Characterization and Therapy for Experimental Infection by Helicobacter mustelae in Ferrets

Background. Numerous clinical trials evaluating the efficacy of various antimicrobial compounds against Helicobacter pylori infection have been performed in humans. A convenient animal model for Helicobacter infection would facilitate the evaluation of novel therapies. These experiments were performed to evaluate the use of ferrets as a model of Helicobacter infection.
Materials and Methods. Ferrets were infected experimentally with Helicobacter mustelae and subsequently treated with bismuth subsalicylate (BSS) triple therapy (BSS, metronidazole, and amoxicillin), or left untreated. The status of infection and serology was assessed during treatment and for 8 weeks posttreatment. Seven ferrets successfully treated with triple therapy were challenged with H. mustelae and monitored for infection for an additional 5 weeks.
Results. Infection of ferrets by H. mustelae was accompanied by gastritis and a specific antibody response. Treatment of H. mustelae -infected ferrets with BSS suppressed bacterial growth in four of nine animals but did not eradicate infection. Triple therapy eradicated infection in all nine ferrets with a reduction in gastric inflammation. No relapse of infection occurred up to 8 weeks posttherapy. Challenge with H. mustelae of ferrets successfully treated with triple therapy resulted in a 100% rate of reinfection.
Conclusions. H. mustelae infection can be eliminated by triple therapy, but this does not result in protective immunity against reinfection by H. mustelae. This model, using a strain of Helicobacter indigenous to the host, may be useful for assessing therapeutic efficacy of novel therapies for the treatment of human infection by H. pylori.     

Loss of urease activity in Helicobacter mustelae

Urease-negative variants of Helicobacter mustelae were isolated after spontaneous loss of activity during sub-culture. The whole-cell protein patterns showed that the loss of urease activity was linked to the absence of two polypeptides of 29·1 and 65·4 kDa. Restriction endonuclease analysis of chromosomal DNA indicated no substantial differences between the urease negative and positive variants. It is likely that a change in the expression of the gene for urease was responsible for these observations.     

Non-motile mutants of Helicobacter pylori and Helicobacter mustelae defective in flagellar hook production

Flagellar hooks were purified from Helicobacter pylori and Helicobacter mustelae. The 70 × 16nm H. pylori hook was composed of FIgE subunits of 78kDa, while the 72 × 16nm H. mustelae hook was composed of 87kDa subunits. N-terminal sequence was obtained for the FIgH proteins of both species, and for an internal H. mustelae FlgE peptide. Degenerate oligonucleotide primers allowed amplification of a 1.2 kb fragment from the H. mustelae chromosome, which carried part of the flgE gene. The corresponding H. pylori gene was cloned by immunoscreening of a genomic library constructed in λZAP Express, The translated H. pylori flgE sequence indicated a protein with limited homology with the hook proteins from Salmonella typhimurium and Treponema phagedenis. Mutants of H. pylori and H. mustelae defective in hook production generated by allele replacement were non-motile and devoid of flagellar filaments but produced both flagellin subunits, which were localized in the soluble fraction of the cell. The level of flagellin production was unchanged in the mutants, indicating that the regulation of flagellin expression in Helicobacter differs from that in the Enterobacteriaceae.     

Preformed enzyme profiling of Helicobacter pylori and Helicobacter mustelae from human and animal sources

A total of 126 strains of Helicobacter pylori from human and animal (monkey, baboon and pig) gastric mucosa, and four strains of H. mustelae were biotyped using preformed enzyme profiles (API Zym). The strains were from 10 countries and they were predominantly biotype II (85%). The other three biotypes, which were less frequently encountered, were detected with similar frequency (4–6%). There was no evidence of geographical or host-associated biotype differences. Reference strains for each of the biotypes are available from the NCTC for inclusion in future studies.     

Genome conservation in Helicobacter mustelae as determined by pulsed-field gel electrophoresis

Abstract Genomic DNA from 15 strains of Helicobacter mustelae was subjected to pulsed-field gel electrophoresis (PFGE) after digestion with Pac I and S fi I. H. mustelae genome DNA appeared very similar in all strains examined, whether isolated from ferrets or mink or from animals bred in either the USA or in the UK. The H. mustelae genome size was estimated to be 1.7 Mb, similar in size to that of H. pylori . A minor difference in PacI PFGE pattern and genome size was observed between rifampicin-resistant and rifampicin-susceptible derivatives of H. mustelae F251. Another minor difference in genome pattern based on PFGE with S fi I was observed between an H. mustelae strain used to experimentally infect four ferrets which resulted in loss of an S fi I site in strains obtained from the newly infected ferrets. Thus, although minor differences in PFGE pattern were noted, H. mustelae lacks the genomic diversity observed in H. pylori .     

Phylogeny of Helicobacter felis sp. nov., Helicobacter mustelae, and related bacteria.

Strain CS1T (T = type strain) is a gram-negative, microaerophilic, urease-positive, spiral-shaped bacterium that was isolated from the gastric mucosa of a cat. Additional strains which possessed biochemical and ultrastructural characteristics similar to those of strain CS1T were isolated from the gastric mucosa of cats and dogs. The guanine-plus-cytosine content of the DNA of strain CS1T was 42.5 mol%. The 16S rRNA sequences of strain CS1T, strain DS3 (a spiral-shaped isolate from a dog), and Helicobacter mustelae were determined by direct RNA sequencing, using a modified Sanger method. These sequences were compared with the 16S rRNA sequences of Helicobacter pylori, "Flexispira rappini," Wolinella succinogenes, and 11 species of campylobacters. A dendrogram was constructed based upon sequence similarities. Strains CS1T and DS3 were very closely related (level of similarity, 99.3%). Two major phylogenetic groups were formed; one group consisted of strains CS1T and DS3, H. mustelae, H. pylori, "F. rappini," and W. succinogenes, and the other group contained the true campylobacters. The average level of similarity between members of these two groups was 84.9%. Within the first group, strains CS1T and DS3, H. pylori, and H. mustelae formed a cluster of organisms with an interspecies similarity level of 94.5%. The phylogenetic positions of W. succinogenes and "F. rappini" were just outside this cluster. On the basis of the results of this study, we believe that strains CS1T (= ATCC 49179T) and DS3 represent a new species of the genus Helicobacter, for which we propose the name Helicobacter felis.     

Inverse nickel-responsive regulation of two urease enzymes in the gastric pathogen Helicobacter mustelae

The acidic gastric environment of mammals can be chronically colonized by pathogenic Helicobacter species, which use the nickel-dependent urea-degrading enzyme urease to confer acid resistance. Nickel availability in the mammal host is low, being mostly restricted to vegetarian dietary sources, and thus Helicobacter species colonizing carnivores may be subjected to episodes of nickel deficiency and associated acid sensitivity. The aim of this study was to investigate how these Helicobacter species have adapted to the nickel-restricted diet of their carnivorous host. Three carnivore-colonizing Helicobacter species express a second functional urea-degrading urease enzyme (UreA2B2), which functions as adaptation to nickel deficiency. UreA2B2 was not detected in seven other Helicobacter species, and is in Helicobacter mustelae only expressed in nickel-restricted conditions, and its expression was higher in iron-rich conditions. In contrast to the standard urease UreAB, UreA2B2 does not require activation by urease or hydrogenase accessory proteins, which mediate nickel incorporation into these enzymes. Activity of either UreAB or UreA2B2 urease allowed survival of a severe acid shock in the presence of urea, demonstrating a functional role for UreA2B2 in acid resistance. Pathogens often express colonization factors which are adapted to their host. The UreA2B2 urease could represent an example of pathogen adaptation to the specifics of the diet of their carnivorous host, rather than to the host itself.     

Natural and experimental Helicobacter infections

Current information about Helicobacter infections in humans and various domestic, wild, and research animal species that have been used or have the potential to be used as animal models of human disease is presented. The Helicobacter genus now includes at least 26 formally named species, with additional novel species in the process of being characterized. The natural history, host range with zoonosis potential, pathology, and diagnostic techniques are presented, along with examples of how Helicobacter infection has interfered with unrelated in vivo research. Current recommendations for deriving and managing helicobacter-free animal colonies for research are provided.     

Identification and molecular characterization of a major ring-forming surface protein from the gastric pathogen Helicobacter mustelae

The spiral microaerophilic bacterium Helicobacter mustelae is linked to gastritis and gastric ulcers in ferrets. Electron microscopy of H. mustelae showed the presence of a laterally extensive array of 8.5-nm-diameter rings on the cell surface, which was shown to be composed of a 150 kDa protein. This protein was purified, and the sequence of 10 amino-terminal residues was determined. Polyclonal antibody against the purified 150 kDa protein labelled the ring structures on the homologous strain by means of immuno-gold. Cross-reactive proteins were identified in three H. mustelae strains, but not in Helicobacter pylori or Helicobacter felis. The hsr gene encoding this protein was cloned, and the protein expressed in Escherichia coli independently of vector promoters. The 1519-codon nucleotide sequence of the gene was determined, and comparison with the chemically derived protein sequence indicated a 47-residue leader pep-tide, and a mature protein with a molecular weight of 152 300. Thus the cell surface of H. mustelae differs markedly from other members of the genus Helicobacter in being covered by an array of 8.5 nm rings composed of a 150 kDa protein.     

Comparative ultrastructural and functional studies of Helicobacter pylori and Helicobacter mustelae flagellin mutants: both flagellin subunits, FlaA and FlaB, are necessary for full motility in Helicobacter species.

Helicobacter mustelae causes chronic gastritis and ulcer disease in ferrets. It is therefore considered an important animal model of human Helicobacter pylori infection. High motility even in a viscous environment is one of the common virulence determinants of Helicobacter species. Their sheathed flagella contain a complex filament that is composed of two distinctly different flagellin subunits, FlaA and FlaB, that are coexpressed in different amounts. Here, we report the cloning and sequence determination of the flaA gene of H. mustelae NCTC12032 from a PCR amplification product. The FlaA protein has a calculated molecular mass of 53 kDa and is 73% homologous to the H. pylori FlaA subunit. Isogenic flaA and flaB mutants of H. mustelae F1 were constructed by means of reverse genetics. A method was established to generate double mutants (flaA flaB) of H. mustelae F1 as well as H. pylori N6. Genotypes, motility properties, and morphologies of the H. mustelae flagellin mutants were determined and compared with those of the H. pylori flaA and flaB mutants described previously. The flagellar organizations of the two Helicobacter species proved to be highly similar. When the flaB genes were disrupted, motility decreased by 30 to 40%. flaA mutants retained weak motility by comparison with strains that were devoid of both flagellin subunits. Weakly positive motility tests of the flaA mutants correlated with the existence of short truncated flagella. In H. mustelae, lateral as well as polar flagella were present in the truncated form. flaA flaB double mutants were completely nonmotile and lacked any form of flagella. These results show that the presence of both flagellin subunits is necessary for complete motility of Helicobacter species. The importance of this flagellar organization for the ability of the bacteria to colonize the gastric mucosa and to persist in the gastric mucus remains to be proven.