Groups of female BALB/c mice infected by intravenous injection with 50 erythrocytes containing Plasmodium berghei Vincke et Lips, 1948 were sacrificed on days 3 through 12 after infection. Rheumatoid factor-like IgM (RF-IgM) and parasite-specific IgG levels were determined by enzyme-linked immunosorbent assay in serum specimens and in culture medium removed from spleen cell cultures established at sacrifice. All four mouse IgG subisotypes were recognized by RF-IgM molecules induced by Plasmodium berghei infection, and in this regard, the parasite-induced RF-IgM response resembled that induced by lipopolysaccharide polyclonal activation. Plasmodium berghei infection resulted in a biphasic RF-IgM response, with infected animals demonstrating significantly increased levels of RF-IgM early in the infection and significantly decreased levels late in the infection, compared to uninfected control mice. The decreased levels of RF-IgM observed late in infection correlated with increasing parasitaemia levels, and were primarily due to a decrease in RF-IgM specific for mouse IgG2a. Late infection levels of RF-IgM specific for IgGI, IgG2b, and IgG3 were not significantly different from those of control animals.

Cilia-associated respiratory (CAR) bacillus is an unclassified, gram-negative, extracellular bacterium that causes chronic respiratory tract disease in rodents. Infected mice develop microscopic lesions characterized by a primary lymphocytic response followed by macrophage and neutrophilic infiltration. To characterize the lymphocytic subsets that respond to CAR bacillus infection, BALB/c mice were inoculated with 10(5) CAR bacillus bacteria. At seven weeks after inoculation, mice were euthanized and the tracheobronchiolar and hilar lymph nodes were collected and stained for cell surface markers to T cells (CD3, CD4, and CD8), B cells (B220, CD5), natural killer (NK) cells (pan-NK) and intracellular interleukin 10 (IL-10) and interferon-gamma (IFN-gamma). Flow cytometric analysis of lymph nodes from CAR bacillus-infected mice revealed 11% increase in frequency of B cells (R220+), 12% increase in the frequency of double-negative (CD4-CD8-CD3+) T cells, and slight increase in the B-1 subset of B cells (B220+CD5+). There was no change in the frequency of NK cells. The CAR bacillus-infected mice had an overall decrease in the frequency of T cells. Intracellular cytokine staining revealed distinct populations of T cells producing IL-10 and IFN-gamma, and IL-10 production from B cells; NK cells were not a substantial source of IFN-gamma. To our knowledge, this is the first characterization of lymphocytic responses and suggestion that B cells and double-negative T cells may be principally responsible for the lesions associated with CAR bacillus infection.

Putative virulence factors including extracellular proteases, hemagglutinin, hemolysins, and soluble cytotoxins may play significant roles in the pathogenesis of trichomoniasis. The cytotoxicity, hemagglutinating, and hemolytic activity of Tritrichomonas foetus isolate ATCC #30003 and several field isolates were compared. All isolates were hemolytic toward mouse and bovine erythrocytes but not other tested species. The isolates varied significantly in hemagglutinating ability and cytotoxin production. A 40,000 Da soluble cytotoxin was partially purified and characterized. Chromatography separated cytotoxic activity from hemagglutinating and hemolytic activity but not from protease activity. However, protease assays indicated that protease activity was inversely correlated with cytotoxic activity. Characterization studies indicated that cytotoxic activity was destroyed by heat and acidic conditions but repeated freeze/thawing did not diminish activity. Target cell specificity assays showed Henle cells were twice as sensitive to the effects of the cytotoxin as Vero cells. These results suggest that T. foetus isolates vary in the production of virulence factors and produce a soluble relatively stable non-protease cytotoxic protein capable of killing cultured mammalian cells in vitro.

The cilium-associated respiratory (CAR) bacillus is a gram-negative, extracellular bacterium that causes persistent respiratory tract infections in rodents. We have previously demonstrated that BALB/c mice are more susceptible to CAR bacillus-induced disease than resistant C57BL/6 mice, with elevations in pulmonary gamma interferon (IFN-gamma) and interleukin (IL)-4. IL-10 is a type 2 cytokine that can increase host susceptibility to bacterial diseases through its anti-inflammatory effects, including suppression of macrophage function. The purpose of this study was to further describe the cytokine profiles associated with histologic lesions in CAR bacillus-infected mice and to assess the effects of cytokine depletion on the pathogenesis of disease. Six-week-old female BALB/c and C57BL/6 mice and mice with targeted mutations in IFN-gamma and IL-4 were inoculated intratracheally with 10(5) CAR bacillus organisms, and samples were collected at 6 to 7 weeks postinoculation. Lung samples were collected for histopathologic examination and analysis of cytokine mRNA. IFN-gamma, IL-10, and IL-4 mRNA levels in the lungs of infected mice were semiquantitatively measured using a reverse transcriptase-mediated PCR assay and compared to those in uninfected control animals of each strain. BALB/c mice infected with CAR bacillus had a median lung lesion score of 6 and IL-10 and IL-4 mRNA levels were significantly elevated. The majority of C57BL/6 mice were resistant to disease characterized by lung lesions scores of 2 or less and a dominant IFN-gamma mRNA cytokine profile. A few C57BL/6 mice with lesions scores of 5 or greater had elevations in all three cytokines and were susceptible to disease. C57BL/6 IFN-gamma knockout mice had increased disease with elevations in IL-10 and IL-4 mRNA, while BALB/c IL-4 knockout mice infected with CAR bacillus had a mild decrease in lesion severity and an attenuated IL-10 mRNA expression compared to wild-type BALB/c mice. These data indicate that IL-10 and IL-4 predominate in CAR bacillus-induced histologic lesions in mice, while IFN-gamma, may play a role in resistance to disease.

ABSTRACT The cilium-associated respiratory (CAR) bacillus is a gram-negative, gliding bacterium that causes persistent respiratory tract infections in rodents despite histologic and serologic evidence of a marked immune response. To assess humoral immunity and cytokine responses in CAR bacillus disease, 6-week-old female BALB/c and C57BL/6 mice were inoculated intratracheally with 10 5 CAR bacillus organisms. CAR bacillus-specific serum immunoglobulins (immunoglobulin M [IgM], IgG1, IgG2a, IgG2b, IgG3, and IgA) and local pulmonary cytokines (tumor necrosis factor alpha [TNF-α], gamma interferon [IFN-γ], and interleukin-4 [IL-4]) were evaluated by enzyme-linked immunosorbent assay every 7 days for 49 days. BALB/c mice developed CAR bacillus-induced lesions early in the course of disease that became more severe with time. Correlating with increasing disease severity, BALB/c mice had elevations in all antibody isotypes tested, and elevations in pulmonary TNF-α, IFN-γ, and IL-4. C57BL/6 mice developed mild lesions with mild increases in serum IgM, IgG1, IgG2b, and IgG3 levels and minimally detectable IgG2a and IgA. Cytokine perturbations were not detected in C57BL/6 mice. The persistence of infection in BALB/c mice with vigorous serum antibody responses and increased IFN-γ and IL-4 responses suggests that humoral immunity and T-cell responses are ineffective at preventing CAR bacillus disease. Furthermore, the lackluster antibody responses and undetectable cytokine responses in C57BL/6 mice suggest that humoral immunity and T-cell responses are not critical in resistance to CAR bacillus-induced disease.

Clostridium piliforme is an obligate intracellular bacterium that causes enterohepatic disease in many animal species. C. piliforme infections are commonly subclinical in laboratory rats and mice, and little is known about host regulation of disease or of the effects of C. piliforme infections on investigations that use subclinically infected animals. To assess host regulation of subclinical C. piliforme infections and the effects of those infections on laboratory mice, the expression of the pro-inflammatory cytokines tumour necrosis factor-alpha (TNF-alpha) and interferon-gamma (IFN-gamma) was evaluated at 0, 1, 3, 7, 14 and 28 days after inoculation with C. piliforme. Subclinical infection was induced in weanling C. piliforme-susceptible DBA/2 or -resistant C57BL/6 mice with either a toxic or a non-toxic C. piliforme strain. Hepatic lesions and bacteria were demonstrated histologically in both mouse strains for 14 days after inoculation with the toxigenic bacterial strain, but were never demonstrated histologically following inoculation with the non-toxigenic strain. Hepatic TNF-alpha and IFN-gamma mRNA and serum protein levels were similarly elevated in both mouse strains 1 day after inoculation with both C. piliforme strains, as evaluated by reverse transcription PCR and enzyme-linked immunosorbent assays, respectively. Elevation of IFN-gamma persisted for 14 days after inoculation; TNF-alpha remained elevated at 28 days after inoculation.

Clostridium piliforme is an obligately intracellular bacterium that causes enterohepatic disease in many domestic and laboratory animal species. Susceptibility to infection is known to vary with the host immune status, species and strain, but little is known about specific immune mechanisms that regulate this disease. Subclinical infection was induced in weanling C. piliforme-susceptible DBA/ 2 or resistant C57BL/ 6 mice with either a toxic or a non-toxic C. piliforme isolate. Hepatic lesions and bacteria were evident in both mouse strains for 14 days after inoculation with the toxigenic bacterial isolate, but were never demonstrated following inoculation with the non-toxigenic isolate. All mice demonstrated increased interleukin-6 (IL-6) levels that were largely independent of host strain susceptibility to infection or virulence of the bacterial isolate. The severity of C. piliforme-induced hepatic lesions was increased by polyclonal anti-IL-6 treatment in both resistant (DBA/ 2) and susceptible (C57BL/ 6) mouse strains. These data indicate that IL-6 is important in mediating the course of murine C. piliforme infections but is not involved in determining host susceptibility to acute infection, nor is it influenced by the virulence of the C. piliforme isolate.

Background and Purpose: Several rodent helicobacters have been associated with chronic active hepatitis or inflammatory bowel disease. Severe combined immunodeficient (SCID) mice appear to be inherently susceptible to disease attributable to these emerging pathogens. With the advent of polymerase chain reaction (PCR) analysis, it has become clear that several as yet unidentified Helicobacter species may also colonize rodents, but their capacity to cause disease is unknown.Methods: A Helicobacter species isolated from feces of a BALB/c mouse and provisionally named "H, typhlonicus" was used to inoculate helicobacter-free 4-week-old SCID mice (n = 11 males and 11 females). At various weeks after inoculation, mice were sacrificed and liver and intestinal specimens were collected for histologic examination and PCR analyses.Results: The C.B-17 scid/scid mice inoculated with "H, typhlonicus" developed moderate to severe proliferative typhlocolitis, similar to that seen in SCID mice infected with H, hepaticus or H, bills, However, in contrast to mice infected with H, hepaticus or H, bills, lesions of chronic active hepatitis were not detected in mice inoculated with "H, typhlonicus." A similar disease syndrome developed in SCID mice cohabitated with B6D2F1 mice naturally infected with a novel Helicobacter species that was genetically identical to "H. typhlonicus,"Conclusion: "Helicobacter typhlonicus" joins a growing list of helicobacters that are capable of inducing enteric disease in immunodeficient mice.

ABSTRACT Helicobacter hepaticus is a bacterial pathogen that causes chronic active hepatitis and inflammatory bowel disease in mice. The purpose of this study was to develop a recombinant antigen-based enzyme-linked immunosorbent assay (ELISA) to detect H. hepaticus -infected mice. A genomic library of H. hepaticus was constructed and was screened with sera from H. hepaticus -infected mice. A 459-bp open reading frame that coded for an 18-kDa immunoreactive protein, MAP18, was identified. The gene had high identity with genes coding for outer membrane proteins of other bacteria, and the predicted amino acid sequence of MAP18 had a putative membrane-trafficking signal sequence and a putative signal peptidase II cleavage site. The recombinant protein was expressed in Escherichia coli as a glutathione S -transferase (GST) fusion protein, GST-MAP18, and purified by affinity chromatography. The 44-kDa fusion protein was detected on Western blots probed with sera from H. hepaticus -infected mice but was not detected on blots probed with sera from mice infected with Helicobacter muridarum or Helicobacter bilis or with sera from mice free of Helicobacter infection. The GST-MAP18 fusion protein was used as an antigen in an ELISA to detect anti- H. hepaticus antibodies in sera from infected mice. This ELISA was compared to an H. hepaticus -specific ELISA that uses a detergent extract of H. hepaticus as the antigen. Sera from mice naturally and experimentally infected with H. hepaticus , H. bilis , or H. muridarum and sera from mice free of Helicobacter infection were evaluated. Both ELISAs performed with a high specificity (98%); however, the detergent extract-based ELISA performed with a higher sensitivity (89%) than the recombinant protein-based ELISA (sensitivity, 66%). These data indicate that H. hepaticus carries a gene that encodes an immunogenic 18-kDa membrane-associated protein; however, antibodies to this protein are not detected in all infected mice.

Kennett, M. J., Hook, R. R., Jr., Franklin, C. L., and Riley, L. K. 1999. Acanthamoeba castellanii: Characterization of an adhesin molecule. Experimental Parasitology92, 161–169. Acanthamoeba castellanii is a free-living protozoan that causes keratitis in humans and has been associated with pneumonia and granulomatous amebic encephalitis in dogs, sheep, and other species. Adherence of the Acanthamoeba to epithelial cells is critical to the pathogenesis of this disease. In this study, several mouse monoclonal antibodies (MAb) generated to whole Acanthamoeba trophozoites identified surface membrane epitopes by ELISA and IFA. Nine antibodies inhibited adherence of [35S]-methionine-labeled Acanthamoeba trophozoites to hamster corneal epithelial cells by 27–90%. Sodium periodate treatment, but not proteinase K digestion, of whole Acanthamoeba destroyed epitopes recognized by adherence-inhibiting antibodies such as MAb 7H6, suggesting that the adherence epitopes are carbohydrates. Other antibodies, MAb 2A8 for example, recognized surface membrane peptide epitopes that were proteinase K sensitive and sodium periodate resistant. Purified MAb 2A8 was used in an antigen-capture ELISA with peroxidase-labeled MAb 7H6 and demonstrated that the carbohydrate adhesion molecule was linked to the peptide recognized by MAb 2A8. Both MAbs 7H6 and 2A8 recognized a >207-kDa band on a Western blot of eluant from a MAb 2A8 immunoaffinity column, confirming that MAb 7H6 and MAb 2A8 recognize different epitopes on the same adherence molecule. MAbs 7H6 and 2A8 also identified the adhesion molecule in soluble Acanthamoeba membrane preparations and MAb 2A8 immunoaffinity column eluant by ELISA and Western blot. Neither of these antibodies were inhibited from binding to whole trophozoites nor membrane extracts by mannose or mannan in competitive binding assays. When our Acanthamoeba membrane preparations were electrophoresed and immunoblotted with α-d-mannosylated-biotin albumin, no bands were recognized in the >207 kDa range by our adherence-associated antibodies. These results suggest that the Acanthamoeba adhesin is not identical to the mannose binding protein of Acanthamoeba but rather is a distinct surface membrane glycoprotein.

Helicobacter hepaticus is a bacterial pathogen that causes chronic active hepatitis and inflammatory bowel disease in mice. The purpose of this study was to develop a recombinant antigen-based enzyme-linked immunosorbent assay (ELISA) to detect H. hepaticus -infected mice. A genomic library of H. hepaticus was constructed and was screened with sera from H. hepaticus -infected mice. A 459-bp open reading frame that coded for an 18-kDa immunoreactive protein, MAP18, was identified. The gene had high identity with genes coding for outer membrane proteins of other bacteria, and the predicted amino acid sequence of MAP18 had a putative membrane-trafficking signal sequence and a putative signal peptidase II cleavage site. The recombinant protein was expressed in Escherichia coli as a glutathione S -transferase (GST) fusion protein, GST-MAP18, and purified by affinity chromatography. The 44-kDa fusion protein was detected on Western blots probed with sera from H. hepaticus -infected mice but was not detected on blots probed with sera from mice infected with Helicobacter muridarum or Helicobacter bilis or with sera from mice free of Helicobacter infection. The GST-MAP18 fusion protein was used as an antigen in an ELISA to detect anti- H. hepaticus antibodies in sera from infected mice. This ELISA was compared to an H. hepaticus -specific ELISA that uses a detergent extract of H. hepaticus as the antigen. Sera from mice naturally and experimentally infected with H. hepaticus , H. bilis , or H. muridarum and sera from mice free of Helicobacter infection were evaluated. Both ELISAs performed with a high specificity (98%); however, the detergent extract-based ELISA performed with a higher sensitivity (89%) than the recombinant protein-based ELISA (sensitivity, 66%). These data indicate that H. hepaticus carries a gene that encodes an immunogenic 18-kDa membrane-associated protein; however, antibodies to this protein are not detected in all infected mice.

Clostridium piliforme induces enterohepatic disease in many domestic and laboratory animal species. Susceptibility to infection is known to vary with the immune status and strain of the host, but little is known about specific immune mechanisms that regulate this disease. To evaluate host control of C. piliforme infection, we examined the role of interleukin-12 (IL-12) both in the control of and in the response to murine C. piliforme infection. For this study, 3-week-old C. piliforme-resistant C57BL/6 or -susceptible DBA/2 mice were infected intravenously with either the toxic H1 or the nontoxic M1 C. piliforme isolate. Serum and liver samples were collected prior to C. piliforme inoculation (day 0) and at days 1, 3, 7, 14, and 28 postinoculation, Evaluation of hepatic IL-12 p40 mRNA expression by reverse transcription-PCR and of total-IL-12 protein levels in serum by enzyme-linked immunosorbent assay revealed that C. piliforme induced elevations in both hepatic p40 mRNA and serum total-IL-12 levels at all times postinoculation. Elevations were similar with both toxic and nontoxic C. piliforme isolates. Levels of total IL-12 in serum were significantly (P < 0.05) higher in C57BL/6 mice than in DBA/2 mice. Additional experiments were performed in which polyclonal antibody treatment was used to neutralize IL-12 in mice of both strains prior to intravenous inoculation with toxic C. piliforme H1-IL-12 neutralization increased the severity of Tyzzer's disease at day 3 postinoculation in both mouse strains, but the degree of increase was greater in C57BL/6 mice than in DBA/2 mice.

291 Helicobacter hepaticus is a newly recognized gram-negative, helical, microaerobic, pathogenic bacterium of mice (1). Helicobacter hepaticus has been associated with chronic active hepatitis in a variety of mouse strains (1–5), with hepatocellular tumors in A/JCr mice (2, 4), and with inflammatory large bowel disease in immunodeficient (6) and germ-free mice (5). Identification of H. hepaticus-infected mice is essential because H. hepaticus can confound or invalidate research data obtained from infected mice. Samples for diagnostic evaluation are routinely collected from colony mice or sentinel mice to survey mouse colonies for pathogenic agents. Sentinel mice are immunocompetent mice housed in direct or indirect contact with the population to be surveyed so that they will be exposed to infective agents harbored by colony mice. Indirect contact of sentinel mice is commonly used to expose sentinel animals to potential infective agents, and is achieved by regular transfer of soiled bedding from the population to be surveyed (7). Sentinel mice provide a readily available source of animals for health monitoring, avoiding the use of valuable colony mice for this purpose. Sentinel mice are also essential when serologic assays are used to monitor colonies of immunodeficient mice. The purpose of the study reported here was to determine whether sentinel mice could be used to detect H. hepaticus infections. To this end, we exposed mice free of Helicobacter infection to soiled bedding from H. hepaticus-infected mice. Exposed mice were monitored for H. hepaticus colonization by use of polymerase chain reaction (PCR) assay and for development of H. hepaticus-specific serum antibodies by use of enzyme-linked immunosorbent assay (ELISA). Our data indicate that sentinel mice can be used to identify H. hepaticus-infected mice. Four-week-old female C57BL/6NCr mice documented to be free of Helicobacter infections by results of fecal PCR were obtained (Fredrick Cancer Research and Development Center, Fredrick, Md.) and were used as sentinel and control mice. The C57BL/6NCr mice were from a colony that was documented to be free of ectoparasites, endoparasites, known bacterial pathogens, and relevant histologic lesions of the gastrointestinal tract, liver, lungs, and kidneys. Mice from the colony were also consistently seronegative for

PURPOSE. The protozoan Acanthamoeba produces a severe keratitis in a small percentage of people, especially contact lens-wearers. The purpose of this work was to develop and characterize an immortalized line of hamster corneal epithelial cells to be used in studies of the pathogenesis of Acanthamoeba keratitis. METHODS. Hamster corneal epithelial cells were maintained in primary culture and immortalized using simian virus 40 (SV40). Foci of transformed cells were cloned and subsequently characterized by phase-contrast microscopy and immunocytochemistry. Growth characteristics of the clone that were analyzed included loss of dependence on conditioned medium and ability to grow in soft agar. Cytotoxicity experiments were performed, to determine whether the selected clone was susceptible to Acanthamoeba infection in vitro. RESULTS. A cell line which exhibited epithelial morphology, as determined by phase contrast microscopy, was selected and cloned. Immunocytochemistry demonstrated the presence of keratin in the cloned cells, confirming the epithelial nature of the cell line. Immortalization was shown by loss of dependence on fibroblast-conditioned medium, ability to form colonies in soft agar and no apparent senescence following numerous passages in culture. This cell line was found to be sensitive to the cytotoxic effects of a pathogenic strain of Acanthamoeba. CONCLUSIONS. An immortalized line of hamster corneal epithelial cells was developed. This clone is susceptible to infection with Acanthamoeba and will be a useful tool with which to investigate the pathogenesis of Acanthamoeba keratitis.

Mouse monoclonal antibodies (MAbs) developed to a rat isolate (R-3) of cilia-associated respiratory (CAR) bacillus were used to assess antigenic relationships among three rat and five rabbit CAR bacillus isolates. Evaluation of MAbs by enzyme-linked immunosorbent assays (ELISAs) indicated that 87 of 241 hybridomas secreted CAR bacillus-reactive antibodies that could be grouped into four major groups. Group-I MAbs reacted with epitopes expressed by all CAR bacillus isolates and at least two or more nonrelated species of bacteria. Group-II, -III, and -IV MAbs reacted with only one or more of the rat CAR bacillus isolates; no MAbs reacted only with rat and rabbit CAR bacillus isolates. Western blot analyses indicated that 41-, 50-, and 105-kDa peptides of rat CAR bacillus isolates expressed rat CAR bacillus group- and isolate-specific epitopes. Hyperimmune anti-CAR bacillus antiserum and serum specimens from a CAR bacillus histologically positive mouse and rat also reacted with the 41-, 50-, and 105-kDa peptides. Sera from CAR bacillus histologically negative rats did not react with these peptides. These results suggest that the 41-, 50-, and 105-kDa peptides may represent suitable antigens for development of a specific ELISA for detection of rodent CAR bacillus infections. Furthermore, these data indicate that use of crude CAR bacillus preparations for either rat or rabbit CAR bacillus ELISAs is inappropriate.

The cilia-associated respiratory (CAR) bacillus is an unclassified, gram-negative, motile bacterium that has been implicated as an etiologic agent of respiratory disease in laboratory rodents. In the present study, approximately 1,200 bases of the 16S rRNA gene from three CAR bacillus isolates were sequenced. CAR bacillus-specific primers were designed on the basis of the 16S rRNA gene sequence and used in a PCR assay. The PCR assay detected as little as 500 fg of purified CAR bacillus DNA. The expected 267-bp DNA fragment was amplified from respiratory tissue of frozen, formalin-fixed, and paraffin-embedded samples from experimentally and naturally infected rats and mice. In contrast, no product was amplified from respiratory tissues of sham-infected experimental animals or animals that were serologically or histopathologically negative for the CAR bacillus. Our findings indicate that this PCR assay is a rapid, specific, and sensitive detection method for the diagnosis of CAR bacillus infection in rats and mice.

Helicobacter hepaticus is a newly recognized bacterium associated with chronic active hepatitis, hepatic carcinoma, and inflammatory bowel disease in mice. Currently, fecal or tissue PCR, fecal culture, or histologic examination of silver-stained liver sections is used to diagnose H. hepaticus infection. In this report, we describe an enzyme-linked immunosorbent assay (ELISA) for serodiagnosis of H. hepaticus infection in mice with a membrane digest preparation of H. hepaticus as the antigen. Sera from mice positive for H. hepaticus by PCR or histologic examination (n = 88), positive for Helicobacter bilis by PCR (n = 13), positive for other helicobacters (not identifiable to species level) by PCR (n = 25), or negative for all Helicobacter species by PCR (n = 162) were used to evaluate the ELISA. Results indicated that ELISA provided 93.2% sensitivity, 94% specificity, 87.2% positive predictive value, and 96.9% negative predictive value. Cross-reactive antibodies were detected in some mice infected with helicobacters not identifiable to species level. To further define ELISA sensitivity and specificity, groups of 10 C57BL/6 mice were inoculated per os with H. hepaticus, Helicobacter muridarum, or H. bilis. Sera were collected and examined by the ELISA. H. hepaticus-infected mice seroconverted by 2 weeks and maintained ELISA reactivity throughout the 18-week study, while mice infected with H. muridarum and H. bilis were negative by ELISA. These results indicate that this reported ELISA is highly sensitive and specific for the serodiagnosis of H. hepaticus infection in mice.

Clostridium piliforme infection (Tyzzer's disease) induces enterohepatic disease in many domestic and laboratory animals. Murine susceptibility to Tyzzer's disease varies with host strain, age, and immune status However, little is known about the role of the immune system in control of this disease. To investigate the role of host immunity in Tyzzer's disease, mice were depleted of either neutrophils, natural killer cells, or macrophages by antibody administration or chemotherapy. After depletion, DBA/2 mice, which are naturally susceptible to C. piliforme, or naturally resistant C57BL/6 mice were inoculated intravenously with C. piliforme. Animals were euthanized 3 days postinoculation and evaluated for gross and histologic lesions and hepatic bacterial load. In juvenile DBA/2 or C57BL/6 mice, depletion of either neutrophils or natural killer cells increased severity of disease. In adult mice, depletion of natural killer cells significantly increased severity of Tyzzer's disease in the resistant (C57BL/6) but not in the susceptible (DBA/2) strain. Macrophage depletion did not alter the course of infection in either mouse strain. These studies indicate an important role for neutrophils and natural killer cells in the pathogenesis of murine Tyzzer's disease. The role of macrophages in murine C. piliforme infection will require further evaluation.

A fecal PCR assay for detection of Helicobacter infections in laboratory rodents was developed. DNA was isolated from murine fecal pellets, and a region of the 16S rRNA gene conserved among murine Helicobacter species was amplified. The fecal PCR was sensitive and specific. This assay does not require euthanasia of rodents, which is especially important for valuable rodents, such as transgenic mice.