In the present study, all 9 H. suis strains were able to persist in the stomach of 2 different mouse strains at relatively high colonization levels compared to H. pylori. Most likely, colonization with H. suis is achieved more easily compared to H. pylori, which often requires prior adaptation to mice. This was underlined by the higher colonization rates observed for the mouse-adapted H. pylori SS1 strain, compared to the parental pre-mouse SS1 (pMSS1) strain [28, 29]. The ease by which mice become infected with H. suis warrants further research on the role of mice in the epidemiology of H. suis infections in pig herds.
In general, H. suis-infected BALB/c mice showed lower colonization rates compared to C57BL/6 mice. A discrepancy between the colonization of BALB/c and C57BL/6 mice has also been described for other gastric helicobacters such as H. felis and H. pylori[21, 30, 31]. The higher inflammatory response, as observed histologically in the stomach of BALB/c mice infected with H. suis, may create a hostile and unfavourable environment for the bacterium, as has been suggested previously . One possible factor contributing to the lower degree of H. suis colonization observed in BALB/c mice, could be the more pronounced Th17 response in this mouse strain. Although in both mouse strains an infection with H. suis induced an upregulation of IL-17 mRNA, the response was significantly higher in BALB/c mice. A clear inverse correlation between IL-17 mRNA and IL-6 mRNA (promoting Th17) expression levels on the one hand and H. suis colonization on the other hand was observed in this study, which largely corresponds to what has been described for H. pylori. Indeed, most studies reveal a negative correlation between a fully functional Th17 response and H. pylori colonization [33, 34], although some authors suggest the opposite .
Besides a Th17 response, H. pylori has been shown to upregulate the Th1 response both in humans and in mouse models [35–37]. This was indeed confirmed in this study with H. pylori strain pMSS1, although levels of IFN-γ upregulation were increased only mildly. Interestingly, no increased expression of IFN-γ, a signature Th1 cytokine, could be observed in H. suis-infected animals in the present study, neither in BALB/c mice nor in Th1-prone C57BL/6 mice [24, 25]. In contrast, most H. suis strains caused an upregulation of the Th2 signature cytokine IL-4, which was most pronounced in BALB/c mice. A similar upregulation was absent in H. pylori-infected mice, clearly highlighting the differences in the immune response elicited in mice infected with H. pylori (Th17/Th1) compared to mice infected with H. suis (Th17/Th2). Moreover, the results obtained in this study correspond to the histological changes observed in mice and Mongolian gerbils infected for up to 8 months with H. suis strain 5 . In this study, only in BALB/c mice infected with HS5 for 8 months, an increased B cell accumulation was observed, further underlining the involvement of a Th2-polarized response . Interestingly, a Th2 response, rather than a Th1-predominant response, has been associated with the development of low-grade B cell MALT lymphoma [38, 39], which has indeed been associated with gastric non-H. pylori Helicobacter infection, including H. suis.
As discussed above, the overall inflammatory response, including the Th17 response, is stronger in H. suis-infected BALB/c mice, compared to C57BL/6 mice. IL-10, often produced by regulatory T cells (Treg’s), is a suppressive cytokine that acts through downregulation of several pro-inflammatory cytokines and the resulting decrease of inflammatory cell recruitment . For H. pylori, it has been shown that infection suppresses the effective induction of H. pylori–specific Th17 immunity through the induction of a Treg response . A positive correlation between expression of IL-10, a suppressive cytokine, and H. suis colonization could be observed only in BALB/c mice, although similar levels of IL-10 expression were observed in H. suis-infected mice from both strains. So most likely, other mechanisms, such as the difference in IL-6 expression, are involved since this cytokine promotes a Th17, as well as a Th2 response, both of which are more pronounced in BALB/c mice [34, 41].
In a previous experimental infection study in mice with H. suis strain HS5, we observed only a mild increase of macrophage infiltration in the corpus of the stomach at 21 and 63 days after infection, but not at 8 months post infection . In the present study, no significant upregulation of TNF-α and IL-12 expression was observed in animals infected with H. suis for 59 days. Possibly, the absent upregulation of these macrophage-secreted cytokines reflects the return of macrophage infiltration to baseline levels after a possible mild increase during the initial weeks of H. suis infection.
Only in C57BL/6 mice, an IL-1β upregulation was observed in H. suis-infected animals. It has been shown that this pro-inflammatory cytokine also plays an important role in the inhibition of gastric acid secretion and the development of severe hyperplastic and dysplastic glandular changes [42, 43]. However, previous experiments showed that these lesions do not develop in C57BL/6 nor BALB/c mice after long-term (8 months) infection with H. suis. Instead, lymphoid tissue lesions develop both in mice and Mongolian gerbils, which contrasts to the predominant metaplastic/dysplastic changes observed during long-term H. pylori infection [23, 44].
Although all H. suis strains included in this study induced a similar immune response, some remarkable differences could be observed. Most H. suis strains caused no significant upregulation of IL-4 in C57BL/6 mice, although this was the case for HS1. In addition, HS1 induced the strongest upregulation in this mouse strain of IL-6, involved in Th17 and Th2 differentiation. On the other hand, only one strain, HS6, showed a complete lack of IL-4, IL-6 and IL-10 upregulation in both mouse strains. This clearly shows that mild strain differences exist in the immune response evoked by H. suis in the murine host. Although this is most likely also the case in infected humans and pigs, this remains to be investigated.
For H. pylori, differences in the host response evoked by different strains have been associated with the presence, absence or functionality of several virulence-associated genes, for instance those belonging to the cytotoxin-associated genes pathogenicity island (cag PAI) and the vacuolation cytotoxin A encoding gene vacA[45–47]. H. suis lacks most of the genes of the cag PAI and does not produce a functional vacuolating cytotoxin A . One of its main virulence factors, capable of modulating lymphocyte function through its effects on lymphocyte proliferation and cytokine secretion, is the gamma-glutamyl transpeptidase . It remains to be determined if differences in production of this enzyme by different H. suis strains, as has indeed been observed in vitro , play a role in the mild variation of the immune response described in the present study.
In conclusion, all 9 H. suis strains were capable of colonizing mice, but the numbers of H. suis bacteria were lower in the stomach of BALB/c mice. Although differences between H. suis strains were observed, colonization generally caused a predominant Th17 response, mainly in BALB/c mice, accompanied by a less pronounced Th2 response for most H. suis strains. This contrasts with the immune response induced by H. pylori infection, characterized by a Th17/Th1 response and the absence of a Th2 response. Despite a clear immune response evoked in the murine host, infection persisted in all H. suis-inoculated animals.