Encephalitozoon cuniculi Infections in Dogs: A Case Series

Introduction

The obligate intracellular parasite, Encephalitozoon (E.) cuniculi, belongs to the phylum Microspora. The organism is best known as a parasite of rabbits and rodents, where it typically causes subclinical infections; but the parasite may cause encephalitis or chronic renal disease in rabbits on a sporadic basis.^1,2 Occasionally E. cuniculi causes overt disease in dogs and foxes, with sporadic cases reported in other mammals.^1,2 This parasite also has been rarely reported as a cause of systemic, often fatal disease in humans who are positive for human immunodeficiency virus.^3–7

The parasite has a direct life cycle, and microsporidial transmission typically occurs when infective, environmentally resistant spores are shed in the urine or feces of an infected host and are subsequently ingested by a susceptible host. Transplacental transmission of E. cuniculi is well documented in domestic rabbits and has been demonstrated experimentally in dogs.^2,8

A limited number of reports describe natural E. cuniculi infections in dogs. In the 1950s and 1960s, an encephalitis-nephritis syndrome was described in puppies in England and South Africa, and parasites were histologically identified as E. cuniculi.^9–11 Chronic nephritis resulting in renal failure has also been reported in young adult dogs in Africa.^12,13 Only two case reports have documented the parasite in dogs in the United States (US), and those reports are more than 25 years old.^14,15

In addition to those case reports, a limited number of serological surveys have been done with widely varying detection of antiparasite anti-bodies in various dog populations. Studies in Norway, England, South Africa, and Slovakia have reported seroprevalence rates of 0% (none of 1104 dogs), 13.3% (33 of 248 dogs), 18% (40 of 220 dogs), and 37.8% (73 of 193 dogs), respectively, in surveys of dogs using convenience sampling. ^16–19 In an additional study in South Africa that focused on dogs with primary renal disease, 23% (12 of 52 dogs) were seropositive, compared with a 5% (two of 42 dogs) seropositivity rate in control dogs.²⁰ Those serological findings suggest that parasite exposure or infection is more common than clinical reports indicate. Positive serological tests in clinically normal dogs also suggest that asymptomatic infections may occur that are similar to infection patterns widely recognized in clinically normal rabbits and rodents, where chronic infection and parasite shedding in feces or urine have been documented.²

Recently, molecular tools have provided additional information on this parasite. Using molecular characterization of E. cuniculi isolates from various hosts, subtle differences have been found in the internal transcribed spacer (ITS) region of the ribosomal ribonucleic acid (rRNA) gene, which classifies rabbit, rodent, and dog isolates into three genotypes named I, II, and III, respectively.^3,21 Human isolates from the US, Europe, and Japan have been characterized as the dog genotype III, while other reports from Europe relate human isolates to the rabbit genotype I.^3–7,21 Although direct animal-to-human transmission of E. cuniculi has not yet been documented, this parasite should be considered as a pathogen with zoonotic potential.

Since E. cuniculi has not been widely documented as a canine pathogen in the US, we are reporting a series of cases of encephalitozoonosis in puppies and one case in a young adult dog.

Materials and Methods

In this retrospective study, medical records were reviewed for dogs that had confirmed E. cuniculi infections based on postmortem evaluations and molecular analyses.

Results

The bodies of 19 dogs from 12 kennels were evaluated in this study [see Table]. The cases included purebred dogs of eight toy breeds. With two exceptions, the dogs ranged in age from 4 to 10 weeks and included 11 females and six males (gender was not recorded for one puppy). Body weights were available for 11 dogs, and all were ≤1 kg. Body conditions of most puppies were reported as thin or underweight. Seventeen puppies had histories of progressive neurological disease ranging from 4 days to 2 weeks in duration. Commonly reported clinical signs included depression, inappetence, weight loss, ataxia, hypermetria, central blindness, and circling. In most cases, the puppies died naturally; but several puppies were euthanized when moribund. When multiple puppies from the same litter were infected, clinical signs and subsequent deaths were clustered within a few days. In at least two cases (kennels 1 and 2), some puppies within the litter were fatally infected while others remained clinically normal. In a single case, a 5- month-old, female Yorkshire terrier had a history of progressive wasting and neurological disease for an extended duration of approximately 3 months before death [kennel 5; see Table].

Histopathological findings were similar in all puppies, and lesions were consistent with those previously described as constituting an “encephalitis-nephritis” syndrome.^1,10,14 The kidney was a major organ in which lesions were found consistently [Figure 1A]. Renal lesions were characterized by disruption of normal cortical and corticomedullary architecture with extensive interstitial collections of lymphocytes and plasma cells, as well as many macrophages and occasional focal hemorrhage or necrosis. Small (1 × 2.5 μm), Gram-positive, plump rod-shaped organisms—consistent morphologically with E. cuniculi—were widely distributed in the inflammatory foci or in the cytoplasm of renal tubular epithelial cells.

The brain was the second organ where significant and consistent lesions were found [Figure 1B]. Lesions were present in both gray and white matter without detectable regional distribution. A diffuse, lymphocytic and plasmacytic encephalitis was seen with a distinct but not exclusively perivascular distribution. Perivascular cuffs of lymphocytes, plasma cells, and some macrophages also were present. Numerous organisms were in the brain parenchyma—most often in the vascular endothelial cell cytoplasm and infrequently within the cytoplasm of neurons. Focal cellular nodules comprised of lymphocytes, macrophages, and glial cells (some with small foci of central necrosis) were scattered randomly in all affected areas of the brain. Lesions seen in other tissues of some puppies included nonsuppurative hepatitis, gastritis, myocarditis, pulmonary edema, and mild interstitial pneumonia. Encephalitozoon spores were infrequently identified in those tissues.

Encephalitozoonosis was also diagnosed in one young adult, female dachshund [kennel 12; see Table]. The dog was reported by the owner as a “poor doer” with small body size and persistent thin body condition. At 18 months of age, the dachshund was presented to a local veterinary practice in severely depressed and dehydrated condition; euthanasia was performed based on a diagnosis of end-stage renal disease. The ante-mortem diagnostic test results for this case were not available.

In the postmortem gross and histological evaluation of the dachshund, lesions of chronic end-stage renal disease were observed. Severe interstitial fibrosis was evident, accompanied by a loss of renal tubular epithelium, tubular dilatation, destruction of tubules, focal tubular regeneration, and glomerulosclerosis. Occasional focal collections of Gram-positive microsporidial organisms were observed.

In cases for which additional serodiagnostic and/or microbiological tests for other infectious diseases were performed at the TVMDL at the request of the submitting veterinarian, no significant additional findings were reported.

Based on information from referring veterinarians and dog owners, additional cases of encephalitozoonosis were likely associated with these documented cases. Puppy deaths in at least 11 litters from six kennels (kennels 1, 2, 3, 7, 8, 10), totaling at least 23 additional puppies, were anecdotally reported. No detailed histopathological materials or molecular analyses were available to confirm a diagnosis of encephalitozoonosis in these animals; however, clinical histories and physical and temporal proximity of these puppies to confirmed cases suggest that these puppies also died of the parasitic infection.

In most cases, the infected dogs had some history of association with a large group of dogs. Nine of the reported cases were submitted from breeding kennels (kennels 2, 3, 7, 9, 10, 12), and two additional cases were fatally infected puppies that were recently obtained from large breeding kennels [kennels 4, 6; see Table]. Kennel 1 (five cases) had recently purchased a brood bitch (dam of one of the infected litters) from a large kennel. The source of the puppy was unavailable for three cases (kennels 5, 8, 11).

A total of nine parasite isolates were established in tissue culture. In four cases, DNA was extracted directly from homogenized brain or kidney tissue; in four cases, DNA was extracted from spores purified from the primary parasite cultures established in vitro; and in five cases, parasite DNA was sequenced from both tissues and cultured spores.

Molecular data were generated on rRNA genes of parasites from 13 dogs. In eight cases, sequences of the 3′ end of the small subunit (SSU) rRNA gene (GenBank accession numbers AF144246-AF144253) ranged from 252 to 269 bases in length. In eight cases, sequences from the 5′ end of the SSU and ITS regions of rDNA (GenBank accession numbers EU001235-EU001242) ranged from 429 to 586 bases in length.^3,21 For three cases, both gene regions were sequenced. All sequences showed 100% homology to each other and to E. cuniculi genotype III sequences for previously reported isolates in GenBank (accession numbers AJ005581, L29560).

Discussion

In this case series, 19 cases of canine encephalitozoonosis were documented using histopathological and molecular methods. Anecdotal clinical information from littermates and other litters at several kennels suggests that at least 23 additional puppies probably died from this parasitic infection. Typical clinical signs at presentation included depression and the development of progressive neurological abnormalities just before or after weaning.

While clinical disease from E. cuniculi is uncommon in dogs, the pathogen may be underdiagnosed, particularly in young puppies, for several reasons. Mortality in unweaned puppies is not frequently investigated by owners, so a cause of death is never determined. Historically, neurological disease in puppies was often attributed to canine distemper or other viral infections, and appropriate diagnostic testing was not available at the time. Currently, encephalitozoonosis is a rather unfamiliar disease in dogs, and the appropriate diagnostic tests are not easily available to aid in antemortem diagnosis of the infection. Serodiagnostic testing for canine antibodies to the parasite is possible, but few veterinary diagnostic laboratories provide that service.¹ Indirect immunofluorescent assays and enzyme-linked immunosorbent assays have been used experimentally to measure canine antibody titers in serological surveys, but those assays are not commercially available.^16–20 Several companies that provide diagnostic services for research laboratory animals conduct serological screening for rabbits and rodents on a fee-for-service basis, but they do not offer routine canine serological screening tests. One concern in using this method for diagnosing individual cases is the questionable reliability of antibody testing in immunologically immature animals, such as the young puppies identified in this study. Therefore, diagnostic methods that directly identify microsporidian spores in urine sediment, feces, or tissues should be developed further.

Limited experimental data show that parasite spore shedding can be detected antemortem in fecal and/or urine samples from infected dogs (data not shown). Several staining methods have been developed for routine use in medical diagnostic laboratories to diagnose human cases of microsporidiosis.^1,23,24 Chitin-staining fluorochromes, such as Calcofluor White fluorescent stain^e or Fungi-Fluor stain,^f have been used for fungal diagnosis and identification of oval microsporidial spores (1.5 μ × 2.5 μ) in fecal smears or body fluids using microscopy with the appropriate ultraviolet wavelength.^1,23,24 A modified trichrome staining method^g also has been developed for identifying organisms using bright field microscopy.^1,23–25 Unfortunately, these diagnostic services are not readily available from veterinary medical diagnostic laboratories on a fee-for-service basis, because the demand is low for microsporidial diagnostic testing of animal samples. Consequently, canine Encephalitozoon infections are typically diagnosed postmortem, as illustrated by the cases included in this report.

When evaluating histological tissues from necropsy or biopsy specimens, parasites do not have distinctive staining characteristics using H&E stain, and they are easily missed on routine microscopic analysis. Microsporidial spores are Gram positive, but the vegetative proliferative stages of the parasite are not, so special-tissue Gram stains are sometimes helpful in histopathological analysis of tissues.¹

Even if encephalitozoonosis cases are diagnosed antemortem, no treatment has been reported for use in dogs. The benzimidazole drug, albendazole, has been used successfully to treat Encephalitozoon infections in immunocompromised humans using a dosage of 400 mg q 12 hours for ≥3 weeks.^1,5 That drug is not approved for use in dogs, but it might be an extra-label treatment option under appropriate circumstances. Other compounds have shown some in vitro efficacy against the parasite, and a few of those compounds have been used experimentally in humans, but no experimental treatment regimens have been reported in dogs.¹

In humans, the primary route of transmission of microsporidial infections is through ingestion or inhalation of infective parasite spores.^5,6,22 While data are lacking in naturally infected canine cases, infections via ingestion of microsporidial spores and by transplacental transmission methods have been documented experimentally in dogs.^1,8,12,13 In this study and in the limited available literature, most infected animals were associated with breeding facilities containing numerous dogs.^8,13 These data suggest that horizontal transmission of the parasite could be facilitated among a large number of adult animals maintained in close contact within the kennel, particularly if sanitation was not optimum. The puppies in these cases also might have been exposed in utero to infection from their asymptomatic dams, since they developed clinical disease at early ages. Further epidemiological work using parasitological and serological testing would be helpful in understanding the dynamics of microsporidial infections in a dog population.

In this study, all of the cases of Encephalitozoon infection occurred in toy breeds of dogs. However, we do not believe that this trend suggests a breed-associated susceptibility to this parasite. Instead, the animals in these cases probably reflect the trend for large commercial breeding kennels to focus on toy breeds as the optimum market for the pet store trade.

All of the reported cases were identified in the state of Texas; however, we do not believe a particular geographical area is associated with this infection in dogs. These cases represent a compilation of necropsy and histopathological submissions to the state veterinary diagnostic laboratory, which results in the apparent geographic clustering. However, the various cases were submitted from locations as far as 500 miles apart over a period of 10 years.

Encephalitozoon cuniculi is a pathogen with zoonotic potential causing additional concern from a public health standpoint.^1,2,21 Human infections with this microsporidial species have been reported, particularly in immunocompromised patients.^4–7 Anecdotal evidence supports this concern. In a single human case, a 10-year-old girl was reported to seroconvert after close contact with a litter of puppies with encephalitozoonosis.⁸ Previous molecular characterization of rRNA genes suggested that isolates of dog and human origin were identical, while rabbit and rodent isolates of E. cuniculi showed subtle differences.³ Our molecular characterization of 13 canine isolates confirms and extends previous work by analyzing additional canine isolates and showing 100% homology with previously reported sequences.²¹

Additional studies are needed to document E. cuniculi infections as a cause of puppy neurological disease and canine renal disease. A better understanding of transmission patterns is needed in order to control this parasite in breeding kennels and multidog households. Clinical studies are needed to document parasite spore shedding in feces and/or urine of infected animals as a source of canine or human exposure. Further studies from a public health standpoint are also warranted to define the role of the dog as an inapparent carrier of this parasite.

Conclusion

The intracellular pathogen, E. cuniculi, was identified as a cause of fatal encephalitis in 18 puppies from 11 unrelated sources. The parasite was also identified as the cause of end-stage renal disease in a young dog. This infection is probably underdiagnosed because of a lack of familiarity with the parasite and a lack of readily available diagnostic testing methods. Asymptomatic infections probably occur in dogs, and further studies are needed to understand the epidemiology and transmission patterns of this parasite in dogs and to evaluate the zoonotic risk to humans.

Acknowledgments

We thank the following for their submission of case materials included in this report: Drs. Ann Buchanan and Steve Wilson, Glenwood Animal Hospital, Tyler, Texas; Dr. Richard Hoekstra, Animal House Veterinary Clinic, Dallas, Texas; Dr. Laura Lipsey, Booneville Animal Hospital, Bryan, Texas; and Alta Vista Animal Hospital, Keller, Texas. Thanks also to Kathleen Logan for her technical support and expertise in molecular analyses.