Am. J. Trop. Med. Hyg., 78(6), 2008, pp. 917-921
Copyright © 2008 by The American Society of Tropical Medicine and Hygiene
Cryptosporidium muris in a Texas Canine Population
Philip J. Lupo,
Rebecca C. Langer-Curry,
Mary Robinson,
Pablo C. Okhuysen, AND
Cynthia L. Chappell*
Center for Infectious Disease, The University of Texas School of Public Health, Houston, Texas; Center for Laboratory Animal Medicine and Care, and School of Medicine, The University of Texas Health Science Center at Houston, Houston, Texas
 |
ABSTRACT
|
Molecular technology has led to the discovery of previously unrecognized Cryptosporidium species in new hosts, such as C. canis in humans. The notion that dogs may transmit Cryptosporidium species to humans has significant public health implications, and additional studies are merited. The purpose of this study was to examine a group of kenneled dogs to determine the prevalence of Cryptosporidium species infection and to identify parasite species. Prevalence of active infection was 71%. Six positive samples were analyzed by polymerase chain reaction amplification of the 18S ribosomal RNA gene followed by restriction fragment length polymorphism analysis to identify the Cryptosporidium species. Restriction digest patterns identified C. muris as the infecting species in all six dogs; species identity was confirmed by genetic sequencing. To our knowledge, this is the first report of a naturally occurring C. muris infection in a canine host. The finding of C. muris in asymptomatic canines supports the notion of dogs as potential sources of human infection.
 |
INTRODUCTION
|
Cryptosporidium species are intracellular, apicomplexan parasites that infect a wide variety of mammals. In humans, Cryptosporidium is a causative agent of diarrhea and is now recognized in more than 150 different host species throughout the world.1 Cryptosporidium is ubiquitous in human populations, and infection commonly results from exposure to contaminated water or an infected person or animal. Cryptosporidium is difficult to control because the oocysts are stable in the environment, resistant to standard methods of disinfection, and their small size (< 5 µm) makes them difficult to efficiently remove from water by filtration.2
Genotyping of oocysts isolated from infected hosts has confirmed multiple Cryptosporidium species, distinguished distinct genotypes within the various species, and identified new species of Cryptosporidium. In fact, Cryptosporidium species not previously recognized as pathogenic in humans have been found in both immunocompetent and immunocompromised humans. Although C. parvum and C. hominis are responsible for most human infections, C. baileyi, C. canis, C. felis, C. meleagridis, C. muris, and Cryptosporidium pig and cervine genotypes have been detected in humans.3 Most Cryptosporidium species develop in enterocytes, generally causing a self-limited diarrhea in immunocompetent persons. In contrast, C. muris infects the stomach lining and can become chronic, but manifesting few, if any, symptoms.4,5
Stray and domesticated dogs in many countries throughout the world harbor Cryptosporidium,6 although in many studies the particular Cryptosporidium species was not identified. To date, C. parvum and C. canis are the Cryptosporidium species associated with canines.7 However, experimental challenge of dogs with C. muris resulted in a persistent but asymptomatic infection.4,5 Cryptosporidium infections in canines are typically associated with low oocyst excretion and few, if any, clinical signs.8
The possibility that dogs may transmit Cryptosporidium to humans has significant public health implications, and additional studies using sensitive detection methods, particularly in the United States, are merited. The purpose of this study was to examine a group of kenneled dogs to determine the prevalence of Cryptosporidium infection in this cohort and to identify parasite species.
Several studies have ascertained the prevalence of Cryptosporidium in dogs. These studies were conducted in various parts of the world, each with different canine populations (stray versus kenneled) and diagnostic tests (Table 1
). Prevalence estimates ranged from 0.0% to 7.4% (median = 1.0%) for less sensitive detection methods (carbol fuchsin stain, acid-fast stain, sugar flotation),9 and from 0.0% to 23% (median = 6.5%) with more sensitive methods (immunofluorescence assay [IFA], enzyme immunoassay [EIA], polymerase chain reaction [PCR]). Two canine studies conducted in the United States10,11 reported prevalences of 2.0% and 3.8%. Furthermore, a recent study has suggested the possible transmission of C. canis from a dog to children residing in the same household.12 A single seroprevalence study of 20 dogs indicated a high prevalence (80%) of antibodies to Cryptosporidium,13 but the small sample size made generalization to the canine population at large difficult.
 |
MATERIALS AND METHODS
|
Sample population.
The study population was a convenience sample of 70 kenneled dogs (1–3 years of age) from the University of Texas Health Science Center (UTHSC) Kennels located in Huntsville, Texas. Dogs included in the study population were housed in one of two wings of the UTHSC Kennels. The West Wing included adult dogs (n = 29) kenneled in the UTHSC facility from birth. The South Annex housed adult dogs (n = 41) that were retired from their duties as Texas Department of Criminal Justice (TDCJ) tracking dogs for prisons located throughout the state. These dogs typically remain in the kennel from two weeks to one year. All dogs were hounds, and the study population was equal in terms of sex (51% female and 49% male). Fifty-three percent of the dogs were unrelated, and among those that were, there were a total of 10 family groups.
The kennels are concrete and brick buildings surrounded by an open field. All dogs were housed in indoor/outdoor cages on concrete pads as either one to two dogs per cage and were cared for by residents of the Texas Department of Criminal Justice in Huntsville. Cages were adjacent to a corridor (i.e., run) used to exercise the dogs. The cages and runs were washed daily with water and cleaned weekly with bleach. The investigation was reviewed, approved, and conducted in accordance with the guidelines of the University of Texas Health Science Center at Houston Animal Welfare Committee.
Specimen collection, processing, and analysis.
Up to two stool samples were collected per cage in the morning prior to the daily cleaning. When two samples were present in cages with two dogs, the samples were collected and randomly labeled A or B. A total of 70 samples were collected from 56 cages. Each sample was collected in a sterile container and immediately transported to the laboratory. Approximately 10 g of fecal specimen was diluted in 10% neutral buffered formalin (1:4 [v/v]) and stored at 4°C. An additional vial of stool was frozen with no additives for further analysis.
Cryptosporidium antigen in fixed, diluted fecal samples was tested in duplicate with the ProSpecT Cryptosporidium Microplate Assay (Alexon-Trend, Ramsey, MN). The samples were analyzed according to the manufacturers protocol. Optical density at 450 nm was determined (TiterTek MultiScan MCC/340; Titertek, Huntsville, AL), and a mean was calculated for each sample. Negative and positive controls were provided by the manufacturer. Any value > 0.050 was defined per the manufacturer as positive for Cryptosporidium antigen. The assay has a reported sensitivity and specificity of 98% and 100%, respectively.14 It has been tested in animals other than humans, including dogs. Among dogs, specificity and sensitivity were 94% and 71%, respectively.15
An indirect IFA (Crypto-Glo, Waterborne, Inc., New Orleans, LA) was used on selected samples to confirm morphologic features of oocysts. All tests were done in accordance with the manufacturers directions.
PCR–restriction fragment length polymorphism (PCR-RFLP) analysis.
Six canine fecal samples that were positive for Cryptosporidium antigen were randomly selected for molecular analysis along with two C. muris negative controls. Frozen stool samples were thawed, and DNA was isolated using QIAamp DNA Stool Mini Kit (Qiagen Inc., Valencia, CA) according to manufacturers protocol. Isolated DNA was then subjected to 18S ribosomal RNA (rRNA) PCR-RFLP analysis as described.16 Briefly, DNA (10 µL) was amplified in a nested PCR, and products were incubated for one hour at 37°C with either SsP I or Vsp I (New England Biolabs, Beverly, MA). This technique has been used for species- and genotype-specific identification of Cryptosporidium.16 The length of the 18S rRNA gene is 1,733–1,750 basepairs, depending on the species and genotype tested. In the present study, the Cryptosporidium genus-specific primers used in the first reaction were 5'-TTCTAGAGCTAATACATGCG-3' (forward primer) and 5'-CCCTAATCCTTCGAAACAGGA-3' (reverse primer). Species-specific primers used in the second reaction were 5'-GGAAGGGTTGTATTTATTAGATA-AAG-3' (forward primer) and 5'-AAGGAGTAAGGAAC-AACCTCCA-3' (reverse primer). Digested products were resolved by electrophoresis on a 1.5% agarose gel and visualized by staining with ethidium bromide. These products were compared with previously published PCR patterns generated under the same conditions.17 The sample with the most intense RFLP band was subjected to sequencing, and the PCR product was analyzed on an ABI 3100 capillary automated DNA sequencer (Applied Biosystems, Foster City, CA). Sequencing was in one direction and conducted using genomic DNA. The resulting nucleotide sequence was subjected to homology BLAST search against GenBank and aligned with various Cryptosporidium nucleotide sequences using the ClustalX program (ftp://ftp-igbmc.u-stasbg.fr/pub/ClustalX/). A neighbor-joining phylogenetic tree was generated by molecular evolutionary genetic analysis, version 2.118 on the basis of evolutionary distances calculated with the Kimura two-parameter model.
 |
RESULTS
|
Prevalence of Cryptosporidium infections.
Stool samples (n = 70) from 56 cages yielded 40 cages (71%) positive for Cryptosporidium antigen by EIA. Negative EIA values (i.e., those < 0.050) ranged from 0.026 to 0.047. Fifty-one samples were EIA positive with values ranging from 0.054 to 2.170. Nine (17.6%) of the positive samples had absorbance values between 0.050 and 0.075, i.e., near the cutoff point. There were no discrepant results (positive versus negative) between duplicate measurements. The coefficient of variation for the duplicate measurements was 6.6%, which represented minimal within-subject variation. Analysis of the 70 individual samples from these cages yielded 49 positive for Cryptosporidium antigen (prevalence = 70%). All fecal samples were formed, and no gastrointestinal signs were noted in the dogs at the time of collection. The lack of clinical signs is consistent with previous studies assessing Cryptosporidium infection in dogs. Puppies bred in the facility were not assessed for infection but were also asymptomatic.
Stools from six infected dogs randomly selected for genotyping were also tested by IFA (Crypto-Glo, Waterborne, Inc.) to examine the size and shape of the oocysts. Labeled oocysts were morphologically consistent with those of C. muris.
One wing of the kennel housed dogs that were retired from duties as tracking dogs for prisons located throughout the state. Given their assignment as tracking dogs, these animals were exposed to numerous sylvatic environments. Another wing housed breeding dogs. Some dogs were transferred between wings. Although the dogs did not have daily contact with one another (except for cage mates), they used some of the same run yards and had contact during breeding. Cryptosporidium prevalence estimates were similar for both wings.
Genotyping Cryptosporidium species.
Stools from six infected dogs were randomly selected for PCR-RFLP analysis along with two C. muris-negative controls. The PCR products indicated that all samples contained the genus Cryptosporidium (Figure 1
). To determine the Cryptosporidium species, amplicons were digested (Figure 1
) and compared with previously published banding patterns unique to Cryptosporidium spp.17 The RFLP analysis using two restriction endonucleases (Vsp I and Ssp I) each yielded a pattern indicative of Cryptosporidium muris (Figure 2
). Furthermore, the PCR product from canine SA39 was sequenced (GenBank accession no. EU549784), and a BLAST search showed that the 18S rRNA PCR fragment revealed greater than 99% identity with a known C. muris DNA fragment from the same locus. Sequencing was conducted using genomic DNA yielding a 707-basepair amplicon. The sequence from canine SA39 was compared with a known C. muris DNA fragment (C. muris mouse genotype gi3873253). Identity of the infecting Cryptosporidium species was further confirmed by neighbor-joining phylogenetic analysis, which indicated that the C. muris dog isolate clustered with previously reported C. muris isolates. A phylogenetic tree (Figure 3
) was constructed from the canine SA39 sequence and 10 Cryptosporidium species/isolates included in the analysis: C. muris 1 (gi3873252), C. muris 2 (gi3873253), C. muris 3 (gi22128355), C. muris 4 (gi49359175), C. muris 5 (gi6997), C. parvum (gi7001), C. serpentis (gi5870846), C. meleagridis (gi5714421), C. felis (gi5714422), and C. canis (gi5714423).

View larger version (80K):
[in this window]
[in a new window]
|
FIGURE 1. Polymerase chain reaction analysis for Cryptosporidium species in stool samples. Six canine samples and two controls were amplified with primers specific for the 18S ribosomal RNA gene. Lane 1, molecular mass markers; lane 2, C. hominis; lane 3, C. parvum; lane 4, dog SA38; lane 5, dog SA39; lane 6, dog SA40A; lane 7, dog SA40B; lane 8, dog W15; lane 9, dog W16.
|
|

View larger version (43K):
[in this window]
[in a new window]
|
FIGURE 2. Polymerase chain reaction (PCR)–restriction fragment length polymorphism analysis for Cryptosporidium species in stool samples. Six canine samples and two controls were amplified with primers specific for the 18S ribosomal RNA gene. The PCR products were then digested with either Vsp I (lanes 2a–9a) or Ssp I (lanes 2b–9b). Lane 1, molecular mass markers; lane 2, C. hominis; lane 3, C. parvum; lane 4, dog SA38; lane 5, dog SA39; lane 6, dog SA40A; lane 7, dog SA40B; lane 8, dog W15; lane 9, dog W16.
|
|

View larger version (9K):
[in this window]
[in a new window]
|
FIGURE 3. Comaprison of Cryptosporidium muris fragment isolated from canine SA39 (GenBank accession no. EU549784) with other published Cryptosporidium species sequences by neighbor-joining analysis of small subunit ribosomal RNA sequences. The Kimura18 two-parameter model was used in distance calculation, and the numbers on branches indicate percent bootstrap values. Sequences were obtained from GenBank: C. muris 1 (gi3873252), C. muris 2 (gi3873253), C. muris 3 (gi22128355), C. muris 4 (gi49359175), C. muris 5 (gi6997), C. parvum (gi7001), C. serpentis (gi5870846), C. meleagridis (gi5714421), C. felis (gi5714422), and C. canis (gi5714423).
|
|
 |
DISCUSSION
|
To our knowledge, this is the first report of a natural infection of C. muris in canines. The large percentage of infected dogs in this kennel setting suggested a common source of exposure or transmission by fecal cross-contamination of the environment. Because the facility has no evidence rodent infestation and dog food is kept in sealed plastic containers, direct transmission between cage mates as well as fecal contamination of the exercise area are the most likely sources of spread within the colony. Although there was no visible evidence of rodent infection, the cages were open, and accessible for rodent contamination. One may also speculate that tracking dogs were exposed to rodent droppings during their training or tracking activities, which often included access to wooded areas. Rodents have previously drawn attention as potential sources of Cryptosporidium for other host species.19 Another previous study20 described a C. muris outbreak in captive marsupials. In that outbreak, mouse feces in the captive breeding facility contained C. muris, thus leading the investigators to conclude that the marsupials acquired the infection through contaminated food or water.
The 71% prevalence of Cryptosporidium in kenneled dogs was determined with a commercially available EIA that detects Cryptosporidium antigens in fecal samples. Positive results in canines with confirmed C. muris infections suggested significant cross-reactivity between the two Cryptosporidium species and the usefulness of this assay in detecting C. muris infections. Other studies conducted in our laboratory showed that the EIA yielded positive results when purified, intact C. muris or C. parvum oocysts were used in the test (unpublished data). Furthermore, in a preliminary experiment, titration of C. muris oocysts yielded positive results for wells containing 1.6 x 104 or more oocysts, the same level of detection as C. parvum.
In the RFLP analysis, controls (C. hominis and C. parvum) displayed unique banding patterns compared with those from the canines. Stool samples (n = 6) that yielded the highest EIA absorbance values were selected for genetic testing. The PCR results from these dogs were the same and all were consistent with the previously published C. muris pattern.17 Thus, testing of samples from additional dogs was deemed unnecessary. Sequencing of the 18S rRNA PCR fragment from canine SA39 showed > 99% identity with a known C. muris DNA fragment. Figure 3
also supports the species identity of the canine parasite. Furthermore, the isolate from canine SA39 groups with other C. muris isolates, indicating infection with C. muris. The RFLP patterns and sequencing indicate C. muris infection in the six canines, suggesting that this species was circulating in the dog population and responsible for positive EIA results.
Cryptosporidiosis in dogs seems to be a particularly mild infection, regardless of the Cryptosporidium species. Because C. muris infection in dogs appears to be asymptomatic, it would likely have gone unrecognized in the absence of this study. Interestingly, C. muris infection often leads to chronic infection with the parasite. Thus, it is likely that dogs have a persistent infection or may be re-infected from repeated exposures.
The risk of infection among the animal handlers at the TDCJ facility is presently unknown. Further studies will be required to evaluate infections in the exposed workers and, if found, the parasites potential for transmission within the larger prison population. Experimental studies of C. parvum or C. hominis in humans has shown a low infectious dose with some isolates. If the same holds true for C. muris, it is possible that even a low level of infection in dogs may be a source of human exposure.
The zoonotic potential of Cryptosporidium and the possibility of human infection by companion animals have important implications, particularly in at-risk populations, such as those who are immunocompromised. This study should serve to stimulate future research into the Cryptosporidium species that are potentially pathogenic for humans, as well as the sources and risk factors important in transmission of cryptosporidiosis.
Received October 12, 2007.
Accepted for publication March 11, 2008.
Financial support: This study was supported in part by the Environmental Protection Agency STAR grant R-82918001.
* Address correspondence to Cynthia L. Chappell, University of Texas School of Public Health, 1200 Herman Pressler Street, Suite 118A Houston, TX 77030. E-mail: cynthia.l.chappell{at}uth.tmc.edu 
Authors addresses: Philip J. Lupo, University of Texas School of Public Health, 1200 Herman Pressler Street, Room E645, Houston, TX 77030, Tel: 713-500-9221, Fax: 713-500-9020, E-mail: Philip.J.Lupo{at}uth.tmc.edu. Rebecca Langer-Curry, Office of Environmental Safety, Baylor College of Medicine, 2450 Holcombe Boulevard, Room OW200B, BCM175, Houston, TX 77021, E-mail: langercu{at}bcm.tmc.edu. Mary Robinson, Center for Laboratory Animal Medicine and Care, University of Texas Health Science Center at Houston, IMM, Room 241, 1825 Pressler Street, Houston, TX 77030, E-mail: Mary.A.Robinson{at}uth.tmc.edu. Pablo C. Okhuysen, Division of Infectious Diseases, University of Texas Medical School, 6431 Fannin, Room 2.112, Houston, TX 77030, E-mail: Pablo.C.Okhuysen{at}uth.tmc.edu. Cynthia L. Chappell, University of Texas School of Public Health, 1200 Herman Pressler Street, Suite 118A Houston, TX 77030, Tel: 713-500-9026, Fax: 713-500-9020, E-mail: Cynthia.L.Chappell{at}uth.tmc.edu.
 |
REFERENCES
|
- Fayer R, 2004. Cryptosporidium: a water-borne zoonotic parasite. Vet Parasitol 126: 37–56.[Web of Science][Medline]
- Betancourt WQ, Rose JB, 2004. Drinking water treatment processes for removal of Cryptosporidium and Giardia. Vet Parasitol 126: 219–234.[Web of Science][Medline]
- Caccio SM, 2005. Molecular epidemiology of human cryptosporidiosis. Parassitologia 47: 185–192.[Medline]
- Iseki M, Maekawa T, Moriya K, Uni S, Takada S, 1989. Infectivity of Cryptosporidium muris (strain RN 66) in various laboratory animals. Parasitol Res 75: 218–222.[Web of Science][Medline]
- Aydin Y, Ozkul IA, 1996. Infectivity of Cryptosporidium muris directly isolated from the murine stomach for various laboratory animals. Vet Parasitol 66: 257–262.[Web of Science][Medline]
- Xiao L, Ryan UM, 2004. Cryptosporidiosis: an update in molecular epidemiology. Curr Opin Infect Dis 17: 483–490.[Web of Science][Medline]
- Miller DL, Liggett A, Radi ZA, Branch LO, 2003. Gastrointestinal cryptosporidiosis in a puppy. Vet Parasitol 115: 199–204.[Web of Science][Medline]
- Abe N, Sawano Y, Yamada K, Kimata I, Iseki M, 2002. Cryptosporidium infection in dogs in Osaka, Japan. Vet Parasitol 108: 185–193.[Web of Science][Medline]
- Irwin PJ, 2002. Companion animal parasitology: a clinical perspective. Int J Parasitol 32: 581–593.[Web of Science][Medline]
- el-Ahraf A, Tacal JV Jr, Sobih M, Amin M, Lawrence W, Wilcke BW, 1991. Prevalence of cryptosporidiosis in dogs and human beings in San Bernardino County, California. J Am Vet Med Assoc 198: 631–634.[Web of Science][Medline]
- Hackett T, Lappin MR, 2003. Prevalence of enteric pathogens in dogs of north-central Colorado. J Am Anim Hosp Assoc 39: 52–56.[Abstract/Free Full Text]
- Xiao L, Cama VA, Cabrera L, Ortega Y, Pearson J, Gilman RH, 2007. Possible transmission of Cryptosporidium canis between children and a dog in a household. J Clin Microbiol 45: 2014–2016.[Abstract/Free Full Text]
- Tzipori S, Campbell I, 1981. Prevalence of Cryptosporidium antibodies in 10 animal species. J Clin Microbiol 14: 455–456.[Abstract/Free Full Text]
- Garcia LS, Shimizu RY, 1997. Evaluation of nine immunoassay kits (enzyme immunoassay and direct fluorescence) for detection of Giardia lamblia and Cryptosporidium parvum in human fecal specimens. J Clin Microbiol 35: 1526–1529.[Abstract]
- Rimhanen-Finne R, Enemark HL, Kolehmainen J, Toropainen P, Hanninen ML, 2007. Evaluation of immunofluorescence microscopy and enzyme-linked immunosorbent assay in detection of Cryptosporidium and Giardia infections in asymptomatic dogs. Vet Parasitol 145: 345–348.[Web of Science][Medline]
- Xiao L, Morgan UM, Limor J, Escalante A, Arrowood M, Shulaw W, Thompson RC, Fayer R, Lal AA, 1999. Genetic diversity within Cryptosporidium parvum and related Cryptosporidium species. Appl Environ Microbiol 65: 3386–3391.[Abstract/Free Full Text]
- Xiao L, Escalante L, Yang C, Sulaiman I, Escalante AA, Montali RJ, Fayer R, Lal AA, 1999. Phylogenetic analysis of Cryptosporidium parasites based on the small-subunit rRNA gene locus. Appl Environ Microbiol 65: 1578–1583.[Abstract/Free Full Text]
- Kumar S, Tamura K, Jakobsen IB, Nei M, 2001. MEGA2: molecular evolutionary genetics analysis software. Bioinformatics 17: 1244–1245.[Abstract/Free Full Text]
- Fayer R, Morgan U, Upton SJ, 2000. Epidemiology of Cryptosporidium: transmission, detection and identification. Int J Parasitol 30: 1305–1322.[Web of Science][Medline]
- Warren KS, Swan RA, Morgan-Ryan UM, Friend JA, Elliot A, 2003. Cryptosporidium muris infection in bilbies (Macrotis lagotis). Aust Vet J 81: 739–741.[Web of Science][Medline]
- Simpson JW, Burnie AG, Miles RS, Scott JL, Lindsay DI, 1988. Prevalence of Giardia and Cryptosporidium infection in dogs in Edinburgh. Vet Rec 123: 445.[Web of Science][Medline]
- Grimason AM, Smith HV, Parker JF, Jackson MH, Smith PG, Girdwood RW, 1993. Occurrence of Giardia sp. cysts and Cryptosporidium sp. oocysts in faeces from public parks in the west of Scotland. Epidemiol Infect 110: 641–645.[Medline]
- Causape AC, Quilez J, Sanchez-Acedo C, del Cacho E, 1996. Prevalence of intestinal parasites, including Cryptosporidium parvum, in dogs in Zaragoza city, Spain. Vet Parasitol 67: 161–167.[Web of Science][Medline]
- Kim JT, Wee SH, Lee CG, 1998. Detection of Cryptosporidium oocysts in canine fecal samples by immunofluorescence assay. Korean J Parasitol 36: 147–149.[Medline]
- Bugg RJ, Robertson ID, Elliot AD, Thompson RC, 1999. Gastrointestinal parasites of urban dogs in Perth, Western Australia. Vet J 157: 295–301.[Web of Science][Medline]
- Cirak VY, Bauer C, 2004. Comparison of conventional coproscopical methods and commercial coproantigen ELISA kits for the detection of Giardia and Cryptosporidium infections in dogs and cats. Berl Munch Tierarztl Wochenschr 117: 410–413.[Web of Science][Medline]
- Nakai Y, Hikosaka K, Sato M, Sasaki T, Kaneta Y, Okazaki N, 2004. Detection of Cryptosporidium muris type oocysts from beef cattle in a farm and from domestic and wild animals in and around the farm. J Vet Med Sci 66: 983–984.[Web of Science][Medline]
- Huber F, Bomfim TC, Gomes RS, 2005. Comparison between natural infection by Cryptosporidium sp., Giardia sp. in dogs in two living situations in the West Zone of the municipality of Rio de Janeiro. Vet Parasitol 130: 69–72.[Web of Science][Medline]
- Giangaspero A, Iorio R, Paoletti B, Traversa D, Capelli G, 2006. Molecular evidence for Cryptosporidium infection in dogs in central Italy. Parasitol Res 99: 297–299.[Web of Science][Medline]