noncanonical host cells, including but not limited to fibroblasts, adipocytes, myofibres or epidermal cells, such as pigmented cells (e.g., [8,[43][44][45][46]). Since it is well documented that its close relative T. cruzi uses lysosome exocytosis to invade the host cell, including the non-professional phagocytic cells [5,38,39], it is likely that Leishmania could use a similar strategy [8,28].
While in macrophages, Leishmania-mediated lysosome exocytosis has only been reported after the parasite uptake ( Figure 1B) [29]; in different host cells, it can facilitate the invasion process itself ( Figure 1C) [8,28]. This appears to be the case at least for L. amazonensis entry into the fibroblasts [8]. The colocalisation of L. amazonensis promastigotes that have half-entered a fibroblast with the LAMP marker suggests that lysosome recruitment does indeed occur concomitantly with parasite invasion and that lysosomes donate their membrane to form the nascent parasitophorous vacuole [8]. Two hypotheses have been proposed for the L. amazonensis initiation of the host cell plasma membrane damage leading to lysosome exocytosis and the subsequent endocytosis in these host cells [8]: (i) parasite flagellar motility towards the host cell plasma membrane, causing mechanical damage (as in T. cruzi [38]) and/or (ii) the secretion of cytolytic molecules (such as pore-forming cytolysins) leading to plasma membrane permeabilisation [47,48].
Other possible strategies that Leishmania parasites might use to invade non-professional phagocytes should be verified in the future. It is even possible that the promastigote flagellum itself plays a much more important role in invasion than previously thought. According to older in vitro observations, the flagellum could serve as an anchor inside the invaded cell, by which the parasite pulls itself inside when penetrating cells with limited phagocytic capacity [43]. In fact, Leishmania parasites appear to possess several mechanisms that would allow them to enter their host cells, and it is likely that these mechanisms may alternate or even complement each other depending on the type of the cell attacked and the infecting Leishmania species.
It is obvious that Leishmania parasites have tools to invade even the non-canonical host cells that have been neglected as players in the dynamics of Leishmania infection and its persistence. These cells are listed in Table 1 and include, besides other professional phagocytes (mast cells, dendritic cells, eosinophils, and histiocyte-like cell lines), mainly non-professional phagocytic cells, such as lymphocytes, fibroblasts, adipocytes, epithelial and endothelial cells, mesenchymal stem cells, myocytes, and keratinocytes.
To collect the data for Table 1, we adapted the tables of Rittig and Bogdan (2000) [7] and Chang and Fish (2017) [49], which we further expanded and updated using the following strategy. A literature search of relevant articles was conducted between December 2022 and January 2023 using databases such as Web of Science and PubMed. The search was performed independently by both authors using combinations of keywords, such as this Boolean string: "Leishmania* AND (fibroblast* OR adipo* OR fat OR "epithelial cell*" OR epithel* OR myocyte* OR muscle* OR myofibre OR "stem cell*" OR keratinocyte* OR "endothelial cell*" OR endothel*) AND (multiplicat* OR transform* OR uptake OR internali* OR phagocyto* OR intracellular* OR amastigote*)". The search was not restricted by the year of publication. The retrieved articles were selected based on the eligibility criteria including, but not limited to: an original research article, full text access, detailed description of methodology, host cells of vertebrate origin, and clear statement of results. Exceptionally, secondary citations were used for research articles with a unique output, but not accessible in the full text version. References in selected articles were also evaluated. The Leishmania species names listed in Table 1 correspond to the names provided in the original research articles, as older papers, in particular, did not use more accurate molecular techniques for species identification, and it is therefore not possible to complement them with current taxonomy and nomenclature.
The data listed in Table 1 indicate that the outcome of Leishmania internalisation is likely to depend on the host and Leishmania species, the infecting form of Leishmania (promastigote vs. amastigote), and also the origin of the host cell or tissue. This table only includes vertebrate host cells; however, some Leishmania species have also been found to invade insect cell lines [50][51][52]. Of particular interest is that the cell lines prepared from mosquitoes (Aedes aegypti) or sand flies (Lutzomyia spinicrassa) and mainly containing cell types with epithelial and fibroblast appearance, also support the internalisation of Leishmania promastigotes, together with their transformation into the amastigote form [50,51] and even their multiplication [52].
In the following subchapters, we have focused on studies using host cells that are as close as possible to their natural state, since cancerous or mutated cells (such as those used for immortalised cell lines) may have altered metabolism that may also affect Leishmania entry, survival, and multiplication. Mouse adipocytes derived from primary pre-adipocytes from subcutaneous white adipose tissue 

Fibroblasts
Fibroblasts may play a neglected role in latent leishmaniasis and the disease epidemiology because, in some host-parasite combinations, they can support intracellular parasite survival, transformation into amastigotes, and even amastigote multiplication, resulting in viable progeny capable of transforming back to promastigotes [8].
Several in vitro and in vivo studies have reported that fibroblasts harbour Leishmania amastigotes (Table 1) [8][9][10][11]14,81,[83][84][85][86][87]. Conflicting results have been observed regarding the survival of L. amazonensis in fibroblasts, ranging from limited survival of parasites (e.g., [12,91]) to their successful multiplication [8,10,14,15] (Table 1). Co-incubation of enucleated fibroblasts (cytoplasts, cell nucleus artificially removed in vitro) with L. amazonensis revealed that the parasitophorous vacuole biogenesis and parasite multiplication are independent of the host cell nucleus, and showed these cells to be a promising model for studies focusing on the role of the host cell nucleus during the parasite-host interactions (in particular, the modulation of the gene expression) [15]. The amastigotes of L. amazonensis have also been detected in the perineurial cells (=epithelioid myofibroblasts) of BALB/c mice with experimental cutaneous leishmaniasis [96]. It is interesting that L. mexicana and L. infantum were able to multiply in mouse and hamster skin fibroblasts, but not in skin fibroblasts from rats [10]. In contrast, L. donovani does not appear to be capable of long-term survival and multiplication in fibroblasts in any of the host species tested-human [87], mouse [93], nor chicken [11,95] (Table 1).
Ultrastructural and immunohistochemical analysis of skin biopsies from dogs with naturally acquired leishmaniasis revealed the presence of free and occasionally vacuoleenclosed amastigotes within fibroblasts, while some amastigotes in close contact with the host plasma membrane appear to damage it, indicating that amastigotes also have an active invasion potential [81]. The authors speculated that the source of these amastigotes infecting fibroblasts in the deeper layers of the skin were necrotic macrophages. Similarly, L. donovani amastigotes have been seen surrounded by a closely applied membrane of human foreskin fibroblasts cultured in vitro [87].
The potential involvement of fibroblasts in the pathogenesis of cutaneous leishmaniasis is of particular interest, as recent studies indicate their role in Leishmania immune evasion strategies [13]. Fibroblasts are one of the most abundant cells at the site of transmission and are important producers of macrophage-and neutrophil-attracting chemokines. Fibroblasts also interact directly with macrophages during wound healing and can move by diapedesis, thus being capable of Leishmania dissemination [8]. The relatively long lifespan of fibroblasts and their limited ability to eliminate invaders could lead to the persistence of infection [7]. In addition, fibroblasts are capable of phagocytosis but have a limited ability to control the Leishmania infection through NO production [9]. Indeed, in healed mice, approximately 40% of L. major amastigotes were found in skin and draining lymph node fibroblasts, indicating that they may serve as a safe shelter and a site of potential recrudescence [9,58]. In latent leishmaniasis, the balance may be maintained by neighbouring macrophages producing enough NO to destroy Leishmania amastigotes within the fibroblasts [9].
Similar to professional phagocytes, the entry of Leishmania into fibroblasts could be either passive or active ( Figure 1A). An older in vitro study, supported by transmission electron microscopic visualisation, claimed that the infection of fibroblasts by L. braziliensis promastigotes occurs via the parasite-induced phagocytosis, when the parasites enter fibroblasts with their flagellar end through pseudopodia-like formations on the host cell surface [86]. The engulfed promastigotes settled in vacuoles that did not fuse with secondary lysosomes and transformed into amastigotes. Uptake is likely to be receptor-mediated, as skin fibroblasts cocultured with L. amazonensis promastigotes showed an increased expression of the mannose receptor during the early stages of infection, possibly binding mannosylated ligands on the promastigote surface [91]. This modulation of fibroblast mannose receptor expression was reversed concomitantly with the loss of parasite viability, indicating that the presence of viable parasites is required to maintain it [91].
However, the active entry of Leishmania into the fibroblasts has also been reported ( Figure 1). It has been shown that L. amazonensis induces its entry into host fibroblasts by damaging the fibroblast plasma membrane, leading to the internalisation of promastigotes during the subsequent membrane repair process ( Figure 1C) [8]. The invasion process is independent of host actin remodelling (i.e., this process is not a form of induced phagocytosis) and involves Ca 2+ -dependent recruitment/exocytosis of host lysosomes to repair the plasma membrane. During this process Leishmania actively induces lysosome-triggered endocytosis, a cell invasion mechanism based on the transient permeabilisation of the host cell plasma membrane [8]. Similar to T. cruzi interactions with HeLa cells [38], this invasion process has only been observed with the viable metacyclic stage of the parasite; neither dead metacyclic nor the procyclic promastigotes (the developmental stage of Leishmania found only in the vector) were engulfed by fibroblasts [8]. On the other hand, parasiteunrelated membrane injury promotes internalisation of L. amazonensis promastigotes [8] as well as T. cruzi trypomastigotes [38].

Adipocytes
Adipose tissue has also been postulated as a potential reservoir for intracellular pathogens with the ability to induce disease relapses [46]. Indeed, recent studies have confirmed the ability of Leishmania promastigotes to infect adipocytes in vitro and in vivo [45,46]. Adipose tissue could serve as a perfect reservoir for Leishmania, especially considering that this tissue niche is also used by its relatives-extracellular T. brucei [120] and intracellular T. cruzi [121]. Although their survival strategies are not entirely the same as those of Leishmania, we may still find some parallels that can inspire future research. Trypanosoma brucei has been shown to accumulate in the adipose tissue of mice early after infection [120]. These adipose tissue extracellular T. brucei forms, which are transcriptionally distinct from bloodstream forms, can replicate and are capable of infecting a naïve host [120]. Moreover, trypomastigotes of T. cruzi invade human and mouse adipocytes and transform into amastigotes during the acute phase of infection [121]. Although replication in adipocytes has not yet been directly observed in Leishmania, L. amazonensis, and L. brasiliensis amastigotes recovered from infected cells retain the ability to differentiate into replicative promastigotes [45], and recovered L. infantum amastigotes were infectious to another host [46].

Mesenchymal Stems Cells
Mesenchymal stems cells (MSCs) residing in bone marrow could also provide a perfect protective niche for Leishmania parasites and support their persistence in the host organism. Among other factors, these cells are (i) capable of self-renewal and have low-reactive oxygen species properties (ideal for long-term parasite viability), (ii) do not normally express MHC Class II on their surface and their MHC Class I molecules do not trigger effector functions of cytotoxic T-cells, and (iii) express potent drug efflux pumps (probably enabling Leishmania drug evasion) [109]. Indeed, L. infantum promastigotes successfully infected the CD271+/Sca1+ bone marrow MSCs of C57BL/6 mice and transformed into amastigotes in both in vivo and in vitro settings [109]. Moreover, several Leishmania species, including agents of both the cutaneous and visceral leishmaniases (Table 1), have been shown to persist for some time in an inactive form in cultures of adipose tissue-derived MSCs [16]. As stem cells generally remain dormant in the absence of an exogenous stimulus, they may represent ideal reservoir host cells for Leishmania [16]. The mechanism of invasion of mesenchymal stem cells remains to be elucidated, although phagocytic properties have already been reported for adipose tissue MSCs [16].

Myocytes
Muscle cells are highly parasitised by T. cruzi [122] but are overlooked in Leishmania studies. It is therefore of particular interest that L. infantum amastigotes have been detected in the muscle biopsies from dogs with obvious muscle damage, causing a progressive polymyositis affecting the masticatory and skeletal muscles [110]. Moreover, another study reported canine leishmaniasis associated with myositis of adnexal, extraocular and intraocular smooth and striated muscles that were parasitised by Leishmania amastigotes [111].
In addition, there are in vitro studies reporting the internalisation and replication of several Leishmania species in muscle cells of different origins [43,102]. The ability of Leishmania amastigotes to invade muscle fibres was also confirmed by an in vivo experimental study comparing the muscle infection by L. amazonensis in two mouse strains with a different susceptibility to leishmaniasis-susceptible BALB/c mice and resistant C3H.He mice [112]. While the BALB/c mice showed an intense inflammatory infiltrate between the amastigote-infected myofibres, followed by a total muscle destruction at day 90 p.i., the C3H.He mice showed only a mild inflammatory infiltrate without intracellular amastigotes, followed by a muscle repair process [112].
In BALB/c mice experimentally inoculated with L. major promastigotes, we also occasionally observed the presence of amastigote-like structures within muscle fibres (Figure 2, unpublished results). However, these observations require confirmation by transmission electron microscopy or immunolabelling.
phagocytic properties have already been reported for adipose tissue MSCs [16].

Myocytes
Muscle cells are highly parasitised by T. cruzi [122] but are overlooked in Leishmania studies. It is therefore of particular interest that L. infantum amastigotes have been detected in the muscle biopsies from dogs with obvious muscle damage, causing a progressive polymyositis affecting the masticatory and skeletal muscles [110]. Moreover, another study reported canine leishmaniasis associated with myositis of adnexal, extraocular and intraocular smooth and striated muscles that were parasitised by Leishmania amastigotes [111].
In addition, there are in vitro studies reporting the internalisation and replication of several Leishmania species in muscle cells of different origins [43,102]. The ability of Leishmania amastigotes to invade muscle fibres was also confirmed by an in vivo experimental study comparing the muscle infection by L. amazonensis in two mouse strains with a different susceptibility to leishmaniasis-susceptible BALB/c mice and resistant C3H.He mice [112]. While the BALB/c mice showed an intense inflammatory infiltrate between the amastigote-infected myofibres, followed by a total muscle destruction at day 90 p.i., the C3H.He mice showed only a mild inflammatory infiltrate without intracellular amastigotes, followed by a muscle repair process [112].
In BALB/c mice experimentally inoculated with L. major promastigotes, we also occasionally observed the presence of amastigote-like structures within muscle fibres (Figure 2, unpublished results). However, these observations require confirmation by transmission electron microscopy or immunolabelling. The mechanism of Leishmania entry into muscle cells remains to be elucidated. While older studies hypothesised the ability of promastigotes to actively penetrate target cells through the motility of their flagellum acting as an anchor [43], others speculate that Leishmania parasites could penetrate myocytes, rich on fucose-mannose ligands, using the fucose-mannose receptor [112]. Parasite-unrelated injury to muscle cells may facilitate The mechanism of Leishmania entry into muscle cells remains to be elucidated. While older studies hypothesised the ability of promastigotes to actively penetrate target cells through the motility of their flagellum acting as an anchor [43], others speculate that Leishmania parasites could penetrate myocytes, rich on fucose-mannose ligands, using the fucose-mannose receptor [112]. Parasite-unrelated injury to muscle cells may facilitate Leishmania uptake via lysosome-triggered endocytosis during repair of the damaged plasma membrane, as has been shown for L. amazonensis and fibroblasts [8].

Endothelial Cells
Endothelial cell parasitism by Leishmania remains unclear, as only a few studies (some of them older) with conflicting results are currently available [113,115,116,123,124]. How-ever, similar to T. cruzi [122], it is likely that at least some Leishmania species could be able to infect endothelial cells.
While a recent in vitro analysis of a human microvascular endothelial cell line (HMEC-1) co-incubated with L. infantum promastigotes reported the absence of internalised parasites [116], the intracellular localisation of Leishmania has been reported from endothelial cells lining the blood-vessels of the kidney, liver, and colon in visceral leishmaniasis [81,113,114,123]. However, it should be noted that the accurate localisation of amastigotes in routine histopathology (usually used as the sole detection method in older studies) is challenging and may be less sensitive than immunolabelling, as demonstrated in canine cutaneous leishmaniasis [125]. For example, the histopathological examination of a human subcutaneous nodule after experimental inoculation with L. tropica promastigotes revealed the abundant presence of amastigotes within the endothelial cells lining the capillaries near the centre of the lesion [115], but another study reported that L. braziliensis parasites were more likely to be attached to the wall of dermal blood vessels and free in the capillary lumen [124]. Accordingly, the L-SIGN receptor, specifically expressed in liver sinusoidal endothelial cells, acts as a receptor for viscerotropic L. infantum (but not dermotropic L. pifanoi), resulting in the strong binding of amastigotes to the cultured endothelial cells; however, no invasion of these cells was reported [126].

Keratinocytes
As one of the most abundant epidermal cell types and an important source of immunomodulatory signals in the skin, keratinocytes may play a key role in the early stages of insect-borne diseases, including leishmaniasis. These cells appear to be unsuitable for Leishmania replication, the function of keratinocytes being more immunomodulatory. Although human keratinocytes were shown to internalise L. infantum or L. major promastigotes in vitro at low levels, they did not allow efficient amastigote multiplication [117]. Accordingly, several in vivo or ex vivo studies reported the absence of keratinocytes parasitised by Leishmania [55,96,127]. Nevertheless, keratinocytes exposed to extracellular L. infantum and L. major parasites have been shown to alter their transcriptional signatures and appear to be stimulated to release factors that influence monocyte infection [117]. The authors hypothesised that the pro-inflammatory response of keratinocytes induced by L. infantum may limit the local survival of the parasite in the skin environment and thus promote its dissemination, whereas the 'silent' interaction of L. major with keratinocytes may increase its ability to survive locally, leading to cutaneous leishmaniasis [117]. It has therefore been proposed that keratinocytes may initiate or suppress the pro-inflammatory response at the site of infection, thereby influencing tissue pathology [117].

Conclusions and Perspectives
Leishmania well-recognised primary host cells are professional phagocytes, macrophages, and monocytes, but other cell types might also be infected with Leishmania parasites. The presence of L. major transcripts has been shown to be associated with multiple cell types at the site of infection. In addition to the well-known host cells of the myeloid lineage (macrophages, inflammatory monocytes, neutrophils, and dendritic cells), they are also found in endothelial and epithelial cells, fibroblasts, keratinocytes, chondrocytes, and myoblasts [128]. However, the key requirement for the host cell-to support Leishmania survival and multiplication-has only been proven for a few of them (e.g., [8,10,46]).
The most studied non-canonical Leishmania host cells are fibroblasts [8][9][10][11][12][13][14][15][81][82][83][84][85][86][87][88][89][90][91][92][93][94][95][96]. They can migrate within the skin tissue [8], potentially allowing Leishmania to spread from the site of transmission, thereby disseminating the infection and enhancing the possibility of being engulfed by the vector during blood feeding. Due to their relatively long lifespan and low leishmanicidal activity, fibroblasts may serve as an ideal reservoir host cell in latent cutaneous leishmaniasis [9]. In visceral leishmaniasis, adipocytes may also play this role [45,46]. The putative reservoir host cells for latent infection should ideally support the intracellular survival of Leishmania for a prolonged period, ideally also facilitating amastigote replication. If the cell cannot support Leishmania replication, the amastigote may leave the reservoir host cell, multiply in monocytes or macrophages, and find safe shelter in another reservoir host cell. Such a scenario may be possible since Leishmania parasites are not dormant during latent leishmaniasis but are continuously replicating [129], being under the tight immune control of macrophage-derived NO [58].
The promiscuity of Leishmania with respect to host cells has been demonstrated using the cell lines of invertebrate origin [50][51][52], showing the ability of Leishmania parasites to also infect-under artificial conditions-the cells of the insect vector [51], where they naturally occur only extracellularly [130,131]. In humans, such artificial conditions could be induced, for example, by HIV immunosuppression, leading to an unusual and rare localisation of Leishmania within sweat gland epithelial cells in the host dermis [97,98]. To our knowledge, such localisation has not been observed in immunocompetent individuals (Table 1). Moreover, the hidden promiscuity of Leishmania parasites may have unexpected therapeutic implications, affecting the screening of new drug candidates. The widely used in vitro testing on promastigotes (vector-derived developmental stage) [132] may not reveal the full complexity of amastigote presence in different tissues/cells during the mammalian host infection. Pharmacokinetics and pharmacodynamics would be better evaluated in the context of multiple Leishmania host cells.
The ability of Leishmania to invade different cell types may also affect the epidemiology of leishmaniasis, being another factor to be consider during the transmission from the mammalian host to the insect vector. Leishmania is able to persist in the uninflamed skin of the mammalian host, while preserving its infectious potential for sand fly vectors [133]. As these parasites can gradually accumulate in the skin, even in clinically healthy hosts, and remain infectious to their insect vectors [133], it is necessary, at least in endemic areas, to monitor not only cured patients but also potential reservoirs, such as dogs or asymptomatic humans.
Leishmania internalisation into the host cell is well described in professional phagocytes, but the mechanisms in non-canonical host cells are less well understood. The mode of Leishmania entry into the host cells, of whatever type, appears to be multifactorial, depending on the host cell type, Leishmania developmental stage (promastigotes vs. amastigotes), Leishmania virulence, as well as on the immune status of the host (e.g., the presence of anti-Leishmania antibodies as opsonising agents). The Leishmania internalisation could be receptor-mediated (e.g., [5,30,31,91]), where Leishmania appears to be passively engulfed, or actively initiated by Leishmania promastigote by wounding the host cell plasma membrane [8], e.g., via the movement of its flagellar tip [29,43]. At least three possible entry pathways have been described ( Figure 1): (i) actin-dependent phagocytosis, (ii) caveolinmediated endocytosis, and (iii) lysosome-triggered endocytosis associated with the host cell plasma-membrane repair mechanism [7,8,28,[32][33][34]36,37].
Undoubtedly, more studies are needed to reveal all the details of the interactions between Leishmania and its various host cells in the hope of finding a way to better control leishmaniasis, including its latent form.

Data Availability Statement:
No new data were created or analysed in this study. Data sharing is not applicable to this article.

Conflicts of Interest:
The authors declare no conflict of interest. The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript; or in the decision to publish the results.