Montserrat Serra’s research while affiliated with Autonomous University of Barcelona and other places

Storage mites may be considered important allergens in dogs with atopic dermatitis. High sensitization rates to Tyrophagus , Acarus , and Lepidoglyphus species have been reported in atopic dogs, and dry pet food has been suggested as a potential source of storage mite exposure. The aim of the present study was to evaluate commercial dry dog food for contamination with storage mites, and how storage time and conditions could influence the risk of contamination. Ten different premium commercial dry dog foods formulated for skin disorders were selected. Food bags were opened and stored for 6 weeks under two different environmental conditions. At different time points, samples from each bag were collected and analysed by microscopy, guanine test, storage mite‐specific traps, and a modified flotation technique. On opening, two storage mites identified as Acarus siro were isolated from one of the 10 bags by flotation technique, indicating that storage mites can be present in packaged dry dog food bags. After 5 weeks of storage under environmental conditions optimal for mite growth (23.2 ± 2.1 °C and 71 ± 5.6% of relative humidity), mites were detected by microscopic observation in nine of the 10 diets. When mites were identified by the flotation technique, Tyrophagus spp. were found to be the most common contaminating species. These results show that dry dog food can be a suitable substrate for storage mite reproduction, and that environmental and storage conditions may influence food contamination and mite development.

The development of a complex cellular model, which incorporates the basic cell components of the dog skin, would be a useful tool to investigate the biology and pathology of canine skin and also to replace animal testing partially. The aim of the present study was to develop and characterize a canine skin equivalent. Epidermal keratinocytes and dermal fibroblasts were freshly isolated from skin biopsies from healthy dogs. Fibroblasts were embedded into a bio-matrix from collagen type I matrix protein; this built the scaffold where the keratinocytes were seeded, at air exposed conditions. At 3, 7, 15 and 21 days of culture in special growth media, skin equivalents were analysed by histological, immunohistochemical and electron microscopical techniques. At 15 days, keratinocytes underwent differentiation to a multilayer epidermis with stratum basal, stratum spinosum, stratum granulosum and stratum corneum. Expression of epidermal cytokeratins in keratinocytes was detected by immunhistochemistry, and followed the same pattern than in the normal canine epidermis. Fibroblasts from the skin equivalent expressed vimentin as dermal fibroblasts do. A basement membrane (BM) was observed underneath the epidermis; ultrastructurally, it was similar to the normal canine BM and collagen IV and laminin 5 were detected immunohistochemically as major components of this structure. Skin equivalents developed from canine cutaneous cells presented a similar morphological structure than healthy canine skin. Moreover, the immunohistochemical analysis revealed the expression of the major markers of the epidermis (keratins), dermis (vimentin) and BM (collagen type IV, laminin 5).

To assess binding of IgE to native, whole hydrolyzed, and separated hydrolyzed fractions of soy protein in serum obtained from dogs with experimentally induced soy protein hypersensitivity. 8 naïve Beagles (6 experimentally sensitized to native soy protein and 2 control dogs). 6 dogs were sensitized against soy protein by administration of allergens during a 90-day period. After the sensitization protocol was completed, serum concentrations of soy-specific IgE were measured and intradermal skin tests were performed in all 6 dogs to confirm that the dogs were sensitized against soy protein. Serum samples from each sensitized and control dog underwent western blot analysis to assess the molecular mass band pattern of the different allergenic soy fractions and evaluate reactivities to native and hydrolyzed soy protein. In sera from sensitized dogs, a characteristic band pattern with 2 major bands (approx 75 and 50 kd) and 2 minor bands (approx 31 and 20 kd) was detected, whereas only a diffuse band pattern associated with whole hydrolyzed soy protein was detected in the most reactive dog. Reactivity was evident only for the higher molecular mass peptide fraction. In control dogs, no IgE reaction to native or hydrolyzed soy protein was detected. Data suggest that the binding of soy-specific IgE to the hydrolyzed soy protein used in the study was significantly reduced, compared with binding of soy-specific IgE to the native soy protein, in dogs with experimentally induced soy hypersensitivity.

Versican is a large chondroitin sulfate proteoglycan produced by several tumor cell types, including malignant melanoma, which exists as four different splice variants. The presence of versican in the extracellular matrix plays a role in tumor cell growth, adhesion and migration, which could be altered by altering the ratio between versican isoforms. We have previously shown that overexpression of the V3 isoform of versican in human melanoma cell lines markedly reduces cell growth in vitro and in vivo, since V3-overexpressing (LV3SN) cultured cells as well as primary tumors arising from these cells grow slower than their vector-only counterparts (LXSN). In the present work, we have extended these observations to demonstrate that the delayed cell growth is due to multiple events since differences in proliferative index as well as in apoptosis are observed in LV3SN cells and tumors compared to LXSN. For example, LV3SN melanoma cells exhibit delayed activation of MAPK in response to EGF, we have also characterized further the primary tumors originated in nude mice from V3-transduced melanoma cells to determine if other events affect the V3 tumor phenotype. For example, hyaluronan content of LV3SN tumors was higher than in LXSN tumors, whereas other related matrix components and vascularization were unaffected. Furthermore, lung metastasis in nude mice occurred only in animals carrying LV3SN tumors, indicating a dual role for this molecule, both as an inhibitor of tumor growth and a metastasis inductor.

To assess whether dogs with experimentally induced type I hypersensitivity against soy protein would respond to soy hydrolysate and develop cutaneous or gastrointestinal tract reactions after intradermal and oral challenge exposure. 12 naïve Beagle pups (9 sensitized and 3 control dogs). 9 dogs were sensitized against soy protein by administration of allergens during a 90-day period. After the sensitization period, serum concentrations of soy-specific IgE were determined and an intradermal test was performed to confirm the dogs were sensitized against soy protein. An intradermal challenge test and an oral challenge test with native and hydrolyzed soy protein were conducted on 6 sensitized and 2 control dogs. High serum concentrations of soy-specific IgE and positive results for the intradermal test were observed for the 9 sensitized dogs after completion of the sesitization process. Sensitized dogs challenge exposed with hydrolyzed soy protein had a reduced inflammatory response after intradermal injection and no clinical response after an oral challenge exposure, compared with responses after intradermal and oral challenge exposure with native soy protein. Soy-sensitized dogs did not respond to oral administration of hydrolyzed soy protein. Thus, hydrolyzed soy protein may be useful in diets formulated for the management of dogs with adverse reactions to food.

Versican is a large chondroitin sulfate proteoglycan produced by several tumor cell types, including malignant melanoma. The expression of increased amounts of versican in the extracellular matrix may play a role in tumor cell growth, adhesion and migration. We have expressed the V3 isoform of versican in human and canine melanoma cell lines. Retroviral overexpression of V3 did not change the morphology of any of the cell lines but markedly reduces cell growth in the V3 versican expressing melanoma cells. The V3-overexpressing melanoma cells retain their diminished growth potential in vivo because primary tumors arising from these cell lines growth more slowly than their vector only counterparts. This effect was accompanied by increases in cell adhesion on hyaluronan and an enhanced ability to migrate on hyaluronan-coated transwell chambers. This enhanced migration is blocked when cells are preincubated with soluble hyaluronan, or anti-CD44 antibodies, suggesting that V3 acts by altering the hyaluronan-CD44 interaction. Hyaluronan content and CD44 expression are not altered in V3-overexpressing cells compared to vector-transduced cells. Our results show that V3 overproduction modulates the in vitro behavior of human and canine melanoma cell lines and reduces their tumorigenicity in vivo.

CD44, the main cell surface receptor for hyaluronan (HA), is often overexpressed in tumour cells, and its presence has been related to cell proliferation and migration. Many of the functions of CD44 are mediated through its interaction with hyaluronan. This study investigated the expression of CD44 in CML-1 and CML-10c2 canine melanoma cell lines and melanoma biopsies, and the production of hyaluronan and versican by the canine melanoma cell lines. Versican is an extracellular proteoglycan that binds hyaluronan, forming a tridimensional pericellular coat surrounding the cells. Both canine melanoma cell lines expressed CD44 and produced HA, but only CML-1 produced versican. Cells expressing all three components (CD44, HA and versican) formed abundant extracellular matrices as demonstrated by a particle exclusion assay. CD44 was present within benign and malignant melanomas, but its expression was more intense in malignant melanomas (P < 0.01). In high CD44-expressing tumours, CD44 tended to be present in the periphery of malignant melanomas, whereas its expression was homogeneous in benign melanomas.