At present, preharvest Salmonella monitoring programs are primarily based on serological analysis, an inexpensive method that allows rapid screening of high numbers of pig herds . The onset of seroconversion after Salmonella infection in pigs is difficult to predict and can range from 6 days up to 25 weeks after oral inoculation, depending on, among others, the inoculation strain and inoculation dose used and piglet age [7, 9, 33, 34]. Yet, no data are available about the underlying mechanisms that account for these differences.
Seroconversion is initiated when antigen-presenting cells like macrophages and dendritic cells present pathogen-derived peptide antigens to T cells in 2 different ways. In the endogenous pathway, polypeptides that access the host cell’s cytosol are degraded by the proteasome and are presented at the surface of antigen presenting cells through MHC I molecules. Salmonella, however, enters the host cell through the endocytic route, causing its peptides to be presented via MHC II molecules after lysosomal degradation [35, 36]. Antigens bound to MHC II are presented to naïve CD4+ helper T cells that will develop into effector T cells. The latter than differentiate into two major subtypes of cells known as type 1 and type 2 helper T cells (Th1 and Th2 cells, respectively), depending on, among others, the chemo- and cytokine environment. Th1 cells predominantly activate macrophages while Th2 cells will induce maturation of B cells to antibody producing plasma cells. Because the MHC II antigen presenting pathway is able to mount both a cellular and/or humoral immune response, bacteria that interfere with the MHC II expression might radically influence the host’s immune response towards infection. Generally, the expression of MHC II molecules on the surface of macrophages is upregulated following ingestion of microorganisms and recognition of their foreign molecular pattern . We found that Salmonella Typhimurium strain 112910a specifically downregulated MHC II expression on the surface of macrophages and that the MHC I level of macrophages remained unaffected 24 h post inoculation (pi). These results suggest that Salmonella specifically interferes with the MHC II presentation pathway in porcine macrophages, leaving the MHC I pathway undisturbed. Furthermore, downregulation of MHC II expression on porcine macrophages turned out to be Salmonella specific, since we found that an Escherichia coli strain, another closely to Salmonella related member of the Enterobacteriaceae family, was unable to induce MHC II downregulation. Inoculation of bovine macrophages with Salmonella Typhimurium did not cause downregulation of MHC II expression . Although porcine alveolar macrophages represent a quite specific model which might explain their differential behaviour compared to macrophages or other cell types from other host species, our findings might provide further evidence for host-specific Salmonella behaviour.
Salmonella Pathogenicity Islands (SPIs’) are clusters of genes encoding virulence factors that play an important role at different stages of the pathogenesis of Salmonella infections in pigs . We examined whether SPI-1 and/or SPI-2 were involved in the observed Salmonella induced downregulation of MHC II expression on macrophages. Salmonella Typhimurium SPI-1 plays a crucial role in the colonization and invasion of the porcine gut , while SPI-2 is predominantly involved in replication and survival of Salmonella Typhimurium in porcine macrophages . HilA and SsrA/B are major regulators of SPI-1 and SPI-2, respectively, and both regulators induce transcriptional activation of downstream genes in response to a variety of stimuli [23, 40]. We found that Salmonella Typhimurium induced downregulation of MHC II expression was at least partly dependent on SPI-1 and SPI-2. Since recent evidence shows that SPI-1 is also involved in intracellular behaviour of Salmonella in macrophages, besides its main role in cell invasion , the fact that MHC II downregulation is inhibited when SPI-1 is abolished is not a complete surprise. Both SPI-1 and SPI-2 encode a type 3 secretion system, T3SS-1 and T3SS-2 respectively, a needle-like structure that is used to inject so called ‘effector proteins’ into the host cell’s cytosol, for uptake of the bacterium or to adapt the intracellular environment in function of Salmonella survival. Mitchell and colleagues already showed that SPI-2 effector SifA plays a role in downregulation of MHC II expression on human Mel JuSo cells, possibly by recruiting vesicles that transport MHC II molecules to the cell membrane, to maintain the integrity of the Salmonella containing vacuole (SCV) . In support of the hypothesis of redistribution of MHC II molecules from the cell surface to the SCV, our data from the dot blot assay suggest no change in the total amount of MHC II molecules in uninfected and Salmonella infected macrophages 24 h post inoculation. It is likely that SPI-1 and SPI-2 secreted effector proteins cause the Salmonella induced downregulation of MHC II expression on porcine macrophages. Altogether, we showed an explicit role for SPI-1 and a more discreet role for SPI-2 in downregulation of MHC II expression on porcine macrophages. However, because MHC II expression levels in ΔhilA and ΔssrA/B infected macrophages were not completely restored, other factors than SPI-1 and SPI-2 might be involved as well.
It has commonly been assumed that resistance to facultatively intracellular bacteria like Salmonella Typhimurium predominantly requires a cellular immune response. However, there appear to be multiple mechanisms by which antibodies can influence the course of infection with such pathogens . We found that, when Salmonella Typhimurium strain 112910a was opsonized with Salmonella specific antibodies prior to inoculation of macrophages, the bacterium lost its capacity to interfere with MHC II expression and was less able to proliferate in porcine macrophages. A proposed mechanism by which antibodies might interfere with Salmonella virulence is by sterically hindering the proper insertion of T3SS-1 and/or T3SS-2, since both secretion systems are required for successful intracellular proliferation of Salmonella.
Salmonella can actively invade macrophages using T3SS-1 or can be taken up passively through Fc-receptor mediated phagocytosis . Besides an effect through sterical hindrance of T3SS, it is thus possible that opsonization of Salmonella with serum antibodies favoured phagocytosis over Salmonella controlled invasion. It was found that when Salmonella was opsonized prior to inoculation of macrophages, approximately 2 to 2.5 times more bacteria were internalized than in macrophages that were inoculated with non opsonized Salmonella. Salmonellae internalized into macrophages reside within Salmonella containing vacuoles (SCV) and the environment in these SCV after Fc-receptor mediated phagocytosis is different from that after T3SS-1 dependent invasion . These different environmental changes might subsequently result in differences in the expression of hilA and other Salmonella genes, and in this way alter the intracellular behaviour of the microorganism [32, 43, 44]. Since we showed that downregulation of MHC II expression is SPI-1 and SPI-2 dependent, differential gene expression caused by the used internalization mechanism might affect the extent of Salmonella interference with the MHC II expression on macrophages. The mechanism of Salmonella internalization might furthermore affect the macrophage activation level. Indeed, macrophages that phagocytosed Salmonella that was opsonized with Salmonella specific antibodies produced more ROS than macrophages infected with bacteria that were not opsonized, or opsonized with negative pig serum, indicating increased macrophage activation . Furthermore, enhanced ROS production can result in growth inhibition of intracellular bacteria and even destruction of these bacteria , which might explain why Salmonella was less able to proliferate intracellularly after opsonization with Salmonella specific antibodies. Inhibition of downregulation of MHC II expression when Salmonella was opsonized with antibodies prior to inoculation of macrophages might, in addition to sterically hindering the proper insertion of T3SS-1 and/or T3SS-2, thus be the result of enhanced phagocytosis and/or ROS production and subsequent changes in Salmonella virulence gene expression. The findings that opsonization of Salmonella with specific antibodies prior to inoculation of porcine macrophages limits intracellular proliferation and inhibits downregulation of MHC II expression emphasize the importance for Salmonella to circumvent antibody production for successful persistence in the pig host.
Since strain and host specific behaviour is a key characteristic of Salmonella pathogenesis [46, 47], we determined if other Salmonella isolates exhibited the same phenotype as strain 112910a. We found that the extent of MHC II downregulation differed considerably among various Salmonella strains. Salmonella Typhimurium strain MB2216, a pig isolate, and strain DAB69, a pigeon isolate, and serovar Infantis and Derby strains, both pig isolates, exhibited no downregulation of MHC II expression in the current setup.
Subsequently, it was determined if the strain specific differences in MHC II downregulation capacity were inherent to macrophage infection by that particular strain, or correlated with Salmonella invasion capacity in, Salmonella induced cytotoxicity on and ROS production by macrophages. The Salmonella invasion capacity of the tested Salmonella strains did not significantly differ, except for strain MB2498 that was significantly impaired in infecting macrophages. This might be due to the fact that this strain was markedly less motile than the other strains, as assessed by growing the strains in semi solid agar [48, 49]. Infection with all tested Salmonella strains was cytotoxic for macrophages 24 h pi. Finally, the macrophage activation status indicated by ROS production measurement did not differ between Salmonella strains. Because we found no significant correlation between the extent of MHC II downregulation on one hand, and Salmonella invasiveness, cytotoxicity or induction of ROS production on the other hand, we can conclude that strain specific differences in the capacity to downregulate MHC II expression on porcine macrophages are irrespective of strain specific differences in invasion capacity and cytotoxic effects on or induction of ROS production by macrophages.
Since we showed that SPI-1 and SPI-2 are involved in Salmonella Typhimurium strain 112910a induced downregulation of MHC II expression, we wanted to examine if differential SPI-1 and/or SPI-2 gene expression accounted for the observed differences in MHC II downregulation between Salmonella strains. Surprisingly, no correlation was found between the expression of SPI-1 genes hilA and sopB for a certain strain, although this can be due to the fact that expression of the effector protein SopB is only partly regulated by HilA . Furthermore, no sopB expression was detected in Salmonella MB2498 and Derby strains, indicating the absence of the sopB containing SPI-5 in these strains [13, 50]. Expression of SPI-2 genes ssrA/B and ssaH markedly differed between Salmonella strains, and in contrast to hilA and sopB expression, a significantly positive correlation was found between ssrA/B and ssaH expression in a certain strain. However, no correlation was found between SPI-1 or SPI-2 gene expression and the capacity to downregulate MHC II expression. Although Salmonella gene expression in LB medium and inside macrophages might vary, the fact that SPI-1 and SPI-2 gene expression patterns of different Salmonella strains did not correlate with their capacity to interfere with the MHC II expression level of macrophages might indicate that other factors than SPI-1 and SPI-2 are also involved in Salmonella induced MHC II downregulation.
In conclusion, we showed that Salmonella Typhimurium strain 112910a specifically downregulated MHC II expression on porcine macrophages in merely a SPI-1 and SPI-2 dependent way. Furthermore, opsonization of Salmonella with antibodies counterbalanced the bacterium’s capacity to downregulate MHC II expression and to proliferate intracellularly, emphasizing the importance for Salmonella to interfere with the host’s immune response for successful persistence. The extent of MHC II downregulation on macrophages differed among Salmonella strains and these strain specific differences did not correlate with differential SPI-1 and/or SPI-2 expression, providing evidence that besides SPI-1 and SPI-2 other factors are also involved in Salmonella induced downregulation of MHC II expression. Our findings might imply that Salmonella strains that are capable of circumventing the pig’s immune response can better persist in pigs, interfering with serological screenings that are used in most preharvest Salmonella monitoring programs in Europe to date.