“
Candidatus Helicobacter heilmannii,” previously known as
Gastrospirillum hominis (
14), infects animals and humans, but the incidence is much lower than the incidence of
Helicobacter pylori in humans. “
Candidatus Helicobacter heilmannii” has not been reliably cultured to date, but it can be maintained in mouse stomachs (
2,
23). In humans, “
Candidatus Helicobacter heilmannii” infection has been associated with minimal gastritis, occasional gastroduodenal ulcers, and mucosa-associated lymphoid tissue (MALT)-type gastric B-cell lymphomas (MALT lymphomas) (
15). Both
H. pylori and “
Candidatus Helicobacter heilmannii” infections have been reported to be related to these MALT lymphomas, while the relative significance of
H. pylori and “
Candidatus Helicobacter heilmannii” for MALT lymphoma formation is controversial. Morgner et al. showed that 1.47% of “
Candidatus Helicobacter heilmannii”-infected patients and 0.66% of
H. pylori-infected patients had primary gastric MALT lymphomas (
15). On the other hand, Stolte et al. reported a milder form of gastritis and less common MALT lymphomas in “
Candidatus Helicobacter heilmannii” gastritis patients than in
H. pylori-infected gastritis and duodenal ulcer patients, who are known to have less gastritis (
26). Jhala et al. reported less gastritis and fewer lymphoid aggregates in “
Candidatus Helicobacter heilmannii”-infected patients, but their method to prove the presence of “
Candidatus Helicobacter heilmannii” depended solely on morphological observation using hematoxylin and eosin or Warthin-Starry silver staining or immunostaining with polyclonal antibodies against
H. pylori (
11). Suzuki et al. reported that nearly all MALT lymphomas were
H. pylori positive, but their diagnosis was dependent solely on the
H. pylori serum antibody titer, which could not distinguish between
H. pylori and “
Candidatus Helicobacter heilmannii” (
27).
In animal models, MALT lymphomas have been reported following infection with
Helicobacter mustelae in ferrets (
7), following infection with
Helicobacter felis in normal BALB/c mice (
6), following infection with
H. pylori in neonatally thymectomized BALB/c mice (
9), and following infection with “
Candidatus Helicobacter heilmannii” in normal BALB/c mice (
19,
21). In this study, we used the “
Candidatus Helicobacter heilmannii” previously obtained from a cynomolgus monkey to infect C57BL/6 mice for up to 24 months. We observed the development of low-grade MALT lymphomas in the fundic regions of the stomachs of mice infected with this pathogen. The characteristics of the pathological changes were determined by immunohistochemical, electron microscopic, and in situ hybridization methods.
DISCUSSION
There have been at least four reports of animal studies demonstrating the formation of gastric MALT lymphomas induced by
Helicobacter species. Erdman and colleagues described a case study of
H. mustelae-associated gastric MALT lymphoma in four ferrets, and they found two low-grade lymphomas and two high-grade lymphomas in the antrum (
7). Enno et al. reported observing
H. felis-associated gastric MALT lymphomas in normal BALB/c mice, and they found lymphoepithelial lesions in 88% of the animals and lymph follicles in 38% of the animals 22 months after infection (
6). Fukui and colleagues reported observing
H. pylori-associated gastric MALT lymphomas in neonatally thymectomized BALB/c mice, and they found lymph follicles and lymphoepithelial lesions in 60% of the mice in the absence of Bcl-X immunoreactivity (
9). In addition, Bcl-X-immunoreactive cells were found in these mice 6 and 12 months later. O'Rourke and colleagues reported finding “
Candidatus Helicobacter heilmannii”-associated gastric MALT lymphomas in normal BALB/c mice; furthermore, 15 months after infection with “
Candidatus Helicobacter heilmannii”-like isolates from bobcats and other animals, they reported finding lymphoepithelial lesions in up to 80% of the mice, and at 18 months after infection they found lymphomas in up to 25% of the mice (
19). The present study clearly demonstrated that “
Candidatus Helicobacter heilmannii” infection with cynomolgus monkey mucosal and mucous homogenates resulted in MALT lymphoma formation in 50% of the C57BL/6 mice 3 months after infection and in 100% of the animals 6 months after infection; therefore, this could be a useful animal model of gastric MALT lymphomas, provided that these bacilli can be successfully cultured. In addition, the “
Candidatus Helicobacter heilmannii” used in the present experiments was passaged in mice for up to 12 years, and it could possibly be adapted to mice to increase virulence.
H. pylori infection has been thought to be a prerequisite for the formation of MALT lymphomas, and chronic inflammation is probably a factor leading to malignant transformation via chronic stimulation of the lymphoid tissue (
5,
11,
26,
27). However, the diagnosis of
H. pylori infection in these previous reports was dependent on pathological data showing the presence of bacteria that were longer and more spiral than
H. pylori obtained by hematoxylin and eosin or Warthin-Starry silver staining and immunostaining with polyclonal antibodies against
H. pylori. Now that the 16S rRNA gene sequence of “
Candidatus Helicobacter heilmannii” is available, as it has been since 2003, the previous conclusions should be reevaluated by performing PCR analyses.
The difference between the length of the period before MALT lymphoma occurrence in the study of O'Rourke et al. (
19) and the length of the period in our study may have been due to differences between the experimental animals and the bacteria used. When the animal species used in the studies are considered, the differences may have resulted from the Th1/2 status of the Th1-predominant C57BL/6 mice and the Th2-predominant BALB/c mice. In this context, the report of Fukui and colleagues (
9) is of interest, because these workers demonstrated the formation of MALT lymphomas induced by
H. pylori in neonatally thymectomized BALB/c mice, which suggested that Th1 predominance was important in the formation of MALT lymphomas; however, in that study, Fukui et al. did not provide an explanation for why MALT lymphomas did not form in normal C57BL/6 mice, which are neonatally Th1 predominant.
Another common feature of “
Candidatus Helicobacter heilmannii”-infected C57BL/6 mice and thymectomized BALB/c mice is the hypoacidity induced by parietal cell damage. In neonatally thymectomized BALB/c mice, autoimmune gastritis has been found (
18). Administration of
H. pylori to these mice decreased parietal cell damage, but antiparietal cell antibodies were still detectable. In the present study, direct invasion of parietal cell intracellular canaliculi and the cytoplasm by “
Candidatus Helicobacter heilmannii” was observed. Several previous reports demonstrated this characteristic of “
Candidatus Helicobacter heilmannii” infection in rhesus monkeys and C3H mice, although identification of the bacteria was based on morphology (
4), which could have a tremendous effect on parietal cell function; it is well known that parietal cells change their cytoskeletal structure according to acid secretion status (
8). When differences between bacillus species are considered, more prompt and effective formation of MALT lymphomas by “
Candidatus Helicobacter heilmannii” is expected. O'Rourke and colleagues used “
Candidatus Helicobacter heilmannii”-like isolates whose DNA sequences have yet to be identified (
20), whereas we used genetically well-defined “
Candidatus Helicobacter heilmannii,” which was classified as having cluster 1 16S rRNA and cluster A urease genes, although it should be noted that in the present study, the bacilli were obtained from a cynomolgus monkey stomach.
We previously observed a sustained increase in myofibroblasts (i.e., activated fibroblasts) in “
Candidatus Helicobacter heilmannii”-infected C3H mice (
16); this finding is in good agreement with the bacterial copy number observed in the present study. The continuous influence of “
Candidatus Helicobacter heilmannii” on the gastric mucosa may also be related to the formation of MALT lymphomas.
With regard to the mechanism of formation of gastric MALT lymphomas, soluble FAS antigen and apoptosis have been reported to be associated with the malignancy grade of MALT lymphomas. Thus, the relationship between parietal cell damage and apoptosis still needs to be clarified. In situ hybridization revealed the presence of some bacilli in the lymphoma tissues, which was suggestive of a direct influence of this bacterium or its DNA on the formation of lymphomas. Relevant to this point, Serna et al. (
22) reported the presence of
Helicobacter-like organisms within the lymphoid follicles in cats, suggesting that the bacteria could serve as a stimulus for the development of lymph follicles.
The immunological cross-reactivity of
H. pylori with “
Candidatus Helicobacter heilmannii” (
11) and “
Candidatus Helicobacter suis” is well known and has been considered one of the criteria for identification of these helicobacters (
20). The present study also clearly revealed the cross-reactivity of “
Candidatus Helicobacter heilmannii” with the anti-
H. pylori antibody.