|Home | About | Journals | Submit | Contact Us | Français|
Helicobacter pullorum is a bacterial pathogen in humans. By using microaerobic culture techniques, H. pullorum was isolated from the feces of barrier-maintained mice and identified, on the basis of biochemical, restriction fragment length polymorphism, and 16S rRNA gene sequence analyses. This finding presents an opportunity to study H. pullorum pathogenesis in mice.
Helicobacter pullorum, an enterohepatic Helicobacter species, has been isolated from humans (2, 3, 14-16) as well as avian species (1, 5, 14). Some of the human clinical isolates were from patients with enteritis, while others were from clinically normal individuals (3, 14, 16). Commercial broiler and layer chickens have exhibited vibrionic hepatitis and enteritis attributed to H. pullorum (14). Chickens experimentally infected with two human and two avian isolates of H. pullorum did not develop clinical signs, but both gross and microscopic cecal lesions were noted in experimentally infected chickens (4).
During routine microbiological testing, Helicobacter spp. were detected by PCR in the feces of C57BL/6NTac and C3H/HeNTac mice housed within one isolated barrier unit. Mice maintained in the affected barrier at Taconic, Inc., in Germantown, NY, were shipped to the Massachusetts Institute of Technology for evaluation. Fresh fecal samples from mice as well as cecum and colon specimens were aseptically collected at necropsy and cultured under microaerobic conditions, and bacterial isolates were subjected to biochemical characterization conditions as previously described (9). After 72 h of incubation, individual colonies from 15 C57BL/6NTac mice and 2 C3H/HeNTac mice were observed on Columbia blood agar and CVA plates containing cefoperazone, vancomycin, and amphotericin (BBL Campy agar; BBL, Sparks, MD). Analysis of samples with phase microscopy revealed spiral, motile bacteria. The bacteria were Gram negative and curved to spiral in shape. The bacteria were oxidase positive, catalase negative, urease negative, alkaline phosphatase hydrolysis negative, indoxyl acetate hydrolysis negative, and gamma-glutamyl transpeptidase negative. The isolates grew at 37°C and 42°C, but not at 25°C. Isolates were resistant to cephalothin (30 μg/disc) but sensitive to nalidixic acid (30 μg/disc). These biochemical and antimicrobial characteristics are identical to those of H. pullorum human isolate MIT 98-5489.
The sequences of the 16S rRNA genes of two isolates (MIT 09-6633 and 09-6634) cultured from mice feces were analyzed using previously published techniques (8). Phylogenetic trees were constructed by the neighbor-joining method (13). The total RNA sequences of the two isolates (accession numbers MIT 09-6634 and MIT 09-6635) were analyzed. Comparison of the 16S rRNA sequence with those of other bacteria in our database indicated that the H. pullorum isolate was most closely related to the H. pullorum isolate (NCTC 12826) of human origin. The next most closely related was an H. pullorum chicken isolate (NCTC 12824). The percent similarity values are indicated in the phylogenetic tree constructed with the neighbor-joining method (Fig. (Fig.1).1). Both H. pullorum isolates exceeded 99% similarity to NCTC 12826.
Helicobacter genus-specific primers C97 (5′-GCT ATG ACG GGT ATCC) and C05 (5′-ACT TCA CCC CAG TCG CTG) were used to amplify a 1.2-kb PCR product from the 16S rRNA gene from both mouse and goose fecal samples collected near the exterior of the barrier structure (7). All of the cultures from the 17 mouse samples were positive for Helicobacter-specific PCR with genus-specific primers (7). PCR-amplified 1.2-kb fragments (20 μl) were digested with 10 U of restriction endonucleases AluI and HhaI (New England BioLabs, Beverly, MA). Restriction patterns of the 1,200-bp Helicobacter genus-specific PCR product were compared to H. pullorum human isolate MIT 98-5489, H. typhlonius, H. bilis, H. muridarum, H. hepaticus, and H. rodentium. Three Helicobacter isolates from the Taconic mice were identical to the H. pullorum human isolate MIT 98-5489 (Fig. (Fig.2).2). The 16S rRNA gene sequence of the 1.2-kb Helicobacter spp.-specific DNA fragments from goose feces indicated 100% homology with H. brantae (9).
H. pullorum isolates were assayed for cytolethal distending toxin as previously described by our laboratory (6, 18). The cell sonicates of H. pullorum isolates and the H. cinaedi cytotoxin positive control strain were positive on HeLa83 cells for cytopathic effects consistent with changes seen with cdt activity. These effects included giant cell formation, nuclear enlargement and fragmentation, and multinucleation (Fig. (Fig.3).3). Primers specific for H. pullorum cdtB (F1, 5′-GTCTTTTGAGTGGATTGGATTCT-3′; R2, 5′-CACTCCGGGTGCTTGTGTAT-3′) were used to amplify a 148-bp fragment (11). H. pullorum cdtB-specific PCR indicated the presence of the cytolethal distending toxin gene in all five DNA samples from H. pullorum-infected mice. These DNA samples were obtained from two strains of H. pullorum isolated from infected mice and three DNA samples extracted from feces of two infected mice (Fig. (Fig.3).3). The liver and the entire gastrointestinal tract were fixed in 10% buffered formalin. All tissues taken were routinely processed, embedded in paraffin, sectioned at 5 μm, and stained with hematoxylin and eosin. Histological evaluation of the gastrointestinal system did not reveal any significant lesions.
In 1994, 16 Campylobacter-like organisms were isolated from both humans and chickens with gastroenteritis. In chickens, the organisms were isolated from the liver, duodenum, and cecum. 16S rRNA gene analysis, relative DNA homology, and total protein electrophoretic patterns identified seven of these bacteria to be of the same species, which was subsequently named H. pullorum (14).
H. pullorum also has been isolated from humans with gastrointestinal lesions (14, 18) and cultured from the stools of humans without clinical signs of enteritis (3, 18). H. pullorum has been identified in a male patient with diarrhea of 1-month duration who also had elevated liver enzymes (2). Avian species infected with H. pullorum, including commercial broiler and layer chickens, have been diagnosed with hepatitis and enteritis. Chickens experimentally infected with H. pullorum did not develop clinical signs, but gross and histologic cecal lesions were noted (4, 12, 14). It is unknown whether H. pullorum in humans is acquired by eating uncooked poultry, as is the case with Campylobacter jejuni acquired zoonotic infection (10, 17).
The presence of cytolethal distending toxin may play a role in H. pullorum pathogenesis, as reported for C. jejuni and H. hepaticus. That this bacterium can infect humans, avian species, and now mice suggests that mice can be used to study its pathogenesis. Utilizing the mouse model should provide a better understanding of the genetic components of H. pullorum responsible for pathogenicity and determine how this enteric pathogen can infect a broad diversity of species as well as induce lesions in susceptible mammalian species. The source of H. pullorum infection in the barrier-maintained mice is unknown. Personnel confidentiality agreements precluded screening of employees working in the barrier facility for enteric infection with H. pullorum. Analysis of possible sources for entry of H. pullorum yielded negative results. Helicobacter spp.-positive DNA from goose feces was determined to be H. brantae, previously isolated from Canada geese residing on the east coast of the United States (9). Whether H. pullorum persistently colonizes the feces of wild rodents and whether H. pullorum-infected rodents present a zoonotic risk require further studies.
This work was supported by the NIH grants R01 CA-067529, R01 DK-052413 (F. E. Dewhirst), and P30 ES-02109 (J. G. Fox).
We thank Melissa Mobley and Amanda Potter for technical assistance.
Published ahead of print on 10 March 2010.