The role of innate immunity is gaining interest in the field of lentiviral infections [33, 34]. An important component of innate immunity is the MR (CD206), a group VI C-type lectin, present at the cell surface and endosomes . This study identifies the ovine MR genetic characteristics, the corresponding nucleotide and deduced amino acid sequences and the putative protein structure, all of which were found very close to those found in other mammals [35, 36]. Expression of the ovine MR is being attempted at present through transfection of SRLV non-permissive cells.
In the host, differences in MR oligomerization , heterogeneity of MR N-glycosylation , as well as variation in individual genetic makeup  and health status  may account for differences in MR expression. Like in other species, the ovine MR expression differed among cell types, mRNA specific transcripts being present in macrophages (BMDM) and synovial membrane (GSM) cells, but not in skin fibroblasts (OSF). In SRLV pathogenesis, GSM cells are known to be infected in vivo  and if these cells do differentially express MR on the cell surface in vivo as they do in vitro, MR could represent an entry pathway of ENV-mannosylated viruses through carpal joint infection.
The polyclonal reagent available to detect MR protein expression was produced against a 57-amino acid peptide of the human MR, but the limited size of this peptide (about 3.9% of the whole MR protein) and sequence differences (12 mismatches) between this human MR peptide and the homologous peptide of ovine origin may have led to the evident lack of cross-reactivity of the polyclonal reagent with the ovine MR. In contrast, this polyclonal antibody reacted with mouse MR (with 16 mismatches in the peptide region compared to human MR). The kind and site of amino acid substitutions may account for the differences in cross-reactivity of the polyconal reagent to the ovine vs. the mice peptide.
Evidence of VMV infection in CHO cells transfected with MR (from species like mice) supports the hypothesis that MR expression is sufficient in vitro for VMV infection in particular cell types. In a previous work  CHO/mouse somatic cell hybrid lines became permissive to SRLV if they included mouse chromosome 2 or 4. The finding that mouse chromosome 2 contains the MR gene, originally named Mcr in that species , may explain this finding and suggests that the involvement of the membrane associated C-type lectin DC-SIGN (dendritic cell-specific ICAM-grabbing non-integrin) in this permissiveness (and not that of MR) could be excluded, since DC-SIGN is encoded by chromosome 8 (and not chromosomes 2 or 4). Similarly in our study, the involvement of DC-SIGN was unlikely, since dendritic cells were not used. However, redundant or alternative pathways of virus capture via lectins may coexist .
The observation that CHO cells become infected, upon transfection with MR from species not susceptible to VMV (mice), may indicate a non species-specific viral interaction with the MR. However, the CHO-MR cells infection by VMV was not productive, indicating that subsequently to viral integration, factors involved in viral production appear to differ when comparing CHO-MR with GSM cells (expressing MR), the latter being capable of productive infection.
There must be routes of virus entry into skin fibroblasts (OSF) other than mannose binding lectins, as in these cells blockade of infection by mannan did not take place but a productive VMV infection was observed. Accordingly, OSF appeared to bind, via a yet unidentified receptor different from MR, ENV viral protein glycosylated residues, taking into account the need of ENV glycosylated residues for cell fusion , and the observation that viral replication was strongly inhibited in these cells by ConA and restored upon addition of mannan (likely by mannan binding to ConA). This ConA-mediated inhibition of virus entry and syncytium formation is in agreement with results in human cells using different carbohydrate-binding agents and HIV-1 strains .
The macrophages (BMDM) used in this study represent another category of cells, which express MR transcripts but appear to produce at least two types of VMV receptors, MR and an additional unknown receptor. Mannan, when added to the virus preparation, abrogated ConA effects but only partially in these cells, according to RT activity and provirus quantification. Furthermore, when mannan was added to the cells, it inhibited infection (as it occurred in GSM cells), but only partially. MR has been proposed in different species as a main virus binding site in particular cells  (this might apply to GSM cells used in this study) and/or an auxiliary molecule in other cells (this would apply to BMDM employed in this work). In non-phagocytic cells, MR is not acting as a professional phagocytic receptor, since it does not lead to clearance of virus . Accordingly, in our study the VMV entry into GSM cells via MR subsequently led to a productive infection. However in human cells such as macrophages, results on significance of MR-mediated HIV-1 endocytosis are inconclusive . This may be due in part to the fact that lentiviruses such as HIV-1 bind cells at least via two independent pathways that may coexist in macrophages, the currently accepted infectious route by plasma membrane protein receptors and the route mediated by the endocytic MR, through which HIV-1 epitopes may be subjected to exogenous MHC class I presentation (cross-presentation) . If this applies to the VMV model, the second route would not be present in skin fibroblasts (OSF), as they lacked MR transcripts and were not susceptible to mannan-mediated blocking, but would exist in GSM cells and BMDM.
However, caution should be taken when studying MR expression in BMDM, as it may vary along the individual's life, having implications in pathogenesis. Based on the macrophage phenotype classification in subclinical vs. clinical stages  and according to this and our previous studies , macrophages of SRLV asymptomatic sheep (such as the BMDM tested in this study) would exhibit increased B7 transcript production, whereas those of clinically affected sheep would be expected to have an increased MR expression and viral infection. The known downregulation of B7 molecule expression  and a Th2-biased antibody response to the viral infection  occurring in VMV clinical disease, including arthritis , would be compatible with an upregulation of MR expression in particular target organs such as carpal joints. Antibodies against MR are currently being developed for immunohistochemical studies at different stages of VMV infection.
Besides the cellular receptor, the genetic makeup of the virus may determine the virus-cell interactions, since the number and distribution of ENV amino acids susceptible to mannosylation, may affect viral entry through membrane lectins and consequently the viral production and appearance of disease. Further studies on ENV composition and viral entry are warranted.
In conclusion, we report in vitro studies demonstrating that concerning viral entry there are at least three main patterns in target cells capable of generating a productive infection: i) particular cell types such as synovial membrane (GSM) cells may use MR as a VMV main infection route; ii) other cells such as fibroblasts (OSF) use a route other than MR to bind the glycosylated ENV allowing the virus entry to the cell; and iii) there are cells like macrophages (BMDM), a classical SRLV target, that use MR and an additional receptor for VMV entry. The three cell types may be used as in vitro models to explore the mechanisms and relative relevance of the different entry routes in VMV infections and provide the basis for studies in vivo, on tropism, viral receptors and MR expression aimed to understand viral pathogenesis and host progression from asymptomatic to clinical stages.