Leishmania infantum, the etiological agent of canine leishmaniosis (CanL) in Europe, was responsible of the largest outbreak of human leishmaniosis in Spain. The parasite infects and survives within myeloid lineage cells, causing a potentially fatal disease if left untreated. The only treatment option relies on chemotherapy, although immunotherapy strategies are being considered as novel approaches to prevent progression of the disease. To this aim, a deeper characterization of the molecular mechanisms behind the immunopathogenesis of leishmaniosis is necessary. Thus, we evaluated, for the first time, the host immune response during L. infantum infection through transcriptome sequencing of the popliteal lymph nodes aspirates of dogs with CanL. Differential expression and weighted gene co-expression network analyses were performed, resulting in the identification of 5,461 differentially expressed genes (DEGs) and four key modules in sick dogs, compared to controls. As expected, defense response was the highest enriched biological process in the DEGs, with six genes related to immune response against pathogens (CHI3L1, SLPI, ACOD1, CCL5, MPO, BPI) included among the ten most expressed genes; and two of the key co-expression modules were associated with regulation of immune response, which also positively correlated with clinical stage and blood monocyte concentration. In particular, sick dogs displayed significant changes in the expression of Th1, Th2, Th17 and Tr1 cytokines (e. g. TNF-α, IFN-γ, IL-21, IL-17, IL-15), markers of T cell and NK cell exhaustion (e. g. LAG3, CD244, Blimp-1, JUN), and B cell, monocyte and macrophage disrupted functionality (e. g. CD40LG, MAPK4, IL-1R, NLRP3, BCMA). In addition, we found an overexpression of XBP1 and some other genes involved in endoplasmic reticulum stress and the IRE1 branch of the unfolded protein response, as well as one co-expression module associated with these processes, which could be induced by L. infantum to prevent host cell apoptosis and modulate inflammation-induced lymphangiogenesis at lymph nodes. Moreover, 21 lncRNAs were differentially expressed in sick dogs, and one key co-expression module was associated with chromatin organization, suggesting that epigenetic mechanisms could also contribute to dampening host immune response during natural L. infantum infection in the lymph nodes of dogs suffering from clinical leishmaniosis.

SummaryFour loci seem responsible for the dilution of the basic coat colours in horse: Dun (D), Silver Dapple (Z), Champagne (CH) and Cream (C). Apart from the current phenotypes ascribed to these loci, pearl has been described as yet another diluted coat colour in this species. To date, this coat colour seems to segregate only in the Iberian breeds Purebred Spanish horse and Lusitano and has also been described in breeds of Iberian origin, such as Quarter Horses and Paint Horse, where it is referred to as the ‘Barlink Factor’. This phenotype segregates in an autosomal recessive manner and resembles some of the coat colours produced by the champagne CHCH and cream CCr alleles, sometimes being difficult to distinguish among them. The interaction between compound heterozygous for the pearl Cprl and cream CCr alleles makes SLC45A2 the most plausible candidate gene for the pearl phenotype in horses. Our results provide documented evidence for the missense variation in exon 4 [SLC45A2:c.985G>A; SLC45A2:p.(Ala329Thr)] as the causative mutation for the pearl coat colour. In addition, it is most likely involved as well in the cremello, perlino and smoky cream like phenotypes associated with the compound CCr and Cprl heterozygous genotypes (known as cream pearl in the Purebred Spanish horse breed). The characterization of the pearl mutation allows breeders to identify carriers of the Cprl allele and to select this specific coat colour according to personal preferences, market demands or studbook requirements as well as to verify segregation within particular pedigrees.

Insulin deficiency diabetes (IDD) in dogs is an endocrine disease similar to human type 1 diabetes. There are breeds more commonly affected, such as Yorkshire Terrier and Samoyed, suggesting an underlying genetic component. However, the genetic basis for canine diabetes mellitus (DM) is not fully established. We conducted both whole-genome scans for selection signatures and GWASs to compare the genomes of 136 dogs belonging to 29 breeds previously described at low or high risk for developing DM. Candidate variants were tested in dogs with a diagnosis of IDD and controls attending the Complutense Veterinary Teaching Hospital. The only genomic region under selection (CFA8:72 700 000-74 600 000; CanFam3.1) retrieved by our analyses is included in the immunoglobulin heavy chain gene cluster, which has already been related to human human type 1 diabetes susceptibility. This region contains two non-synonymous variants, rs852072969 and rs851728071, showing significant associations with high or low risk for IDD, respectively. The first variant, rs852072969, alters a protein poorly characterised in the dog. In contrast, rs851728071 was predicted to block the synthesis of an immunoglobulin variable (V) domain in breeds at low risk for DM. Although a large and diverse V gene repertoire is thought to offer a fitness advantage, we suggest that rs851728071 prevents the formation of an auto-reactive immunoglobulin V domain probably involved in the pathophysiology of IDD and, thus, decreases the risk for the disease. These results should be interpreted with caution until the functional roles of the proposed variants have been proved in larger studies.