Small Demodex populations colonize most parts of the skin of healthy dogs
Background – It is unproven that all dogs harbour Demodex mites in their skin. In fact, several microscopic studies have failed to demonstrate mites in healthy dogs.
Hypothesis/Objectives – Demodex canis is a normal inhabitant of the skin of most, if not all, dogs. This hypothesis was tested using a sensitive real-time PCR to detect Demodex DNA in the skin of dogs.
Animals – One hundred dogs living in a humane society shelter, 20 privately owned and healthy dogs and eight dogs receiving immunosuppressive or antineoplastic therapy.
Methods – Hair samples (250-300 hairs with their hair bulbs) were taken from five or 20 skin locations. A real-time PCR that amplifies a 166 bp sequence of the D. canis chitin synthase gene was used.
Results – The percentage of positive dogs increased with the number of sampling points. When a large canine population was sampled at five cutaneous locations, 18% of dogs were positive for Demodex DNA. When 20 skin locations were sampled, all dogs tested positive for mite DNA. Our study indicates that Demodex colonization of the skin is present in all dogs, independent of age, sex, breed or coat. Nevertheless, the population of mites in a healthy dog appears to be small. Demodex DNA was amplified from all 20 cutaneous points investigated, without statistically significant differences.
Conclusions and clinical importance – Using a real-time PCR technique, Demodex mites, albeit in very low numbers, were found to be normal inhabitants of haired areas of the skin of healthy dogs.
Understanding canine demodicosis remains one of the most difficult challenges in veterinary dermatology. Despite the prevalence and severity of the disease, many aspects of the pathogenesis of this entity remain obscure or poorly documented. Reference textbooks repeatedly make the following two statements:1,2 (i) demodex mites are part of the normal fauna of the dog, and mites are present in the hair follicles of healthy dogs; and (ii) a genetically preprogrammed immunological defect is responsible for the exaggerated replication of mites in demodicosis.1
The ‘fact’ that all dogs harbour Demodex mites in the skin has not been proved using reproducible scientific methods. Most textbooks refer to the classic research done by Gaafar et al., in which Demodex mites were found in the skin of 5.4% of healthy dogs.3,4 The authors concluded that Demodex mites could be found in the skin of healthy dogs and that ‘follicular mange’ is a complex condition, but not that all dogs harbour mites in their skin. More recently, Fondati et al.,5 using trichoscopy, could not detect Demodex canis mites in any of 78 dogs examined, and found a single Demodex injai mite in one dog. Furthermore, aspects such as the preferred anatomical location of the mites on the canine skin, the mite density, and the influence of age and breed on the Demodex mite population remain unknown. Recently, we have developed a highly sensitive real-time PCR to detect D. canis DNA.6 In that study, we were able to amplify the DNA of D. canis in nine of 51 dogs (17.6%), after sampling only two points of the skin (lateral face and interdigital skin).6 Interestingly, this is a much higher percentage than previously reported.7,8
The prevalence of Demodex mites in the skin of healthy humans is close to 100%, with a mean mite density of 0.7 mites/cm2 (facial skin).9–14 Demodex mites are assumed to be normal inhabitants of the skin of most mammals.9,14,15 Therefore, considering data from other species and the biology of Demodex mites, the goal of the present study was to determine whether D. canis mites are normal inhabitants of the skin of most, if not all, dogs or if only a subset of the canine population harbours the mites, and these dogs serve as a reservoir of the parasite.
Materials and methods
One hundred dogs living in a humane society shelter were used in this study. All were adult dogs, ranging in age from 9 months to 13 years, and included dogs of both sexes and of different breeds, with a large majority of dogs being of mixed breed. The shelter housed between 250 and 300 dogs, and only dogs with normal physical and dermatological examinations were used. As canine leishmaniosis is endemic in the region, serological testing was performed on all 100 dogs, and 17 tested positive. Dogs were sampled on three occasions, for experiments 1, 2 and 3. The Board of the Centre gave written permission for the study to be carried out, provided that that all sampling procedures were done under the supervision of the shelter veterinarians.
Twenty privately owned healthy adult dogs presented to the Veterinary Teaching Hospital for preventive medicine examinations and veterinary counselling were included in the study. In addition, eight dogs being treated at the Veterinary Medical Teaching Hospital for different diseases that required immunosuppressive or antineoplastic therapy for more than 2 months were also sampled. None of these eight dogs had skin lesions suggestive of demodicosis. All owners were informed of the nature of the study and gave their written consent.
Hair sampling and DNA extraction
Hair samples (n = 250-300) were obtained by gentle plucking of hair in the direction of the growth so as to include the hair bulb (root) in the sample. Each sample included 250-300 hairs. The number of sampling locations in each experiment is shown in Table 1. Hair samples were maintained in phosphate-buffered saline and stored at -20°C untilDNA extraction. For the DNA extraction, samples were centrifuged in a microcentrifuge at maximal speed for 30 min; once the supernatant was removed, 200 μL of digestion buffer (50 mmol/L Tris-HCl, pH 8.5; and 1 mmol/L EDTA) and 4 μL of proteinase K solution (10 mg/mL) were added, and samples were incubated at 56°C overnight. After inactivation of the proteinase K for 10 min at 95°C, the samples were centrifuged for 10 min at maximal speed. Supernatant was transferred to a new tube and diluted 1:10 for PCR amplification.
Five different sampling experiments were conducted (Table 1). Briefly, in experiment 1, 100 healthy dogs living in an animal shelter had hair samples collected from the following five cutaneous locations: head, dorsal area, foreleg, abdomen and hindleg. In experiment 2, 16 dogs with positive samples from experiment 1 and 30 dogs with negative samples were resampled 6 months after the initial sampling. In experiment 3, five healthy dogs living in the shelter were sampled from the following 20 cutaneous locations: lip and periocular skin (four points), perinasal skin, temporal area, chin, ventral and dorsal neck, dorsum (two points), sternum, abdomen (two points), thigh (two points) and interdigital area (four points, one on each foot). As shelter dogs may not be representative of a normal canine population, five healthy, privately owned dogs were sampled in a similar manner (experiment 4). Finally, in experiment 5, eight dogs receiving immunosuppressive or antineoplastic therapy were sampled from five cutaneous locations.
PCR technique to detect Demodex DNA in canine hairs
The technique has been described in detail elsewhere.6 Real-time PCR was carried out in a final volume of 20 μL using FastStart Universal SYBR Green Master (Roche Diagnostics GmbH, Mannheim, Germany), 0.3 μimol/L of each primer and 4 μL of diluted DNA. Primer pairs used were as follows: D. canis forward, 5′-GATGAAGCGGCGAGTAATGTTC-3′; and D. canis