Infections in mice with tachyzoites and bradyzoites of Neospora caninum (Protozoa: Apicomplexa)

Tachyzoites of 2 isolates of Neospora caninum (NC-1 and NC-2) were inoculated subcutaneously (s.c.), intraperitoneally (i.p.), or orally into mice to compare the effects of route of inoculation on pathogenicity. Mice developed more severe disease, and disease occurred sooner when inoculated with the NC-1 isolate compared to the NC-2 isolate. Deaths occurred earlier in mice inoculated i.p. with either isolate. Mice inoculated orally or s.c. with tachyzoites responded similarly to infection. Tissue cysts of the NC-2 isolate produced infections in mice following oral or s.c. inoculation. Lesions seen in mice inoculated with tachyzoites or bradyzoites were primarily acute pneumonia, myositis, encephalitis, ganglioradiculoneuritis, and pancreatitis. In vitro studies demonstrated that tachyzoites of both isolates were killed by incubation in pepsin-HCl solution but not 1% trypsin solution. Bradyzoites of the NC-2 isolate were able to withstand treatment with pepsin-HCl solution.     

Biologic, morphologic, and molecular characterisation of Neospora caninum isolates from littermate dogs

Isolation and biologic and molecular attributes of Neospora caninum from three littermate dogs are described. Tissue cysts were confined to the brain and striated muscles. N. caninum was isolated (isolates NC-6, NC-7, and NC-8) in rodents and cell culture that had been inoculated with brain tissue from the dogs. Schizont-like stages reactive with N. caninum antibodies were seen in cell cultures seeded with bradyzoites released from Percoll-isolated N. caninum tissue cysts from the brain of one dog. Tissue cysts were infective orally to mice and gerbils, but not to cats and dogs. The isolates were also identified as N. caninum by PCR and sequence analysis.     

Characterization of temperature-sensitive strains of Neospora caninum in mice

Temperature-sensitive (ts) strains of the Neospora caninum tachyzoites were selected by chemical mutagenesis and selection for growth at 32 C. Three ts strains and the parental, N. caninum wild-type strain, NC-1, were examined in the present study for their ability to cause disease in inbred BALB/c mice, outbred ICR mice, and chemically immunosuppressed ICR mice. In BALB/c mice, all 3 strains failed to induce clinical disease, whereas infection with the NC-1 strain caused central nervous system disease and death in some mice. No disease was observed in ICR mice inoculated with the 3 ts strains or the NC-1 strain. All immunosuppressed ICR mice inoculated with the NC-1 strain died, whereas no immunosuppressed mice inoculated with the NCts-4 strain and only 1 of 5 mice inoculated with the NCts-8 and NCts-12 strains died. The NCts-4 and NCts-12 strains reverted to a wild-type phenotype when grown at 37 C. Vaccination of BALB/c mice with live, but not frozen NCts-8 strain tachyzoites induced significant (P < 0.05) protection following NC-1 strain challenge.     

First isolation of Neospora caninum from the feces of a naturally infected dog.

Neospora caninum is a major cause of abortion in cattle worldwide. Cattle become infected with N. caninum by ingesting oocysts from the environment or transplacentally from dam to fetus. Experimentally, dogs can act as definitive hosts, but dogs excrete few oocysts after ingesting tissue cysts. A natural definitive host was unknown until now. In the present study, N. caninum was isolated from the feces of a dog. Gerbils (Meriones unguiculatus) fed feces from the dog developed antibodies to N. caninum in the Neospora caninum agglutination test, and tissue cysts were found in their brains. Neospora caninum was isolated in cell culture and in gamma-interferon gene knockout mice inoculated with brain homogenates of infected gerbils. The DNA obtained from fecal oocysts of the dog, from the brains of gerbils fed dog feces, and from organisms isolated in cell cultures inoculated with gerbil brains was confirmed as N. caninum. The identification of N. caninum oocyst by bioassay and polymerase chain reaction demonstrates that the dog is a natural definitive host for N. caninum.     

Vertical transmission of Neospora caninum in BALB/c mice determined by polymerase chain reaction detection.

Vertical transmission of Neospora caninum was evaluated in BALB/c mice using an N. caninum-specific polymerase chain reaction (PCR) assay as a means of detecting parasite transmission to offspring. BALB/c mice were infected with the NC-1 isolate of N. caninum during pregnancy (days 8-15 gestation). Transmission of parasite, detected by PCR, was determined in 2- to 23-day-old offspring. When dams were infected on days 13-15 of gestation, transfer of parasites was detected in only a proportion of the litter. Infection between days 8 and 12 of gestation resulted in a high frequency of parasite transmission; every offspring from all litters was infected. The tissue locations of parasites in pups of different ages were determined. In young pups (2- to 4-days-old), the predominant sites of infection were the lungs and the brain. In older pups (7- and 23-days-old) the predominant site of infection was the brain. This study shows that PCR may be useful for evaluation of candidate vaccines against horizontal N. caninum infection, vertical transmission, or both.     

First isolation of Neospora caninum from an aborted bovine fetus in Spain

Neospora caninum was isolated from the brain of a 6-mo-old aborted bovine fetus from Galicia, Spain. The fetal brain homogenate was inoculated intraperitoneally into cortisonized mice. The peritoneal exudate from the infected mice, along with mouse sarcoma cells (Tg180), was inoculated into a second group of mice, and parasites were harvested from the peritoneal exudate. The parasites were adapted to in vitro growth in Vero monolayers. The tachyzoites from the peritoneal exudate reacted positively with anti-N. caninum antibodies and not with anti-Toxoplasma gondii antibodies on indirect fluorescent antibody test. The tachyzoites were lethal to interferon gamma gene knock out (KO) mice and could be identified immunohistochemically in the tissues. The identity of the parasite was also confirmed by polymerase chain reaction amplification of N. caninum-specific fragments. The sequences of the amplified gene 5 fragments (GenBank AY494944) were found to be identical to that of an Austrian isolate of N. caninum but not to that of NC-1. This is the first isolation of viable N. caninum from Spain.     

Neospora caninum (Protozoa: apicomplexa) infections in mice

Groups of mice were given 0 mg, 4 mg, or 2 mg of methylprednisolone acetate (MPA) 7 days prior to, the day of, and 7 days after subcutaneous inoculation with 0 or 2 x 10(5) tachyzoites of Neospora caninum. Clinical signs of disease were seen only in mice given both MPA and N. caninum tachyzoites. Mice given 4 mg MPA and N. caninum tachyzoites developed severe disseminated neosporosis and most died or were killed when comatose 11-13 days postinoculation (PI). Acute pneumonia, polymyositis, encephalitis, hepatitis, and pancreatitis were the main lesions in these mice. Mice given 2 mg MPA and N. caninum developed mild pneumonia and many mice began showing neurological signs 14 days PI. Neurological signs consisted mainly of pronounced head-tilting and associated impairment of movement. Grossly visible 1-2-mm single or multiple, white areas of discoloration were seen in the brains of many of these mice. Encephalitis, ganglioradiculoneuritis, pneumonia, and polymyositis were the main changes seen in these mice. Tissue cysts of N. caninum were only seen in mice given 2 mg MPA and were first seen 21 days PI. Tissue cysts were 16-34 by 13-29 microns and had a 1.5-3.0-microns-thick cyst wall. Tissue cysts were seen only in the brain. Mice given 4 mg MPA and tachyzoites and host cells that had been frozen for 1 wk did not develop clinical signs of infection, indicating that freezing kills tachyzoites and that viruses or other agents were not involved in the genesis of disease seen in mice given MPA and viable tachyzoites.(ABSTRACT TRUNCATED AT 250 WORDS)     

Growth of and competition between Neospora caninum and Toxoplasma gondii in vitro

Competitive interactions between Neospora caninum and Toxoplasma gondii were studied because both species appear to have identical ecological niches in vitro. Tachyzoites of N. caninum (NC-1 isolate) and T. gondii (RH isolate) were compared in three in vitro studies: (1) rate of penetration of host cells; (2) generation time; and (3) competition between the two species when grown together in the same flask and allowed to compete for space. When tachyzoites of the two species were inoculated onto human foreskin fibroblasts, 3.24-times more N. caninum tachyzoites penetrated cells by 1 h p.i. At 3 h p.i., there were 2.87-times more N. caninum intracellular tachyzoites than T. gondii tachyzoites. The generation times for N. caninum (NC-1 isolate) and T. gondii (RH isolate) were approximately 14-15 h and 8-10 h, respectively. Before exponential growth occurred, both species displayed a lag period, which was 10-12 h for N. caninum and 8-10 h for T. gondii. To observe competition, equal numbers of tachyzoites of each species were mixed and inoculated into flasks of host cells, and the monolayers were allowed to proceed to >90% lysis before the next transfer. Competition was analysed for 31 days by labelling samples of each flask with a species-specific monoclonal antibody and determining the ratio of each species. In all trials, T. gondii outcompeted N. caninum. By 4 days p.i., 70% of the tachyzoites were T. gondii; this percentage increased to 97% by 23 days p.i. When the starting inoculum contained 75% N. caninum and 25% T. gondii tachyzoites, T. gondii was still competitively superior. When infected monolayers that were labelled with T. gondii-specific antibodies were examined, it was noted that both species can occupy and undergo endodyogeny in the same host simultaneously.     

Immunohistochemical and ultrastructural evidence for Neospora caninum tissue cysts in skeletal muscles of naturally infected dogs and cattle.

Protozoan stages were detected in the skeletal muscles of four dogs suffering from neosporosis and two neonatal calves with confirmed Neospora caninum- infection which could be immunohistochemically labelled by an antiserum against the bradyzoite-specific antigen BAG-5. In one calf, a tissue cyst was labelled by an antiserum against the N. caninum isolate NC-1. Ultrastructurally, a 0.3-1 microm-thick cyst wall surrounded the labelled parasites. The cysts were located within myofibres and contained varying numbers of bradyzoites each measuring 5.2(+/-0.6) x 1.6(+/-0.3) microm. The encysted stages showed typical ultrastructural features of N. caninum bradyzoites with subterminal nuclei, electron-dense rhoptries, and micronemes that were orientated perpendicular to the zoite pellicle. Immunohistochemistry and serology did not reveal any evidence for co-infection with Toxoplasma gondii. The detection of tissue cysts in skeletal muscle of N. caninum-infected intermediate hosts is of major epidemiological importance for the understanding of how definitive or intermediate carnivorous hosts become infected with N. caninum.     

Naturally-occurring Neospora caninum infection in an adult sheep and her twin fetuses

Neospora caninum tissue cysts were found in the brains of surgically delivered twin fetuses at 119 days of gestation. In the brains of both fetuses, there was an inflammatory reaction involving perivascular cuffings of mononuclear cells, glial nodules. The dam of these fetuses died because of metritis. Histopathological examination of the ewe revealed N. caninum tissue cysts and focal gliosis with mononuclear cell cuffings. A N. caninum-specific DNA fragment was detected in a brain homogenate of the ewe by the polymerase chain reaction method. This is the first report of N. caninum infection in twin ovine fetuses and in an adult sheep.     

Characterization of the Oregon isolate of Neospora hughesi from a horse

Neospora hughesi was isolated in cell cultures inoculated with homogenate of spinal cord from a horse in Oregon. Tachyzoites of this Oregon isolate of N. hughesi were maintained continuously by cell culture passage and tachyzoites were infective to immunosuppressed mice. Gamma interferon gene knockout (KO) mice injected with tachyzoites developed fatal myocarditis and numerous tachyzoites were seen in lesions. Gerbils (Meriones unguiculatus) inoculated with tachyzoites developed antibodies (> or = 1:500) as indicated by the Neospora caninum agglutination test but did not develop clinical signs, and Neospora organisms were not demonstrable in their tissues. Tissue cysts were not found in gerbils, nude mice, KO mice, immunosuppressed outbred Swiss Webster mice, or BALB/c mice injected with the Oregon isolate of N. hughesi. Ultrastructurally, tachyzoites of the Oregon isolate from the myocardium of infected KO mice and from cell culture were similar to N. caninum tachyzoites. Western blot analysis using NcSAG1 and NcSRS2 polyclonal and monoclonal antibodies and characterization of the internal transcribed spacer 1 sequences from the equine isolates and different isolates of N. caninum from dogs and cattle indicated that the Oregon isolate of N. hughesi is distinct from N. caninum isolates from cattle and dogs.     

Immune factors influencing the course of infection with Neospora caninum in the murine host

This paper investigates the role of specific immune factors on the course of infection in genetic knockout (gko) mice infected with 3 different strains of Neospora caninum. Survival time and parasite persistence were examined in interferon-gamma (IFN-gamma), tumor necrosis factor receptor-2 (TNFR2), interleukin 10 (IL-10), beta 2 microglobulin (beta2M), and inducible nitric oxide synthase (iNOS2) gko or wild-type (wt) mice following infection with either pathogenic (NC-1 or NC-2) or attenuated (NCts-8) N. caninum strains. Infection with NC-1 was 100% lethal in IFN-gamma gko mice, as evidenced by mean survival times of 10-13 days, depending on the challenge dose used. TNFR2 and beta2M gko mice infected with NC-1 or NC-2 strain demonstrated partial susceptibility to disease, as evidenced by histopathology and survival curves. TNFR2 or beta2M gko mice were not susceptible to infection with NCts-8, on the basis of absence of pathology and lack of mortality. Lack of mortality and minimal histopathology scores demonstrated that NC-1, NC-2, and NCts-8 infections were avirulent in IL-10 and iNOS2 gko mice. Adoptive transfer of immune cells from NCts-8 vaccinated normal syngeneic mice into IFN-gamma gko mice significantly (P < 0.05) prolonged mean survival times at all 3 challenge doses of NC-1 but failed to protect against mortality. Interestingly, there was a notable change in the tissue tropism of tachyzoites from the lung and brain in immunocompetent wt, TNFR2 gko, IL-10 gko, beta2M gko, and iNOS2 gko mice to the liver and spleen in IFN-gamma gko mice when challenged with N. caninum. On the basis of these results in gko mice, IFN-gamma is a critical cytokine in the host response against acute neosporosis.     

Comparative ultrastructure of tachyzoites, bradyzoites, and tissue cysts of Neospora caninum and Toxoplasma gondii

The ultrastructure of tachyzoites, bradyzoites and tissue cysts of the NC-1, NC-5 and NC-Liverpool strains of Neospora caninum are reviewed and compared with those of the VEG and ME-49 strains of Toxoplasma gondii. While each stage of N. caninum and T. gondii shared many ultrastructural characteristics, each parasite stage also had certain features or organelles that could be used to distinguish the two parasites. Some of the most prominent ultrastructural differences occurred in the number, appearance and location of rhoptries, looped-back rhoptries, micronemes, dense granules, small dense granules and micropores. The tissue cysts of both parasites were also basically similar, being surrounded by a cyst wall and not compartmentalised by septa. The cyst wall of N. caninum was irregular and substantially thicker, 0.5-4 microm, than those of T. gondii which were smooth and 0.5 microm thick.     

Vaccination with gamma-irradiated Neospora caninum tachyzoites protects mice against acute challenge with N. caninum

Neospora caninum, an apicomplexan parasite, is a leading cause of bovine abortions worldwide. The efficacy of gamma-irradiated N. caninum strain NC-1 tachyzoites as a vaccine for neosporosis was assessed in C57BL6 mice. A dose of 528 Gy of gamma irradiation was sufficient to arrest replication but not host cell penetration by tachyzoites. Female C57BL6 mice were vaccinated with two intraperitoneal inoculations of 1 x 10(6) irradiated tachyzoites at 4-wk intervals. When stimulated with N. caninum tachyzoite lysates, splenocytes of vaccinated mice, cultured 5 and 10 wk after vaccination, secreted significant (P<0.05) levels of interferon gamma, interleukin (IL)-10, and small amounts of IL-4. Antibody isotype-specific ELISA of sera from vaccinated mice exhibited both IgG1 and IgG2a isotypes of antibodies. Vaccinated mice were challenged intraperitoneally with 2 x 10(7)N. caninum tachyzoites. All vaccinated mice remained healthy and showed no obvious signs of neosporosis up to the 25th day post-challenge when the study was terminated. All unvaccinated control mice died within 1 wk of infection. Gamma-irradiated N. caninum tachyzoites can serve as an effective, attenuated vaccine for N. caninum.     

Experimental infection of sheep with Neospora caninum oocysts

The purpose of the present study was to investigate the potential of Neospora caninum oocysts to infect sheep and determine whether N. caninum DNA could be detected by polymerase chain reaction (PCR) assay in blood and brain of sheep after oocyst inoculation. Six ewes were inoculated per os with 10(4) N. caninum oocysts, whereas 2 ewes served as uninoculated controls. All sheep were bled weekly for 7 wk after inoculation. Blood was analyzed for the presence of N. caninum DNA by 2 different PCR assays, as well as for the presence of antibodies to recombinant and native N. caninum antigens. Neospora caninum DNA was detected in 2 sheep as early as 7 days after oocyst inoculation (DAOI). All 6 sheep were PCR positive by 32 days and remained positive until the end of the study at 49 DAOI. Aside from 1 ewe, all sheep inoculated with N. caninum oocysts contained detectable N. caninum DNA in the brain tissue collected at 49 DAOI. Unlike with PCR, no lesion or parasite was detected by immunohistochemistry. Antibodies were detected by enzyme-linked immunosorbent assay, Neospora agglutination test, or immunoblotting to either native or recombinant N. caninum antigens in sheep inoculated with oocysts.     

Toxoplasma gondii: blood and tissue kinetics during acute and chronic infections in mice

Acute lethal infections were obtained in mice by intraperitoneal (IP) injection of 10(2) or 10(4) tachyzoites of the virulent RH and C56 strains. Chronic infections were obtained by IP injection or peroral (PO) gavage of 20 cysts of the avirulent C strain. Mice were sacrificed at varying intervals after infection and parasite burdens were quantitated in blood, brain, and lungs using a tissue culture method. Acutely infected mice died within 6 to 10 days postinfection as a function of the strain and inoculum size. With either strain, tachyzoites were first detected in lungs on either Days 2 or 4 postinfection, according to the inoculum size, then in brain and blood at Days 4 or 6; parasitic loads remained constantly at a higher level in lungs than in brain until the date of death. Bradyzoites could only be detected in lungs, from Days 4 or 6 until death. In chronic infections, similar results were obtained for IP and PO infected mice. Both tachyzoites and bradyzoites were first detected in lungs and brain from Day 7 after infection; tachyzoites remained at a higher level in lungs than in brain until Day 10, then subsequently decreased in lungs. At Day 50, tachyzoites were not detectable in lungs, whereas bradyzoites remained at a constant level; in brain, both parasitic stages were detectable at a similar level throughout the follow-up period. These results indicate that infection with a virulent Toxoplasma strain is characterized by an early involvement of lungs, with pneumonia as the principal cause of death.(ABSTRACT TRUNCATED AT 250 WORDS)     

Intrauterine Neospora caninum inoculation of heifers and cows using contaminated semen with different numbers of tachyzoites

OBJECTIVE: To investigate the potential of different Neospora caninum tachyzoite doses to infect heifers (experiment 1) and cows (experiment 2) when administered in utero by artificial insemination via contaminated semen. METHODS: In experiment 1, five groups of 5, 7, 8, 9, and 5 cyclic heifers were hormonally synchronized and artificially inseminated with semen containing 0 (A, controls), 10(2) (B), 5 x 10(3) (C), 5 x 10(4) (D), and 5 x 10(5) (E) live N. caninum NC-1 isolate-tachyzoites, respectively. Experimental infection was followed for 100 days. Parasitaemia and specific serum IgG, and interferon-gamma (IFN-gamma) responses were studied. In experiment 2, four groups of 9, 10, 9, and 9 adult multiparous cows with confirmed infertility problems of diverse aethiology were hormonally synchronized and artificially inseminated with semen containing 0 (a, controls), 10(2) (b), 5 x 10(3) (c), and 5 x 10(5) (d) live N. caninum NC-1 isolate-tachyzoites, respectively. Experimental infection was followed for 63 days. Parasitaemia and specific serum IgG responses were studied. RESULTS: In experiment 1, parasitaemia was detected in 1, 2, and 3 heifers from groups B, C, and D, respectively, between 9 and 23 days after insemination. Persistent specific serum antibody responses were detected in 2 and 3 heifers from groups D and E, respectively. Transient specific serum antibody responses were detected in 2, 1 and 1 heifers from groups C, D, and E, respectively. In addition, 1 heifer from group B showed a serum-specific antibody level higher than cut off value at 21 days post-insemination. Heifers seroconverted between 23 and 47 days after insemination. Specific IFN-gamma levels were detected in 1, 4, 6, and 3 heifers from groups B, C, D, and E, respectively, between 9 and 55 days after insemination. Pregnancy rate in the control group (60%) was higher than those observed in inoculated heifers (0-42.9%). Pregnancy rates in inoculated heifers were lower when the tachyzoite dose was increased (B 42.9%, C 12.5%, D 11.1%, and E 0%). In experiment 2, no Neospora DNA in blood nor specific serum IgG to N. caninum were detected in any of the cows studied, except in one cow inoculated with 5 x 10(5) tachyzoites (group d) which showed a relative index x100 (RIPC) values of 9.4, 18.9, and 18.1 at 42, 56, and 63 days after insemination, respectively. CONCLUSIONS: This study provides evidence that the intrauterine infection via contaminated semen using 5 x 10(4) and 5 x 10(5) tachyzoites caused persistent serum-specific antibody responses in some heifers. On the basis of serological data, a dose-response effect was also observed. In addition, N. caninum would be a probable cause of early foetal death in inoculated heifers. In contrast, results obtained in a similar experiment with cows showing confirmed infertility indicate that higher doses, such as of 5 x 10(5) tachyzoites, were necessary to induce seroconversion in at least one animal.     

Experimental neosporosis in bulls: parasite detection in semen and blood and specific antibody and interferon-gamma responses

AIM: To investigate the presence of Neospora caninum in semen and blood, and the development of specific antibody and interferon-gamma (IFN-gamma) responses in experimentally infected bulls. METHODS: Eight bulls were intravenously infected with 10(8) live N. caninum tachyzoites of NC-1 isolate. The presence of N. caninum in semen and blood was assessed using a nested-PCR procedure. PCR-positive semen samples were bioassayed using a BALB/c nu/nu mouse model. Specific anti-N. caninum antibody and IFN-gamma responses were also examined. In parallel, eight seronegative bulls were studied as non-infected controls. All bulls were monitored for 26 weeks. RESULTS: All eight experimentally infected bulls showed N. caninum DNA in their semen and/or blood samples at some time during the course of the study. Parasite load in semen ranged from 0.1 to 14.5 parasites/ml (mean 6.0). N. caninum could not be detected in BALB/c nu/nu mice inoculated with PCR-positive semen samples. A significant increase in mean serum specific IgM antibody response to N. caninum was detected between 10 and 28 days post-infection (p.i.). Serum specific IgG, IgG1, and IgG2 antibody levels in experimentally infected bulls were significantly different after 21, 10, and 14 days p.i. as compared to controls, respectively. Specific anti-N. caninum IgG were detected in seminal plasma from infected bulls and values obtained were different from controls after 25 days p.i. Mean specific IFN-gamma responses in experimentally infected bulls were significantly higher than controls 3 days p.i. CONCLUSIONS: This is the first study to report the presence of N. caninum DNA in the semen and blood of experimentally infected bulls. Our observations indicate an intermittent presence of N. caninum in low numbers in semen and associated with chronic stage of the infection. This study is also the first to report the detection of anti-N. caninum IgG in seminal plasma of experimentally infected bulls.     

Transplacental Neospora caninum infection in cats

Transplacental transmission of Neospora caninum was studied in 2 pregnant cats (queens). Queen 1 was inoculated subcutaneously with 2 x 10(6) cell culture-derived N. caninum tachyzoites on day 47 of gestation. She gave birth to a full-term kitten on the 17th day after inoculation. The kitten died the second day after birth due to generalized N. caninum infection. The mother cat was killed on the third day after parturition and was found to have a macerated kitten in the uterus. Severe placentitis, metritis, hepatitis, and nephritis due to N. caninum were seen in tissues from the queen. Queen 2 was fed N. caninum tissue cysts and mated 111 days later. She gave birth to 3 healthy full-term kittens. The kittens were necropsied at 2, 22, and 30 days of age. Neospora caninum was recovered from the organs and was seen in histologic sections in 1 of the 3 kittens. Results indicate that N. caninum can be transplacentally transmitted in cats during acute and chronic stages of infection. Neospora caninum-specific IgG antibodies were demonstrated in the sera of inoculated cats and nursing kittens.     

Effects of re-infection with Neospora caninum in bulls on parasite detection in semen and blood and immunological responses

Three bulls with experimentally induced primary infection with Neospora caninum were re-infected intravenously with 10(8) live N. caninum tachyzoites of the NC-1 isolate at 300 days post-infection to investigate the presence of N. caninum in semen and blood, and the associated immune responses. In parallel, three bulls with experimentally induced primary infection with N. caninum and three non-infected bulls were also monitored. Re-infected and infected bulls showed an intermittent presence of N. caninum DNA in semen with a parasite load ranging from 0.1 to 15.6 (mean 4.4) and 0.1 to 11.1 (mean 4.1) parasites/ml, respectively. Re-infected bulls showed significant and persistent serum-specific IgM and IgG antibody responses. Specific IgG levels were detected in seminal plasma of all infected bulls, but the magnitude of the response was significantly higher in re-infected rather than in chronically infected animals. The mean specific IFN-gamma levels in re-infected bulls were significantly increased as early as 3 and 7 days after experimental infection when compared to bulls in other groups. This study showed that the intermittent presence and parasite load of N. caninum in the semen of re-infected bulls is very similar to that reported in chronically infected animals. The protozoa could not be isolated from BALB/c nu/nu mice inoculated with PCR-positive semen samples and inseminated heifers with pooled semen samples did not show seroconversion. Plasma IFN-gamma level seems to be a good indicator of a recent N. caninum infection.