Plenary Session Abstracts

Plenary Session Abstracts: Thursday, 31 AugustTheme: Cutaneous Biology

State of the Art Address

Structure and function of the basement membrane zone


Program in Epithelial Biology, Stanford University, Stanford, California, USA

OVERVIEW Many stratified squamous epithelial tissues such as the skin and oral mucosa contain a complex basement membrane zone (BMZ) composed of a group of specialized components which combine together to form anchoring complexes. At the superior aspect of the BMZ, keratin containing intermediate filaments of the basal cell cytoskeleton insert on basal cell plasma membrane condensations termed hemidesmosomes. Anchoring filaments extend from the basal cell plasma membrane into the extracellular environment and span the lamina lucida connecting hemidesmosomes with the lamina densa and anchoring filaments. At the most inferior aspect of the BMZ, type VII collagen containing anchoring fibrils extend from the lamina densa into the papillary dermis and combine with the lamina densa and anchoring plaques, trapping interstitial collagen fibrils. Thus the cutaneous BMZ connects the extensive basal cell cytoskeletal network with the abundant network of interstitial collagen fibrils in the dermis.

KERATIN FILAMENTS Keratins 5 and 14 assemble together to form intermediate filaments in basal keratinocytes. Keratins contain a central α-helical rod with several nonhelical interruptions as well as nonhelical carboxy and amino terminal regions. The regions of highest conservation between the keratins are located on the ends of the keratin rod in the helix boundary motifs. Keratin intermediate filaments insert upon electron dense structures known as hemidesmosomes.

HEMIDESMOSOMES These structures contain intracellular proteins including plectin and BP230. Plectin (HD1) is a 500-kDa protein that binds intermediate filaments. BP230, also known as BPAG1, is a 230-kDa protein that has homology both to desmoplakin as well as to plectin. BP230, like plectin, functions in the connection between hemidesmosomes and intermediate filaments. Hemidesmosomes also contain the intracellular portions of the transmembrane proteins collagen XVII (BP180) and α6β4 integrin. β4 integrin subunit performs a central role in hemidesmosome formation and contains an especially large cytoplasmic domain, which interacts with other proteins of the hemidesmosomal plaque. Collagen XVII is a transmembrane collagenous protein that interacts with β4 integrin and BP230 intracellularly and with laminin 5 extracellularly.

ANCHORING FILAMENTS These structures contain the extracellular cellular portions of collagen XVII (BP180) and α6β4 integrin. In addition, anchoring filaments contain the molecules laminin 5 and laminin 6. Like all members of the family of laminin proteins, laminin 5 is a large heterotrimeric molecule, containing α3, β3 and γ2 chains. Laminin 5 forms a disulphide bonded attachment to laminin 6, the other known anchoring filament laminin, which contains α3, β1 and γ1 chains. Laminin 5 also forms a strong association with type VII collagen, which serves to connect anchoring filaments with anchoring fibrils.

ANCHORING FIBRILS Type VII collagen is the major component of anchoring fibrils. Type VII collagen contains a large N-terminal globular domain (NC-1) which interacts with laminin 5 in the lamina densa, a long collagenous domain and a smaller C-terminal globular domain (NC-2), which is proteolytically cleaved during anchoring fibril formation. Type VII collagen chains form a triple helix, then two molecules join together in an antiparallel fashion. Anchoring fibrils are then formed by lateral associations of antiparallel dimers. Anchoring fibrils wind around the dermal interstitial collagen fibrils and reinsert back upon the lamina densa, thus attaching the BMZ to the underlying dermis.

Supporting Review

Collagen XVII (BP180, BPAG2) is the most common epidermal basement membrane autoantigen in human beings and animals


College of Veterinary Medicine, North Carolina State University, Raleigh, North Carolina; Program in Epithelial Biology and Department of Dermatology, Stanford University, California; Palo Alto VA Health Care System, Palo Alto, California, USA

Collagen XVII (BP180, BPAG2) is produced by keratinocytes and exists, in vivo, as two distinct forms. The 180 kDa transmembrane form is composed of an intracytoplasmic globular head, bound to integrin α6-β4, plectin and BP230 (BPAG1), and an extracellular rod and a flexible tail which binds to laminin-5 chains. This extracellular segment of collagen XVII is processed by proteolytic cleavage into a soluble 120 kDa protein also known as LAD-1. Collagen XVII is now known to represent the most common skin basement membrane autoantigen in human beings and animals. It is one of the proteins targeted by autoantibodies in patients affected with bullous pemphigoid, mucous membrane (cicatricial) pemphigoid and linear IgA disease. In humans, dogs, cats, pigs and horses, bullous pemphigoid (BP) is an autoimmune blistering disease characterized by skin vesiculation that occurs predominantly during adulthood. Microscopically, there is subepidermal vesiculation associated with neutrophil and eosinophil superficial dermal inflammation. Direct immunofluorescence (IF) reveals linear deposition of IgG and activated complement along the epidermal basement membrane. Indirect IF methods confirm the presence of circulating IgG autoantibodies targeting transmembrane collagen XVII. In humans and animals with BP, the major antigenic epitope is situated in the NC16A domain, a noncollagenous segment situated just outside the keratinocyte’s membrane. Mucous membrane pemphigoid (MMP), previously known as cicatricial pemphigoid, is an autoimmune blistering dermatosis that predominantly affects mucous membranes and mucocutaneous junctions. In humans, dogs and cats, skin lesions consist of vesiculation, ulcerations and occasionally scarring. These lesions are observed in or around the oral cavity, eyes, genitalia or anus, and there is minimal haired skin involvement. Histological examination of skin biopsy specimens reveals ulceration and dermal scarring and/or subepidermal vesicles with few inflammatory cells. Direct IF findings are similar to those of BP. Indirect IF studies establish that circulating IgG autoantibodies target various epitopes on transmembrane collagen XVII (NC16A domain or carboxy-terminus) or laminin-5/laminin-6. Linear IgA disease (LAD) is a rare autoimmune blistering dermatosis identified in humans and dogs. Clinical signs include skin vesiculation and ulceration distributed on the skin and mucous membranes. Microscopic skin lesions consist of subepidermal vesiculation with variable neutrophilic inflammation. Direct and indirect IF studies establish the presence of skin-fixed and circulating IgA (and sometimes IgG) autoantibodies that target LAD-1, the 120 kDa soluble processed extracellular segment of collagen XVII. Additionally, autoantibodies can bind to a 97-kDa degraded form of LAD-1.

Recent studies have confirmed the immunological identity of BP, MMP and LAD in humans and dogs. Because similar segments of collagen XVII are targeted by autoantibodies in human and animal species, they must exhibit unique biochemical properties that render them immunogenic across genera.

Supporting Original Study

Distribution and expression of desmosomal proteins, desmoglein 1 and 3 in canine skin and mucous membrane


Veterinary Medical Teaching Hospital, Gifu University, Gifu; Keio University School of Medicine, Tokyo, Japan

Pemphigus is an autoimmune skin disease in which autoantibodies against desmoglein (Dsg) 1 and Dsg 3 play a pathogenic role in inducing the loss of cell adhesion between keratinocytes. Recently, it has been reported that the expression of desmogleins may relate to the distribution of lesions in PF and PV. In dogs, the target antigens in PF and PV are believed to be Dsg 1 and Dsg 3. Lesions of canine PF usually affect muzzle, ears and foot pad, whereas canine PV patients primarily have mucosal involvement, especially in the oral cavity. These site differences for lesions in canine PF and PV may relate to the distribution and expression of desmogleins. In this study, we investigated the expression of Dsg 1 and Dsg 3 in canine epidermis and mucosal epithelia by immunofluorescence (IF) and immunoblotting. IF studies were performed on normal canine skin from predisposed and nonpredisposed skin and on the mucous membrane in cases of PV and PF. Immunoblot studies were also performed with extracted proteins on the epidermis of the muzzle and abdomen, and mucous membrane of oral cavity, oesophagus, trachea and urinary bladder. Dsg 1 and Dsg 3-specific antibodies were produced from human PV serum by immunoabsorbing with either recombinant human Dsg 1 or Dsg 3 baculoprotein. In IF studies, the expression of Dsg 1 was observed in all stratified squamous epithelia, whereas the expression of Dsg 3 was seen only in the oral cavity, oesophagus and anus. However, neither Dsg 1 nor Dsg 3 was seen in trachea, small intestine and urinary bladder. In skin, anti-Dsg 1 antibody stained the superficial layer most intensely, especially in muzzle and ear, whereas anti-Dsg 3 antibody staining was seen throughout the epithelia in oral cavity, oesophagus and anus. Dsg 1 was also detected in the middle layer of mucous membrane in the oral cavity, oesophagus and anus. In immunoblot studies, the same amount of proteins in each lane were separated by SDS-PAGE, and after immunoblotting with anti-Dsg 1 or Dsg 3 antibody, the density of the visualized bands was analysed with computer-assisted image analysis. The 160-kDa bands detected by anti-Dsg 1 antibody were strong in protein extracts from muzzle skin, but faint in extracts from abdominal skin. The 130-kDa bands detected by anti-Dsg 3 antibody were clearly apparent in proteins oral and esophageal mucosa, but less in the muzzle skin. Therefore, it is conceivable that canine PF mainly affects the muzzle and ears because these areas appear to be rich in Dsg 1 compared with other areas of the skin or mucosal epithelia. On the other hand, the expression of Dsg 3 seemed to be highest in the oral cavity and anus when compared with other epithelia, possibly explaining why these areas are affected in canine PV. This work was supported in part by a Grant-in-Aid from the Ministry of Education, Science, Sports and Culture of Japan (#095566068).

Supporting Original Study

Age- and breed-related differences in canine sebum quality and quantity

R. W. Dunstan, T. H. Herdt, L. Mei, K. M. Credille, R. A. Kennis, R. L. Maier, J. Dziezyc, S. Castle, G. A. Reinhart and G. M. Davenport College of Veterinary Medicine, Texas A & M University, College Station, Texas; College of Veterinary Medicine, Michigan State University, East Lansing, Michigan; The Iams Company, Lewisburg, Ohio, USA

Sebum, the product of sebaceous glands, is composed of various lipids that act as an epidermal and/or follicular lubricant/protectant. Although the lipids produced by dog sebocytes have been defined, factors affecting canine sebum secretion have not been described. The objective of this study was to determine if age and breed influence canine sebum type and quantity. The evaluation was performed on a colony of 18 intact, male dogs (six Siberian huskies, six miniature poodles, and six Labrador retrievers). Starting at 10 weeks of age and concluding when they were 28 weeks old, dogs were sampled at 3-week intervals using the Sebutape method. From these tapes, sebum was reconstituted and thin-layer chromatography (TLC) was performed with results compared against working standards. Sebum was also quantified from the same dogs 6 months after the end of the 18-week study with a Sebumeter SM 810. TLC analysis confirmed the major sebaceous lipids in the dog were cholesteryl esters, wax diesters and cholesterol. Triglycerides and free fatty acids represented minor components. To define breed and time differences, TLC data were reported as a percentage of the total lipids in the sample and as a percentage of the week 0 values. When expressed as a percentage of total lipids, all breeds had essentially the same percentage of total lipids at week 0. There was a significant increase in cholesteryl esters and wax esters as a percentage of total lipids and a significant decrease in the minor lipids in all breeds over the course of the study. There was a significant decrease in cholesterol as a percentage of total lipids in Labrador retriever dogs and miniature poodle dogs. When data were expressed as percentages of the week 0 values, the percentages of cholesteryl esters and wax esters were significantly increased. There also was a significant increase in wax esters as a percentage of week 0 values in Labrador retriever dogs over the course of the study. Triglycerides also increased as a percentage of week 0 levels in all breeds. Fatty acids as a percentage of week 0 values were consistently lower in the Siberian husky dogs from week 3 of the study until week 18. Sebumeter evaluation found that Labrador retriever dogs had the most surface sebum (14 μg cm−2) followed by Siberian husky dogs (12 μg cm−2); both values were much higher than miniature poodle dogs (3 μg cm−2). In summary, based on TLC analyses, total sebum production appears to increase with time in puppies and this increase is predominantly due to wax esters and cholesteryl esters. The decrease in cholesterol in Labrador retriever dogs and miniature poodle dogs as percentages of total lipids and the increase in wax esters in Labrador retriever dogs as a percentage of week 0 values suggest breed differences in sebum composition exist. Sebumeter analysis demonstrated that different breeds also secrete different quantities of sebum. We were intrigued that Labrador retriever dogs, an ‘aquatic breed’, had the highest quantity of surface sebum and that miniature poodle dogs, a breed known for a lack of a ‘dog smell’, had the least. The Iams Company funded this study.

Supporting Original Study

What happens when a dog loses its puppy coat? Functional, developmental and breed-related changes in the canine hair follicle


College of Veterinary Medicine, Texas A & M University, College Station, Texas; Texas A & M Agricultural Research Extension Center, San Angelo, Texas; The Iams Company, Lewisburg, Ohio, USA

This study represents an attempt to define anatomic and functional differences associated with age and breed in a colony of 18 intact, male dogs (six Siberian huskies, six miniature poodles, six Labrador retrievers). Starting at 10 weeks of age and concluding when they were 28 weeks old, these dogs were sampled at 3-week intervals and follicular morphology and function was assessed using the following techniques: (1) examination of site-matched, horizontally sectioned 6 mm skin punch biopsies to define the number of hairs per follicular unit and anagen:telogen ratios; (2) optical-based fibre diametre analysis (OFDA) to determine the diameter and curvature of hair shafts; (3) DermScope examination to evaluate the eruptive morphology of the hair follicles; and (4) measurements of clipped hair to determine the rate of hair growth. For each breed, the period from 10 to 28 weeks of age was associated with an increase in the number of hair shafts per follicular unit: the greatest increase in Siberian husky dogs, the smallest in Labrador retriever dogs. Throughout the study, Siberian husky dogs had the largest number of hairs per follicular unit; miniature poodle dogs had the least. The percentage of telogen hairs increased then decreased slightly for Siberian husky dogs and Labrador retriever dogs, which at 28 weeks of age had 50% and 69% of follicles in telogen, respectively. The percentage of telogen follicles declined in miniature poodle dogs and at 28 weeks of age, 13% of their follicles were in telogen. On OFDA analysis, hair shaft diameter declined for Siberian husky dogs, remained constant for miniature poodles, rose and then declined for Labrador retriever dogs. At 28 weeks of age, the curvature of Siberian husky dog hair approximated that of miniature poodle dog and both these breeds had curlier hair than Labrador retriever dogs. DermScope examination correlated with the histology of horizontally sectioned biopsies and diameter analysis based on OFDA. Of the three breeds, the hair of the Siberian husky dog grew the fastest. These results indicate that there are marked breed differences in follicular morphology based on breed and time. This is especially apparent in the period when the pelage of a dog changes from a puppy to an adult coat. In poodle dogs, this period is characterized by a proportionate increase in the number of large and small diameter hairs of slightly greater curliness. In Siberian husky dogs, this period is characterized by an increase in the number of smaller diameter hairs with slightly greater curliness. In Labrador retriever dogs this period is characterized by an increase in straight, larger diameter hairs. These results also suggest that OFDA analysis of hair shaft curvature and diameter produces a ‘follicular fingerprint’ that can be used as a means of breed identification, and DermScope evaluation may offer a noninvasive means to assess the pattern and severity of alopecic diseases. Finally, the number of hair shafts per follicular unit, and the hair cycle are extremely breed-specific and breed differences must be taken into consideration for accurate assessment of follicular health and disease. The Iams Company funded this study.

Supporting Original Study

Radiography, otoscopy, pneumotoscopy, impedance audiometry, and endoscopy for the diagnosis of otitis media in the dog


College of Veterinary Medicine, The Ohio State University, Columbus, Ohio, USA

The purpose of this study was to assess and correlate results of radiography, otoscopy, pneumotoscopy, impedance audiometry (tympanometry and acoustic reflex testing), and endoscopy with the presence of otitis media diagnosed by myringotomy in dogs with chronic recurrent otitis externa. Forty-five dogs with chronic bilateral otitis externa were entered into the study. Chronic otitis externa was defined as recurrent or continuous external ear infections of at least 6 months’ duration. A diagnosis of otitis media was made on the basis of exudate in the middle ear canal and cytological or cultures evidence of bacterial or yeast infection. Diagnostics performed under anaesthesia included cytology and culture and susceptibility testing from the external ear canal, tympanic bulla radiographs, otic flushing of the external ear canal, otoscopy, pneumotoscopy, tympanometry, acoustic reflex testing, endoscopy, myringotomy, cytology and culture and susceptibility testing from the middle ear, and otic flushing of the middle ear. Sensitivity, specificity, positive predictive value, negative predictive value, and accuracy of radiography, otoscopy, pneumotoscopy, tympanometry, and the acoustic reflex for the diagnosis of otitis media compared to myringotomy were calculated. Radiographic signs suggestive of external ear disease were found in 75/90 (83.3%) ears with clinical otitis externa, which included narrowing of the auditory canal, calcification of the auditory canal, or obliteration of the auditory canal by a soft tissue opacity. Thirteen of 90 (14.4%) ears had radiographic signs of otitis media that included thickening of the bulla, lysis of the bulla, or opacity in the bulla. The tympanic membrane was visualized in 66/87 (75.9%) and 58/87 (66.7%) of ears examined otoscopically and endoscopically, respectively. An effusive otitis media was present in 3/58 (5.2%) middle ears, while the remaining 55 middle ears had a ceruminous exudate. The specificity of radiography, pneumotoscopy, tympanometry, acoustic reflex testing, and otoscopy for the diagnosis of otitis media were 100%, 71.4%, 71.4%, 71.4%, and 100%, respectively, and the positive predictive values for these diagnostic techniques were 100%, 93.5%, 90.9%, 93.8%, and 100%, respectively. The sensitivity and negative predictive values ranged from a high of 57.7% to a low of 16.3% and a high of 18.5% to a low of 9.5%, respectively. Due to the high false negative rate of the five tests evaluated in this study, myringotomy appears to be the optimum technique for the diagnosis of otitis media. However, an abnormal test radiographically, pneumotoscopically, or by impedance audiometry, or a nonintact tympanic membrane are good predictors of otitis media. Supported by Bayer Animal Health and The Ohio State University Canine Research Funds.

Plenary Session Abstracts: Friday morning, 1 SeptemberTheme: Infectious and Parasitic Diseases

State of the Art Address

Changing trends in ectoparasite control


College of Veterinary Medicine, Auburn University, Auburn, Alabama, USA

The decade of the 1990s brought astounding advances in our knowledge, treatment and control of ectoparasitic diseases of companion animals. Treatment is the foremost among these as a result of the introduction of novel ectoparasiticides with improved efficacy. Other improvements include a broadened spectrum of activity, convenience of use, and safety. This has led to increased acceptance of these products by the veterinary community and increased compliance by pet owners. Novel and widely used products have been developed for both dogs and cats and apply to all taxa of ectoparasitic insects and acarines.

Successful flea control has been the primary target of pharmaceutical companies. This has been achieved with topical or systemic adulticidal formulations of phenypyrazole (fipronil), nitroguanidine (imidacloprid), and avermectin (selamectin) compounds. In addition, formulations of ovicidal and/or larvicidal benzoylphenyl urea (lufenuron), pyridine (pyriproxyfen) and terpenoid (methoprene) compounds also have contributed to successful flea control. Use of these novel compounds has revolutionized our approaches to flea control and has rechanneled pet owner interests toward the ethical veterinary product market. Many of the aforementioned compounds, and also ivermectin, moxidectin and milbemycin oxime, also have either label claims or documented ‘off-label’ efficacies against certain ectoparasitic ticks and mites. New agents within one or more of these chemical groups and some novel compounds also are currently in development. These advances also have benefited dermatological specialists in that many parasites and dermatological diseases that were at one time difficult to control are now much less common or have been completely controlled in some regions of the world.

Certain trends are evident in the development of recent ectoparasiticides. Newer products clearly target a broader spectrum of parasites. This has been achieved by the combination of existing agents or by development of new single-entity, broad-spectrum products. Many of the newer products also possess claims against important nematode parasites such as heartworms and intestinal worms. I believe that this trend will continue and will eventually yield products that target a variety of ectoparasites and many helminths, including nematodes, flatworms such as tapeworms, and perhaps some flukes.

One challenge that confronts this industry is the need to develop strategies for deterrence of resistance to available agents. The presumption that we can prevent resistance is likely too optimistic. The recognition that resistance is to some extent inevitable is the first step toward controlling it. A recent industry/academic collaborative monitoring initiative is an excellent first step in dealing with this resistance challenge.

Another challenge that we face as parasite control experts is to quiet the increasingly loud voice of advocates for over-the-counter delivery of ectoparasiticides. It can be argued that many of these products can be used safely by conscientious pet owners and that in many cases ectoparasites are easily recognized and eliminated from pets. However, there are many situations involving ectoparasites and disease that require the knowledge of trained veterinarians and veterinary specialists. Additionally, we in the veterinary community can better monitor efficacy, the development of resistance as mentioned above, noncompliance, and safety of these products if they remain under our purview.

Supporting Review

Pathogenesis of Malassezia dermatitis


Royal Veterinary College, Hawkshead Lane, North Mymms, Hatfield, Herts, UK

It is now widely accepted that Malassezia pachydermatis is an opportunistic pathogen of the dog, with an important role in both otitis externa and dermatitis. However, the pathogenic mechanisms involved in canine Malassezia dermatitis are poorly understood. The outcome of colonization will depend upon the virulence factors expressed by the yeast and the nature of the immunological response in the host. This paper reviews the limited information available with emphasis on studies relating to dogs.

Malassezia pachydermatis is a normal inhabitant of healthy canine skin and mucosae. Dogs with Malassezia dermatitis often have 100–10 000 fold increases in skin population densities in comparison to healthy dogs. Skin and mucosal populations in healthy basset hounds were also found to exceed those of healthy dogs of other breeds. Some atopic dogs have high populations in both affected and unaffected areas. The factors that make the skin of some dogs more favourable to yeast colonization are not clear. However, an in vitro study of adherence showing an inverse relationship between population densities in vivo and corneocyte adherence in vitro, suggests that increased corneocyte receptivity for the yeast is not important in the pathogenesis of Malassezia dermatitis in basset hounds. Features that confer variability in virulence characteristics have not yet been identified amongst strains of M. pachydermatis. In one study, the adherence in vitro of strains isolated from healthy dogs did not vary significantly from strains obtained from dogs with dermatitis. There are differences in protein profiles and enzymatic activity between strains but these have not been directly associated with disease. Malassezia furfur and related lipid-dependent yeasts are the cause of pityriasis versicolor and are implicated in the pathogenesis of seborrhoeic dermatitis and some cases of atopic dermatitis. Immediate and delayed hypersensitivity responses to Malassezia antigens have been reported in humans with atopic dermatitis. Immunoblotting studies have shown that 39 proteins may be stained by IgE. Similarly, immediate intradermal test reactivity to M. pachydermatis has also been observed in atopic dogs. Human patients with pityriasis versicolor showed significantly lower lymphocyte transformation responses to M. furfur. Under certain conditions, the incubation of yeast cells with human peripheral blood mononuclear cells reduced the expression of the pro-inflammatory cytokines IL-6, IL-1β and TNF-α when compared with control cultures. A depressive effect upon pro-inflammatory cytokine production may confer a biological advantage to a saprophytic yeast. In dogs, lymphocyte transformation responses following exposure to M.pachydermatis antigen in vitro in seven healthy basset hounds exceeded those of eight seborrhoeic basset hounds. Serum titres of M. pachydermatis-specific IgG in affected dogs exceeded those of healthy dogs. Western immunoblotting showed that 14 bands of immunoreactivity could be identified, ranging from 219 to 32 kDa. Total serum IgA concentrations in affected dogs were not lower than those of healthy dogs, suggesting that serum IgA deficiency is not a factor in the pathogenesis of the disease. Complex interactions between yeast and skin determine the outcome of colonization. Enhanced understanding is required so that novel therapies and preventative measures can be developed for dermatitis associated with this unusual organism.

Supporting Original Study

Application of the skin-xenograft-mouse model in veterinary dermatology research – modelling canine demodicosis


Ontario Veterinary College, University of Guelph, Canada; College of Veterinary Medicine, University of Florida, Gainesville, Florida, USA

The factors controlling Demodex canis populations on the skin of dogs are not well understood. Clinical observations and in vitro lymphocyte function studies support the current hypothesis that cellular immunity plays an important role in mite elimination and lesion development. Thus, dogs with generalized demodicosis are unable to control mite populations because of a genetic and/or acquired defect in lymphocyte function. However, this hypothesis has not been directly tested. The objectives of this experimental study were: (1) develop a reproducible skin xenograft mouse model of canine demodicosis and (2) test the hypothesis that lymphocytes affect D. canis populations in vivo. Three immunodeficient mouse strains (scid/bg, Tge-26 and RAG-1) were first evaluated for their ability to accept canine skin grafts and support Demodex infection. For Objective 2, RAG-1 knockout mice were xenografted with full-thickness canine skin from a healthy adult dog and infected with 150–250 D. canis mites. Sixty days post-infection, Group I mice were inoculated intragraft with 15×106in vitro stimulated autologous peripheral blood mononuclear cells (PBMCs); Group II received 25×107 unstimulated autologous PBMCs. Infected grafts on Group III mice received saline; uninfected control mice (Group IV) received stimulated PBMCs. Canine PBMCs were isolated by Ficoll-paque® density gradient centrifugation and stimulated in vitro with phytohemagglutinin and human recombinant interleukin-2. Ninety days post-infection, the effect of lymphocyte treatment on skin xenograft mite populations and the presence of inflammation were evaluated by NaOH digests and histology. Xenografts were cultured to assess bacterial colonization. Canine immunoglobulin was quantified by ELISA. Statistical analysis was performed using SAS®. The results were as follows. Objective 1: All three strains of mice supported canine skin grafts and developed productive mite infestations. However, ‘leakiness’ in the scid/bg mice and a wasting syndrome in the Tge-26 mice limited their usefulness for long-term experiments. Objective 2: The average number of mites (per gram of xenograft) in Group I was 67370, in Group II, 23251, and in Group III, 47432. A significant difference in mite counts was observed between treatment Group I vs. treatment Group II (P < 0.05), and between treatment Group I vs. control Group III (P < 0.05). Histological examination of the grafts revealed large numbers of mites within follicles but no accompanying inflammation. Staphylococcus intermedius was recovered from nine skin xenografts. Insignificant quantities of canine immunoglobulin were present in sera from lymphocyte-treated mice. In this study we found that D. canis mites proliferated to high numbers within skin xenografts, directly confirming the importance of systemic dog factors in the control of mite populations. However, heavy D. canis infestations do not induce hair loss in the absence of inflammation. Intragraft inoculation of in vitro stimulated lymphocytes was associated with a significant increase in numbers of mites; possible reasons for this unexpected finding will be discussed. The model could be utilized to evaluate the role of humoral immune factors in the pathogenesis of demodicosis and for drug testing. The methodology applied herein demonstrates the applicability of the skin xenograft mouse model in veterinary dermatology research in general. This research was supported by Morris Animal Foundation, American College of Veterinary Dermatology, European Society of Veterinary Dermatology and Pettrust.

Supporting Original Study

The prevalence of feline staphylococci with special reference to Staphylococcus felis amongst domestic and feral cats in the South-east of England


Royal Veterinary College, Hawkshead Lane, North Mymms, Hatfield, Herts, UK

The prevalence of cutaneous staphylococci (SS) was studied in three groups of cats: feral cats (FC) [n = 10], healthy pet cats (HP) [n = 11] and pet cats with skin disease (LP) [n = 9]. Swab and contact plate specimens were obtained from a total of 13 different sites on each cat and also from lesions when present. SS were identified by cultural characteristics, haemolytic activity, catalase and coagulase reactions, DNase activity and biochemical reactions using commercial API ID32 Staph tests. Staphylococcus felis was identified by comparing API profile numbers of type strains (ATCC 49168 and GD521) and those of recognized sucrose fermenting strains (S36 and S141) to those of the isolates. Matching profile numbers in the ATB 32 STAPH profile index and API database were used to identify the other isolates. There was no significant difference (P > 0.05) in the staphylococcal isolation rate between the feral (49%) and the healthy pet cats (38%), nor between the feral and the affected cats (54%). However, when the affected cats were compared to the healthy cats there was a significant difference (P < 0.05). SS were isolated from at least two sites on every cat sampled. The frequency of isolation from the different sites was: forehead (87%), axillae (62%), anterior nares (60%), groin (60%), external ear canal (27%), interdigital skin (40%) and anal ring (13%). A staphylococcal species was isolated from every lesion sampled. Out of the 398 sites sampled, 187 staphylococcal isolates were obtained and 16 different species were identified. Staphylococcus felis was isolated from all three groups (seven FC, seven HP and three LP) and made up 35% of all the isolates. Staphylococcus intermedius was isolated from all three groups (one FC, three HP and three LP) and made up 20% of the isolates. Staphylococcus aureus (2%), S. chromogenes (4%), S. cohnii (3%), S. epidermidis (2%), S. equorum (1%), S. hominis (2%), S. simulans (7%), S. warneri (4%) and S. xylosus (4%) were isolated from more than one cat. Staphylococcus sciuri (2%) was isolated from three sites sampled on a LP. The remaining four species, S. auricularis (0.5%), S. haemolyticus (0.5%), S. saprophyticus (0.5%) and S. kloosii (0.5%), were each isolated from a single site from four different cats. Twenty-five coagulase-negative isolates (14%) were unidentifiable using the API database and the profile index. There was no significant difference in the isolation rate of S. felis between the three groups (P > 0.05). Although most of the S. intermedius isolates came from multiple sites on lesional cats, the actual number of cats in each group was too low to allow any meaningful statistical analysis. This study suggests that S. felis may be a resident on cat skin and mucosae. Staphylococcus intermedius was the most common pathogenic SS isolated and may play a role in perpetuating skin disease in cats. Acknowledgements are due to Bayer PLC for their financial support, to Dr Igimi for providing type strain GD521 and sucrose fermenting strains S36 and S141 and to Biomerieux, UK, for supplying the API test kits.

Supporting Original Study

Evaluation of the specific immune response in dogs infected by Leishmania infantum


Universitat Autònoma de Barcelona, Spain

The immune system plays a key role in the progression of Leishmania infection in dogs. Infected dogs showing signs of illness have a predominantly humoral, nonprotective immune response which is unable to control the infection (Th2). In contrast, infected dogs with no signs of illness have a predominantly cellular, protective immune response (Th1). The observed differences in response to treatment may reflect differences in the patient’s immune response to the infection. Therefore, immune function testing/profiling may be important to individual patients as a prognostic indicator with respect to response to treatment. The purposes of this study were to characterize the immune profiles of dogs infected with Leishmania and to evaluate usefulness of various immunological tests in the clinical evaluation of infected patients. The following immune function tests were done: anti-Leishmania antibodies (IgG1, IgG2 and total IgGs), cutaneous delayed hypersensitivity testing to leishmanin antigen (DTH), peripheral blood mononuclear cell proliferation assay (PBMC-PA) and production of cytokines (γ-interferon, tumour necrosis factor-α) after stimulation by Leishmania antigen. Three groups of animals were investigated: healthy noninfected dogs (n = 24), infected animals without apparent clinical signs of disease (n = 26) and dogs with obvious clinical signs of leishmaniasis (n = 36 before, during and after treatment). Healthy, noninfected animals had consistently negative test results for all the assays performed anti-Leishmania antibodies, DTH, PBMC-PA, and production of cytokines. Infected animals without clinical disease showed two different immune-profiles: negative or low titres of anti-Leishmania antibodies with positive DTH and positive titres of anti-Leishmania antibodies with negative or weakly positive DTH. The PBMC-PA and the production of cytokines by mononuclear cells were variable in these animals. Before treatment, infected animals with clinical disease had high levels of anti-Leishmania antibodies (mainly IgG2 isotype), negative DTH, and inhibition in PBMC-PA. The production of cytokines (γ-interferon) was low or absent. During and after treatment, infected animals with clinical disease presented a variable immune profile. Some showed a change in the profile, changing from a humoral immune response to a cellular immune response. This change was associated with a clinical improvement and a better prognosis. In summary, we conclude that the combination of serology, DTH and measurement of cytokines constitutes a useful, clinically relevant method to evaluate the immune response to Leishmania. According to our results, dogs living in an endemic area can be categorized as follows: (1) dogs with patent leishmaniasis (Th2-like response); (2) noninfected, healthy dogs; (3) infected but resistant dogs (Th1-like response); and (4) infected dogs with a Th2-like response that will develop the disease.

Supporting Original Study

Local cell recruitment and cytokine production following intradermal injection with Microsporum canis antigen in cats


University of Bristol, Langford, UK

The aim of this experiment was to characterize the nature of the cellular infiltrate, and cytokine mRNA profile at the site of intradermal injection of a soluble Microsporum canis crude antigen preparation (SMCA) in cats that had recovered from Microsporum canis infection.

Eight cats that had recovered from experimental M. canis infection, induced 26 weeks previously, were sedated and given intradermal injections of SMCA and saline (control) at a site distant to previously infected skin. The injection sites were scored for gross signs of inflammation at 30 min, 24, 48 and 72 h after injection. At 48 h after injection 6 mm punch biopsies were taken from the injection sites of four of the cats, immediately after euthanasia. This was repeated for the other four cats, 72 h after injection. Sections were stained using haematoxylin and eosin (HE) for routine microscopic examination. Immunohistochemical (IHC) labelling was performed using antibodies specific for: MHC class II, macrophages and PMNs, B cells, plasma cells (IgG, IgM, IgA), T cells, and T cell subsets (CD4, CD8). The total number of cells in the inflammatory infiltrate (HE slides) and the number of positively labelled cells (IHC slides) were graded using a semiquantitative scoring system (0–4). Biopsies were analysed using semiquantitative RT-PCR assays for mRNA coding for IFN-γ, IL-4, IL-10 and IL-12p40. All the SMCA injection sites had evidence of inflammation at all time points after 30 min, whereas the control sites had scores of zero. Superficial perivascular dermatitis with folliculitis, perifolliculitis and mural interface folliculitis was present in the skin following antigen challenge. In these biopsies at 48 h after injection, eosinophils and mast cells predominated in the infiltrate, whereas at 72 h after injection, macrophages and lymphocytes predominated. Very few cells were labelled with the B cell and plasma cell markers in any sections. For the remaining IHC markers the number of positive cells was higher for the SMCA-injected skin than the control skin. Peak numbers of macrophages were seen 48 h after injection, whereas for the MHC class II, CD3, CD4 and CD8 markers peak numbers of positive cells were seen 72 h after injection. mRNA for IL-4 was not detected in any biopsies. mRNA for IFN-γ, IL-10 and IL-12p40 was detected in both SMCA and control biopsies but was present at higher levels in the SMCA biopsies. The nature of the cellular infiltrate and cytokine mRNA profile in skin following intradermal injection of SMCA in cats recovered from M. canis infection is characteristic of a delayed type hypersensi-tivity response. These findings are consistent with studies of systemic immunity that suggest a strong cell-mediated immune response correlates with recovery from, and resistance to, dermatophytosis in the cat.

Plenary Session Abstracts: Friday afternoon, 1 SeptemberTheme: Dysplasia and Neoplasia

State of the Art Address

Histiocytes and the skin


School of Veterinary Medicine, Department of Pathology, Microbiology, Immunology, University of California-Davis, Davis, California, USA

In 1924 the term histiocyte was introduced as an abbreviation for ‘histiogenic wandering cells’, describing their potential to move within tissue and indicating their putative origin from other mesenchymal cells. Later, the histiocytes were identified as bone marrow derived cells and classified in the ‘mononuclear phagocyte system’ and the ‘dendritic cell (DC) system’. The histiocytic subpopulations – monocytes, macrophages and DC – arise from a common CD34+ precursor. Upon differentiation macrophages and DC develop phenotypic and functional differences and are considered separate cell lineages with partially overlapping functions.

Macrophages reside in close proximity to the microvasculature. As efficient effector cells, they perform key functions of the innate and adaptive immune system and cytokine activation increases their activity. Macrophages bind and phagocytize extracellular proteins and opsonized particles and the engulfed material is digested in phagolysosomes. This process induces secretion of proteases, complement factors, cytokines, chemokines, toxic oxygen metabolites and arachidonate derivatives that enhance inflammation and with it the killing of pathogens. Cutaneous infectious diseases – dermatophytosis, leishmaniasis, mycobacterial infections – are characterized by a predominance of macrophages. Moreover, tissue degenerations – sebaceous adenitis, furunculosis – and deposition of extracellular material – xanthomas, amyloidosis – act as ‘endogenous foreign bodies’ and attract macrophages. Although primarily an effector cells, macrophages can also act as professional antigen presenting cells (APC) and induce a specific immune response.

The DC system consists of phenotypically and functionally diverse cells. Lymphoid DC, which has only been confirmed in mice, guides the development and maintenance of self-tolerance. All three subsets of myeloid DC are found in the skin-associated immune system (SALT). Intra-epidermal DC (Langerhans cells), dermal perivascular (interstitial DC) screen their environment, internalize, and subsequently present antigens to T cells. The third myeloid DC subset is found in the draining lymph node of the SALT. Although capable of phagocytosis, myeloid DC primarily are potent APCs and induce the adaptive immune response. DC express MHC I, MHC II and CD1 and hence, efficiently present processed proteins and gycolipids to T lymphocytes. DC coexpress accessory molecules (B-7 family, CD40) which provide the secondary signal for lymphocyte activation, a critical step for the induction of the specific primary immune response. Small numbers of DC are part of any inflammatory process. Predominance of DC is limited to a small group of skin diseases, including reactive histiocytosis of dogs and cats, other sterile immune-dysregulatory processes of unknown aetiology, reactive DC hyperplasia secondary to neoplasia or benign and malignant histiocytic neoplasia.

Immunophenotyping as a diagnostic tool demonstrates that cells of different origin can mimic morphologic features of histiocytes (example: histiocytic lymphoma). The term ‘histiocytes’ as a morphologic term is therefore used in a broader and less accurate sense. However, accurate identification of lesional cell populations is fundamental to the understanding of a disease process and to achieve a correct diagnosis. Identification of macrophages and DC, and their differentiation from other cells, often require ultrastructural and immuno-phenotypic evaluations.

Supporting Original Study

Maturation states of dendritic cells in canine cutaneous histiocytoma


Centre for Veterinary Science, University of Cambridge, Madingley Road, Cambridge, UK; School of Veterinary Medicine, University of California-Davis, Davis, California, USA

Canine cutaneous histiocytoma (CCH) is a common, benign tumour of Langerhans cells. Langerhans cells are peripheral members of the dendritic cell family of antigen presenting cells. Their function in the skin is to acquire foreign antigens, whereupon they mature and migrate to the regional lymph node and present these foreign antigens to T cells. This tumour undergoes an early period of rapid growth over 1–4 weeks, followed by spontaneous regression, in most cases, over the next 1–3 months. This study investigated the functional characteristics of the cells obtained from clinical cases of CCH. Immunophenotyping of cells obtained from excised tumours, by flow cytometry, revealed expression of CD1a, CD1c, CD11a, CD11c, CD18, CD45, CD45R, MHC-I, MHC-II and E-cadherin, confirming their Langerhans cell origin. In addition, the cells also showed uniform staining with CD1b, and expressed CD11b, CD44, CD49d and ICAM-1, consistent with the CCH cells having an activated phenotype. CCH cells isolated from lesions in the early proliferative phase had lower expression of MHC-I, MHC-II and ICAM-1 and higher expression of E-cadherin compared with those cells isolated from lesions with a lymphocytic infiltrate. This difference is similar to the phenotypic changes that occur during maturation and migration of Langerhans cells from the epidermis to the regional lymph node. CCH cells from 12 patients induced a strong proliferative response in allogenic peripheral blood mononuclear cells (PBMCs), consistent with their dendritic cell origin. CCH cells from early lesions induced a weak proliferative response in autologous PBMCs, and only at high cell numbers. Conversely, CCH cells from later lesions induced a significantly (P < 0.05) stronger proliferative response. The great improvement in the stimulatory capacity for autologous PBMCs is similarly consistent with the hypothesis of maturation of the CCH cells with tumour progression. CCH cells from eight patients exhibited a high rate of macropinocytosis as expected for cells of the myeloid/dendritic lineage. CCH cells from later lesions showed a significantly higher rate of macropinocytosis than those from earlier lesions (P < 0.01). However, these data are not in keeping with the simple progression of functionally normal cells from an immature dendritic cell to a mature dendritic cell phenotype. Immature dendritic cells have a high constitutive rate of macropinocytosis, which is downregulated on maturation induced by cytokines and bacterial products. These data may indicate that the maturation phenomenon suggested by the other data represents a transition from a Langerhans cell precursor to an immature Langerhans cell. Alternatively, the neoplastic cells may not be functionally normal or may not pass through these maturation stages in an orderly fashion. Lesions at the regression stage have a lower proliferative rate and higher apoptotic rate than lesions at the early stage. Apoptotic cells tend to be clustered around infiltrating T cells. These data show that the neoplastic Langerhans cells in canine cutaneous histiocytoma are present at different stages of maturation depending on the clinical stage of the tumour. The different maturation states of the neoplastic cells during tumour progression and regression reflect the natural life cycle of normal Langerhans cells.

Supporting Original Study

Use of monoclonal antibodies as an adjunct to the diagnosis of epithelial skin tumors of the dog and cat using the EnVision+® system for the immunohistochemical detection of tissue antigens


School of Veterinary Medicine, University of Pennsylvania, Philadelphia, Pennsylvania, USA

In 1999 the second series of Histological Classification of Epithelial and Melanocytic Tumors of the Skin of Domestic Animals, which defined specific histological criteria for skin tumours, was published by the World Health Organization. Most skin tumours of the dog and cat can be identified on the basis of these defined criteria. Cytokeratin (CK) is the intermediate filament that forms the cytoskeleton of epithelial cells. CKs are classified according to their molecular weight and isoelectric point and numbered. The CKs expressed by various epithelia will depend on the type of cell, their stage of development and the environment in which they develop. CKs are therefore useful markers to evaluate epithelial cell differentiation. This study evaluated the immunohistochemical staining of normal skin and adnexa and skin tumours in the dog and cat and the value of this procedure to ascertain the definitive diagnosis for these tumours. The immunohistochemical evaluation was carried out using the DAKO® EnVision™+ System and the DAKO Autostainer® to standardize the procedure and reduce technician error. Routinely processed, paraffin embedded tissues were cut at 4 microns and mounted on ProbeOn Plus® microscope slides. The tissues were deparaffinized. Several antigen retrieval procedures were undertaken to maximize staining intensity. Primary antibodies used were AE1 (CK 10, 14, 15, 16, 19), AE3 (CK 1–8), CAM 5.2 (CK 8, 18), CK 15 (stem cells), Calponin (myoepithelium) and CD79a (B cells). The normal Avidin-Biotin-Complex (ABC) method was replaced by a dextran polymer conjugated with HRP and goat antimouse secondary antibodies (EnVision™+ System)® for all monoclonal antibodies, except CD79a, where a modified ABC method, Labelled Streptavidin-Biotin 2 System (LSAB2)®, was used. 3,3′-Diaminobenzidine tetrahydrochloride (DAB) was utilized as the chromogen. We were able to identify the following by their specific immunohistochemical staining in routinely processed normal dog and cat skin – apocrine glandular and myoepithelial cells, apocrine duct epithelium, sebaceous glandular and duct epithelium, internal root sheath of hair follicles, Merkle cells – and to identify those skin tumours showing differentiation to these cell types. This is of importance from a clinical perspective, where these tumours exhibit different biological behaviours and require different postoperative therapy. We were able to differentiate apocrine carcinoma from sebaceous carcinoma and hepatoid gland carcinoma in dogs from anal sac gland carcinoma in the dog. Furthermore, we observed differences in the staining affinity of the normal epidermis of the dog and cat, which are most likely due to differences in cytokeratin expression within epidermal keratinocytes. We believe that the immunohistochemical comparative studies with normal canine and feline epidermis and skin appendages and their skin tumours we have performed are of help in establishing the histogenesis of these skin tumours in the dog and cat.

Plenary Session Abstracts: Saturday, 2 SeptemberTheme: Immunological Diseases

State of the Art Address

Atopic dermatitis: basic and comparative observations


Oregon Health Sciences University, Department of Dermatology, Portland, Oregon, USA

Atopic dermatitis (AD) is a common, chronic inflammatory skin disease affecting roughly 10% of humans, usually beginning during infancy and often continuing into adulthood. The frequent coexistence of allergic rhinoconjunctivitis, asthma and IgE reactivity has traditionally focused research on allergic pathogenic mechanisms, although clinically relevant associations with food and airborne allergens have never been convincingly demonstrated on any consistent basis. Clinically, the hallmarks of AD include severe itching and eczematous inflammation, leading to the secondary features of excoriation and lichenification. Histology shows epidermal oedema (spongiosis) and an infiltrate characterized by T cells and eosinophils. The immunological and pharmacophysiological features of atopic dermatitis have stimulated research seeking to identify relevant effector cells and mediators that characterize the chronic skin inflammation. AD can be transferred by bone marrow cells, which carry the genetic atopic abnormalities and are precursors to those cells that infiltrate the skin. Underlying mechanisms may include abnormalities in cyclic nucleotide regulation of marrow-derived cell function or the allergenic over-stimulation that leads to clinical expression. The primacy of one mechanism over the other remains unresolved, but this does not obviate their values in suggesting novel therapeutic targets such as phosphodiesterase inhibition and immune suppression. Recent clinical trials in human AD are revealing new topical agents that provide potent anti-inflammatory activity beyond current corticosteroid limitations. Canine AD appears to be quite comparable to the human disease. We have been impressed with the AD-like condition in a colony of Basenji-Greyhound dogs. Likewise, studies at North-western University of Pennsylvania helped characterize the dog model of atopy. However, the size and variability of the species along with the relative lack of immunological information and reagents (especially compared to murine or rodent species) have impeded development of a dog model. Through the years, other models have been proposed but the general failure of many of these prompts an historical caution in accepting inadequately proven systems, for example, the erroneously diagnosed ‘eczema’ in rats fed a corn oil diet. This led to the widespread misconception that AD could be corrected by evening primrose oil, though in reality they had a scaly skin disease due to essential fatty acid-deficiency. In the past decade, investigators have caused chronic eczematous conditions along with increased IgE antibodies in mice repeatedly challenged with DNCB and ovalbumin and in NC/Nga mice raised in nongerm-free conditions, all models worth pursuing.

Each of these systems has shortcomings and there is little assurance that they are truly identical with AD. However, the development of a useful animal model must remain a high priority for a disease that has high impact for human and veterinary medical efforts.

Supporting Review

Mast cells: a review of their biology and role in cutaneous inflammation


Royal (Dick) School of Veterinary Studies, Roslin, Midlothian, UK

Mast cells are normal residents of connective tissue and are found in highest numbers in areas of the body that interface with the environment, such as the skin, lung and gastrointestinal tract. In contrast to other cells of the haematopoietic stem cell lineage, which differentiate in the bone marrow before being released into the circulation, mast cells do not circulate as mature cells. Morphologically unidentifiable precursors migrate in the blood and invade connective or mucosal tissues where they proliferate and differentiate into mature mast cells. The circulating precursor of the mast cell possesses cytoplasmic granules, expresses RNA encoding mast cell proteases, expresses the c-kit receptor but lacks the IgE receptor. The differentiation and proliferation of mature mast cells from these progenitor cells is regulated by fibroblast-derived stem cell factor and T-cell-derived interleukins-3, 4, 9 and 10. Stem cell factor is critical for the constitutive development of mast cells because mice that lack it (or its receptor, c-kit) are almost totally deficient in mast cells. Furthermore, stem cell factor is a chemotactic agent for mast cells, influences the adhesion of mast cells to the dermis, regulates the synthesis of mast cell mediators and modulates their secretory function. Interleukin-3 (along with interleukins 4, 9 and 10) is more important in the proliferation of mast cells seen in gastrointestinal nematode infections but is also likely to be involved in the proliferation of skin mast cells. Under the influence of these cytokine systems, immature mast cells migrate from the circulation into the connective tissue underlying epithelial surfaces and develop into distinct functional phenotypes with differing biochemical properties, especially with regard to their content of granule proteoglycans and proteases. Once in the peripheral tissues, mast cells play an important role in the defence of the body against parasites and other infectious agents, but they are also involved in more general immune responses and wound healing. Mast cells exert their effects by synthesizing and releasing a host of inflammatory mediators such as histamine, proteases and cytokines. This process is usually triggered via the interaction between allergens and allergen-specific IgE bound to high affinity FcR1 receptors expressed on the mast cell surface. Activation of FcR1 receptors results in mediator release through a complex series of biochemical reactions involving G proteins, tyrosine kinases, inositol-trisphosphate, protein kinase C, calcium channel activation and rearrangement of the cytoskeleton, allowing contact between the mast cell granules and the plasma membrane. Once released into the intercellular milieu, the granule contents participate in an interplay between the microvasculature and other inflammatory cells that results in all the cardinal signs of allergic inflammation. Although intended to be protective, this process is often directed towards harmless environmental antigens and contributes to the development of allergic diseases such as atopic dermatitis.

Supporting Review

Feline atopic dermatitis: a review


Faculty of Veterinary Medicine, University of Bern, Bern; Switzerland; University Medical Center, Utrecht University, Utrecht, The Netherlands; Faculty of Veterinary Medicine, Utrecht University, Utrecht, The Netherlands

Atopic dermatitis (AD) in cats was first described in 1982. As in humans, it is a pruritic and chronic skin disease characterized by remissions and exacerbations that often starts at a young age. Skin lesions in cats with AD seem to have a more variable distribution compared to humans and dogs. Described cutaneous reaction patterns include self-induced alopecia, miliary dermatitis, eosinophilic granuloma complex lesions and pruritus of the head, neck and pinnae. These different clinical diseases share a histopathological reaction pattern with dermal infiltration of mast cells, eosinophils, lymphocytes and macrophages. Atopy in humans is defined as an inherited predisposition to develop allergen-specific IgE antibodies in response to environmental allergens. Comparable to humans, immediate skin test reactivity is present in cats as well. Evidence, however, for the existence of a putative feline IgE was gathered only over the last few years. It was shown that passive transfer of reaginic antibodies in cats is possible and that this reaction is heat sensitive. Recently, a novel polyclonal antibody against feline IgE was developed. In humans, IgE binding on epidermal Langerhans cells (LCs) is thought to facilitate allergen-specific T cell activation. After capture the allergen is processed and presented to T cells in a MHC class II restricted way. The epidermis and dermis of lesional skin of cats with AD contained a significantly increased number of CD1a+ cells (LCs) and MHC class II+ cells compared to skin of healthy control animals. T cells are important role players in the pathogenesis of AD and histological examination of both human and feline AD skin reveals a predominance of CD4+ T cells. More specifically, an increased number of IL4+ T cells was demonstrated in lesional and nonlesional skin of cats with AD. With double labelling methods it was demonstrated that the IL4 was primarily produced by T cells and not by mast cells. The atopy patch test was proven to be a valuable model for atopic dermatitis in humans. An atopy patch test was developed in cats in an attempt to study the allergic inflammation process. In a small study, 3/6 cats with AD showed erythematous skin reactions to topically applied aeroallergens in contrast to healthy control cats that did not have any reaction. With immunohistochemistry a significant increased number of positive staining cells for IL4, CD3, CD1a, and MHC class II was demonstrated. In conclusion, AD in cats is still not a well-defined disease; however, initial findings do not contradict a pathogenesis that is comparable to what is described in humans.

Supporting Original Study

Recombinant chimeric canine anti-IgE monoclonal antibody therapy removes free and total serum IgE and eliminates skin test reactivity in ragweed sensitized dogs


Department of Molecular and Cellular Biology, Idexx Laboratories Inc., Westbrook, Maine, USA

A chimeric canine anti-IgE monoclonal antibody was constructed from a mouse anti-IgE monoclonal antibody produced against native canine IgE. In vitro characterization experiments demonstrated a very high affinity for canine IgE by the mouse monoclonal antibody. Furthermore, it was determined that the mouse anti-IgE monoclonal prevented canine IgE from binding to its high affinity receptor. The mouse heavy and light chain variable regions containing the IgE binding site were fused to a canine constant IgG heavy chain domain. The resulting chimeric monoclonal antibody was expressed as a recombinant protein in a baculovirus expression system. A feasibility study was conducted on a small population of dogs to test the ability of the recombinant chimerized canine anti-IgE monoclonal antibody to bind canine IgE in vivo and affect an allergic response. Five dogs were experimentally sensitized to ragweed antigen beginning at birth. These dogs subsequently produced a positive intradermal skin test response when challenged with ragweed antigen. Three dogs received the anti-IgE therapy and two control dogs remained untreated. Pre-experimental levels of free and total serum IgE were determined using a sensitive ELISA assay. In the course of a 35-day treatment protocol, free IgE was reduced to undetectable levels within 1 h of the initial treatment while total IgE was reduced to undetectable levels over a 2-week period. IgE levels remained undetectable in excess of 60 days post-treatment. There was no change in either free or total IgE in the untreated dogs. In the dogs receiving anti-IgE monoclonal antibody therapy there was a reduction of high affinity IgE receptor expression on circulating basophils determined by flow cytometry. Synthesis of new IgE could not be demonstrated in the treated dogs based on in vitro experiments using isolated PBL from treated and untreated dogs. This result was supported by the fact that no IgE-Chimeric antibody immune complexes could be detected in the treated dogs while free chimeric antibody was clearly present. Additionally, in two post-treatment intradermal skin tests conducted 1 week apart there was no response to ragweed challenge in the treated dogs while skin test sensitivity to ragweed was unchanged in the control dogs. These experiments demonstrate the feasibility of using anti-IgE monoclonal antibody therapy for treatment of canine atopic dermatitis. The model that has been developed also presents a system to study the biology of the allergic response. This work was supported by IDEXX Laboratories.

Supporting Original Study

Effects of PDE4-inhibitors in a model of allergic dermatitis


Institute for Pharmacology, Toxicology and Pharmacy, School of Veterinary Medicine, Hannover, Germany

Allergic dermatitis may be an indication for the use of phosphodiesterase 4 (PDE4)-inhibitors, since a very high PDE4-activity was noted in monocytes and lymphocytes of patients with this chronic immunological disease. Moreover, methylxanthines (e.g. theophylline or pentoxifylline) that nonspecifically inhibit PDE, are used as anti-inflammatory agents in allergic diseases like bronchial asthma in humans. The clinical effects of PDE4-isoenzyme specific inhibitors in atopic dermatitis have been shown. RPR-73401 (Piclamilast) and SB-207499 (Ariflo) are novel inhibitors of the rolipram-sensitive cAMP-specific PDE4. The purpose of this study was to investigate the effects of RPR-73401 and SB-207499 on an experimentally induced allergic skin reaction and compare them to cyclosporin A and dexamethasone (positive controls). In this study, the mouse ear swelling test (MEST) using BALB/c mice (n = 6 per group) was used. To simulate an allergic inflammatory skin reaction, mice were sensitized to 1-chloro-2,4-dinitrobenzene (DNCB) or toluene-2,4-diisocyanate (TDI). After shaving the abdominal skin and one intracutaneous injection of 50 μL Freund’s adjuvant, the horny layer was stripped with adhesive tape and 100 μL 0.5% DNCB or 5% TDI were applied to the epidermis for 4 days. After a sufficient wash out phase, 20 μL 1% DNCB or 0.5% TDI were applied to the ears as a challenge during each of the following treatment experiments. Ear swelling by oedema formation was measured as a clinical parameter of the inflammatory reaction. After repeated topical treatment of the ear skin with 20 μL 3% RPR-73401 or intraperitoneal (IP) injection of RPR-73401 (1 and 5 mg kg−1), DNCB or TDI were applied as a challenge. Subcutaneous treatment with cyclosporin A (30 mg kg−1) or IP injection of dexamethasone (1 mg kg−1) served as positive controls. Vehicle controls showed a high increase in ear thickness over 48 h after challenge. In contrast, RPR-73401 treated mice (topical or IP) significantly decreased the ear swelling. Topical administration of RPR-73401 had a longer lasting effect. Sensitized mice were also treated IP with SB-207499 (1–30 mg kg−1) two times in 24 h or five times in 12 h. In another group, SB-207499 was given orally 10 mg kg−1 once or three times in 24 h. Local treatment of the ear with diflorasone diacetate (20 μL 0.05%) served as a positive control. Again, vehicle controls showed a significant increase in ear thickness over 48 h after challenge. This swelling was significantly reduced after DNCB-challenge by five times IP treatment with at least 3 mg kg−1 SB-207499. A single oral dose of SB-207499 was more inhibitory in TDI challenged compared to DNCB challenged mice. However, repeated treatment of TDI-sensitized mice (three oral doses in 24 h) resulted in an even higher reduction in ear thickness compared to a single treatment. In conclusion, based on the results of this animal model, the results suggest that PDE4-inhibitors may be useful in treatment of immunological dermatitis. Supported by a grant of the Saechsische Aufbaubank, Dresden, Germany (Project-No.4236).

Supporting Original Study

IgG responses to Malassezia pachydermatis antigens in atopic and normal dogs


Royal (Dick) School of Veterinary Studies, University of Edinburgh, Edinburgh, UK

Dermatitis caused by excessive proliferation of the yeast Malassezia pachydermatis is an increasingly recognized condition in canine practice but the interaction between the organism and the host’s immune system has not been extensively investigated. A common underlying cause of Malassezia overgrowth is atopic dermatitis, a disorder in which immunological abnormalities are known to occur. The aim of this study was to characterize one arm of the humoral immune response to M. pachydermatis by comparing IgG responses to separated antigens of the yeast in atopic dogs with and without Malassezia dermatitis, and normal dogs. Sera were collected from 18 atopic dogs with Malassezia dermatitis, 20 atopic dogs without Malassezia dermatitis, and 18 clinically normal dogs. The diagnosis of atopic dermatitis was based on a combination of consistent history and clinical signs, exclusion of other causes of pruritic skin disease and the presence of at least one positive intradermal skin test reaction as assessed by standard criteria. Malassezia dermatitis was diagnosed by microscopical observation of elevated cutaneous M. pachydermatis populations on Diff-Quik® stained tape strips sampled from various anatomical locations. A sample of M. pachydermatis was obtained from the ear canal of a dog with Malassezia overgrowth and cultured on Sabouraud’s agar. Malassezia proteins were extracted by mechanical disruption of the colonies with microglass beads and separated using sodium dodecyl sulphate polyacrylamide gel electrophoresis (SDS-PAGE). The proteins were then transferred onto polyvinylidene fluoride microporous membranes that were sequentially incubated with 1% skimmed milk, individual sera, horseradish peroxidase conjugated goat antidog IgG, and substrate (3, 3′-diaminobenzidine). The molecular weights of IgG binding proteins were determined using standard curves generated by analysing the position of molecular weight markers on Coomassie blue stained membranes. Numerous IgG-binding proteins were detected with sera from atopic dogs with Malassezia dermatitis. Strong bands representing the major antigens were seen at molecular weights of 25 kDa (in 78% of dogs), 29 (67%), 42 (94%), 45–51 (78%), 61 (67%), 82 (89%), and 90 kDa (61%). However, antigens of 42 (95%), 45–51 (80%), 61 (50%), and 82 kDa (65%) were also recognized by most atopic dogs without cytological evidence of Malassezia dermatitis, although to a lesser extent. Approximately half of the normal dogs showed faint IgG binding to multiple indistinct bands between 42 and 68 kDa, but clearer bands could be seen at 42 (83%), 45 (61%), and 82 kDa (67%). These results suggest that most dogs have evidence of an IgG response to multiple antigens of M. pachydermatis. However, the majority of atopic dogs with Malassezia dermatitis had a greater IgG response and showed binding to two additional proteins of 25 and 29 kDa, which were not recognized by most dogs in the other two groups. These proteins may therefore represent clinically important antigens in cases of Malassezia dermatitis associated with canine atopy.