1. Introduction
Canine visceral leishmaniasis (CVL) is a zoonotic disease for which the major etiologic agent is
Leishmania infantum [
1]. In the urban environment, dogs are considered the main reservoirs of the parasite due to the impact of these animals in maintaining the transmission cycle [
2]. Parasitic dermotropism in dogs results in a high parasite load in the skin that is then available to the sandfly vector [
3]. During natural leishmaniasis infections, the immune response begins in the skin, where the first wave of cytokines and chemokines appears, attracting several cell types from vessels to the injury site and directing the inflammatory response to various profiles [
4,
5]. Skin lesions in dogs affected by visceral leishmaniasis have been frequently observed and largely discussed in the literature [
6,
7,
8,
9,
10]. Nunes et al. (2018) [
6] demonstrated a high frequency (66.7%) of detection of parasitic DNA in the skin of infected dogs. The skin parasite load correlated with macrophage infiltration and an increase in TLR-2, NOS2, IL-10, and TNF-α expression, and the xenodiagnosis showed that this increase in parasite load in the skin correlated with an increased risk of transmission to sandflies [
11]. The parasite load in the skin, as well as the inflammatory pattern in the dermis, varies according to the anatomical region in which they are observed [
7].
During inflammatory processes, the influx of immune cells into the injury site results in reduced tissue oxygen levels due to the high O2 demand of these cells [
12]. Low oxygen levels are closely related to the regulation of vascular endothelial growth factor (VEGF) mRNA expression and a reduction in NOS2 expression by myeloid cells [
12]. VEGF expression induces angiogenesis of blood and lymphatic vessels by promoting the proliferation of vascular endothelial cells and consequent expansion of the capillary network, and it is considered an important factor involved in the inflammatory response [
13]. Studies regarding
Leishmania major have shown that the presence of the parasite in the skin promotes local vascular remodeling through the high expression of VEGF-A and VEGF receptor-2 (VEGFR-2) and consequent proliferation of blood and lymphatic endothelial cells [
14]. Thus,
Leishmania infections result in changes in the morphology of vascular networks in the skin [
14].
The head of dogs presents more blood vessels than the abdomen [
15]. The vascularity of the skin is divided into three segments: the superficial plexus, the cutaneous plexus, and the subdermal plexus. In dogs, the subdermal plexus is formed by branches of direct cutaneous vessels parallel to the dermis. This difference distinguishes the vascularization of dog skin from that of human skin, where the subdermal plexus is formed from branches of musculocutaneous vessels in perpendicular orientations [
16]. In addition, dogs have intense collateral blood flow to adjacent areas of the skin due to the extensive connections between these vascular plexuses [
16]. During
L. infantum infection, the parasite load has been observed to be associated with the dermal vascular plexuses of dogs, suggesting hematogenous spread of the parasite [
17,
18]. Therefore, due to the greater collateral vascularity in dog skin and the extensive communication between vascular plexuses, the dissemination of amastigotes and the parasite load could increase. The dissemination of
Leishmania (Viannia) in immune cells migrating from lymphatic vessels has been demonstrated in a murine model [
19]; however, the dissemination of
L. infantum through migrating cells from blood vessels and vascular alterations in the skin of dogs has not been completely elucidated. Therefore, the aim of this study was to evaluate the association of anatomical vascular differences and
Leishmania-induced vascular morphological changes with clinical signs and parasite load by analyzing the ear and abdominal skin from dogs naturally infected with
L. infantum.
4. Discussion
Herein, we evaluated the inflammatory infiltrate in two distinct regions of the skin (ear and abdomen), the load and distribution of parasites in situ, and the organization of extracellular matrix collagen and studied the functionality of inflammatory cells through the expression of NOS2, MAC387, and VEGF. The choice of skin regions to be studied was based on anatomical differences in vascularization. In the present work, these anatomical vascular differences in the two regions of the skin were considered as parameters for the study of parasitic dissemination. We were able to observe amastigotes parasitizing endothelial cells and perivascular cells and showed that they were distributed throughout the tissue. The parasites were also observed inside blood vessels, indicating a possible route for dissemination of amastigotes through the skin via dermal blood vascularization. Using confocal microscopy, we demonstrated the colocalization of amastigotes and VEGF+ endothelial cells, which was demonstrated for the first time in dogs with visceral leishmaniasis. Studies of vascular changes have been carried out in other infection models, such as in the murine model of experimental infection in cutaneous leishmaniasis [
14,
22], but not for canine visceral leishmaniasis. Furthermore, the ear skin presented a more intense inflammatory process leading to an increase in the number and diameter of blood vessels, favoring the spread of the parasite from the blood to the dermis and vice versa. In this sense, the ear skin presented parasites in a greater quantity and a more dispersed pattern in the dermis, which was also previously demonstrated by other authors [
7]. The differences observed between the ear and the abdomen constitute important findings, as they can help veterinarians during the diagnostic process, in which they can select a region of the skin that are more suitable for diagnostic methods, in addition to being less invasive. Thus, the expanded study of immunopathogenic mechanisms in the skin makes it possible to understand the dynamics of parasite distribution in a site accessible to the insect vector, contributing to the clinical management of infected dogs in endemic areas.
Dog skin is a strategic site for disease transmission since, regardless of the occurrence of clinical signs, it keeps viable parasites accessible to sandflies [
23,
24,
25], with a parasitic load higher in those who are symptomatic [
26,
27].
Leishmania presents tropism for the skin, as demonstrated in dogs infected by vertical transmission and living in nonendemic areas [
3]. Once in the skin, parasitic antigens stimulate the inflammatory process, leading to the recruitment of leukocytes to the site, with a relationship between the number of CD14+ monocytes in peripheral blood and the inflammatory manifestation in the skin [
28]. In our study, we observed a relationship between local MAC387+ infiltrating leukocytes, the intensity of the inflammatory process, and the dispersion of parasites in the skin, mainly in the ear and in symptomatic animals. These data are similar to previous data showing that the skin of symptomatic dogs presents a higher parasite load associated with a more intense inflammatory infiltrate and a higher rate of apoptosis than asymptomatic animals [
26,
29]. This inflammatory infiltrate is composed of lymphocytes, plasma cells, mast cells, macrophages, and neutrophils [
30,
31,
32]. Cells parasitized with amastigotes, such as neutrophils and macrophages, are frequently observed in inflammatory infiltration in the skin [
29]. Therefore, we used a marker for MAC387+ infiltrating leukocytes (monocytes and neutrophils) associated with changes in the number and diameter of blood vessels as a way of studying the differences in cell influx into the skin of the ear and abdomen, which impact the spread of amastigotes through the dermis.
During inflammatory processes, blood vessels actively participate in the process of cell migration to injured tissues [
33]. Increased vascular permeability, dilation of blood vessels, and recruitment of immune cells to the site of injury constitute important mechanisms of the inflammatory response [
33]. Recently, infiltrated monocytes and macrophages can be identified using the marker MAC 387 [
34]; however, in our study, in addition to monocytes and macrophages, recently infiltrated neutrophils were also found to be MAC 387+. In a murine model, early cell influx and persistence of neutrophils in ear skin after parasite transmission were important factors for the establishment and progression of the disease during infection by
L. major [
35]. In the present study, the ear skin showed a higher number of recently infiltrated cells than the abdominal skin. Altogether, the vascular and inflammatory changes observed in ear skin suggest a role for dermal vascularization in increasing the influx of inflammatory cells and possibly parasites into the tissue.
The increased influx of cells during inflammatory processes results in increased O2 demand and consequently results in low tissue oxygen levels [
12]. To produce high levels of leishmanicidal NO, the induction of NOS2 is necessary, and low oxygen conditions reduce the production of NO by activated macrophages since these cells depend on the availability of O2 as a substrate to produce NO [
12]. There was no difference in NOS2 expression when the ear and abdominal skin were compared; however, we noted that although the ear skin had a more intense level of inflammatory infiltrate than the abdominal skin, a relatively low level of NOS2 expression was observed. To analyze this finding, we calculated the NOS2/MAC387+ cell ratio since the influx of cells reflects the intensity of the inflammatory infiltrate and these cells are responsible for the expression of this enzyme. Notably, when the proportion of NOS2/MAC387+ cells was compared by skin region, the ear skin showed a lower expression of NOS2 than the abdominal skin despite showing a more intense level of inflammatory infiltrate. A possible explanation is that the inflammatory infiltrate in the ear skin had its function subverted by a high amount of parasites, becoming less capable of expressing NOS2 and, consequently, of controlling parasite growth. Other studies should be conducted to test this hypothesis. Conversely, the lower expression of NOS2 in the ear skin may be explained by the lack of available O2 resulting from the hypoxic environment generated by increased cell influx. Environments with low oxygenation may favor parasitic permanence in the tissue [
12]. Because of local stimuli or low oxygenation, resident and/or infiltrating macrophages seemed to acquire an M2 profile (activated macrophages). Moreira et al. (2016) [
36] showed that the number of M2 macrophages (CD163+ cells) in the ear skin of naturally infected dogs is associated with an increased parasite load [
36]. An alternative activation of macrophages to an M2 profile also favors parasite replication [
37]. Although there was no difference in NOS2 expression between ear and abdominal skin, associations between NOS2 expression and the intensity of the inflammatory infiltrate and the presence of granulomatous reactions were observed only in the abdominal skin, where the parasite load was generally lower. As NOS2 expression can be considered an M1 functional marker, this result suggests that in abdominal skin, in those samples with a high intensity of inflammatory cells, macrophages may be directed more toward an M1 profile compared to those in ear skin. Further experiments must be performed to understand the mechanism involved in this phenomenon. The association between parasite distribution and abundance of MAC387+ cells only in abdominal skin would suggest that the inflammatory infiltrate responds accordingly to the presence of amastigotes, both of which are more restricted in perivascular regions when the parasite load is low.
Low tissue oxygen conditions are related to VEGF expression [
12]. In our study, the ear skin showed higher levels of VEGF than the abdominal skin. VEGF can promote vascular dilation in vitro in a dose-dependent manner [
13]. Our results indicate a role for VEGF in the vascular proliferation and dilation of blood vessels in the skin types evaluated, especially in the ear skin. Another role of VEGF is the induction of fenestrations between endothelial cells, increasing vascular permeability and facilitating the migration of cells from blood vessels to tissues [
38]. In ear skin, VEGF may play a role in vascular remodeling, which would provide greater cell influx in this region, as observed in this study. The same effect was not observed in the abdominal skin, which had lower levels of VEGF and lower numbers of MAC387+ cells. In vitro studies by Lai et al. (2019) [
39] described a three-way relationship between M2 macrophages, VEGF levels, and PDL-1 levels. Thus, VEGF is able to stimulate the polarization of macrophages to an M2 profile [
39]. Furthermore, M2 macrophages are able to express PDL-1 under autocrine VEGF regulation, thereby participating in T-cell immunomodulation [
39]. Due to the role of VEGF in macrophage polarization, the higher expression of VEGF in ear skin may influence the regulation of macrophages to an M2 profile in this region, resulting in increased parasite persistence. The observation of a parasite burden in cells expressing VEGF could indicate the favoring of parasite survival in these cells.
Another important factor associated with parasite dissemination and persistence in the skin is changes in dermal collagen fibers, which may favor the migration of parasitized macrophages into the blood vessels [
7]. Thus, we performed an analysis of dermal collagen disruption. We observed that the abdominal skin showed lower collagen disruption than the ear skin. We observed an association between collagen disruption and amastigote distribution. This association indicates that in skin with higher collagen disruption, the distribution of the parasites in the tissue is facilitated. We assessed whether there was an association between collagen disruption and recently infiltrated MAC387+ cells. We were able to observe an association between the two parameters together and the location of the skin. The secretion of matrix metalloproteinases (MMP-9, MMP-3, MMP-2) by local and infiltrating cells and parasites plays an important role in the degradation of the extracellular matrix [
7,
40]] and in increasing cell migration into the tissue [
38]. In the skin of dogs with visceral leishmaniasis, increased levels of matrix metalloproteinases, especially MMP-9, have been observed [
7]. This justifies the increase in migrating cells in regions with disrupted dermal collagen. Taken together, these results suggest an important participation of dermal vascularization in parasite dissemination, mainly in ear skin. In addition to the hematogenous spread observed, the increased cell influx in ear skin in association with high levels of VEGF may result in a macrophage-rich environment polarized to an M2 profile, resulting in increased parasite dissemination and persistence in this region. The ear skin is an area that is more exposed to phlebotomine sandfly bites and other ectoparasites than abdominal skin. This constant stimulation may promote a basal microenvironment predisposed to polarization to a Th2 immune response. As reviewed by Arumugam et al. (2022) [
41], parasites show high persistence at the primary bite site. Additionally, the secretions of sandfly salivary glands may modulate the immune response, favoring parasite persistence [
42]. Thus, exposure to multiple bites could favor an increase and persistence of the parasite load. The abdominal skin is a region less exposed to sandfly bites, and the ratio of NOS2 to MAC387+ cells in the abdominal skin indicates a more effective response against the parasite, in which recently infiltrated immune cells are directed toward an M1 profile. Although the ear skin seems to be a potential site for vector infection, other studies should be conducted to better test the hypothesis described above.