Over the past few years, many observations of overwhelming post splenectomy bacterial infections have been reported. Streptococcus pneumoniae is the aetiologic agent in about 80% of cases, but also gram-negative bacteria are involved in the development of fatal infections in splenectomized patients. Functionally, the spleen plays a fundamental role in bacterial clearance either by antibody response or macrophage bactericidal capacity. At the same time, there is evidence that the spleen also contributes to bacterial endotoxin detoxification. Finally, the mechanisms responsible for gram-positive and gram-negative sepsis in the splenectomized host and possible therapeutical approaches able to neutralize bacterial products endowed with noxious effects are discussed.

The pathophysiology of hypergastrinemia in H. pylori infection has been largely investigated and different reports clearly show that the infected antrum has a marked inflammatory response with a suggestive local production of cytokines. Notwithstanding, a few data are available on the circulating levels of cytokines and gastrin in the asymptomatic people carrying H. pylori infection. Thus, aim of the study was to evaluate circulating proinflammatory cytokines [Interleukin (IL)-8, Interleukin (IL)-10, Interferon (IFN)-gamma, and Tumor Necrosis Factor (TNF)-alpha] and gastrin levels in H. pylori positive asymptomatic subjects vs. H. pylori negative ones. To this end, thirty healthy volunteers with no digestive symptoms or systemic disease were enrolled and H. pylori infection was identified by a 13C-urea breath test. Plasma levels of gastrin were determined using the RIA kit whereas IL-8, TNF-alpha, IL-10, and IFN-gamma levels in serum were measured with a solid-phase ELISA. Fifteen infected people showed significantly higher gastrin and TNF-alpha levels than uninfected subjects. On the contrary, IL-8 levels were significantly higher in the uninfected subjects than in H. pylori positive ones (P < 0.0422). IFN-gamma and IL-10 circulating levels were not affected by H. pylori presence, being not significantly different in the two groups.

Endotoxins or lipopolysaccharides (LPS), major components of the cell wall of Gram-negative bacteria, once released from the bacterial outer membrane bind to specific receptors and, in particular, to a membrane-bound receptor, the CD14 (mCD14) and the toll-like receptor 4 present on monocytes/ macrophages. In turn, LPS-activated monocytes/ macrophages release in the host tissue an array of so-called proinflammatory cytokines and, among them, Tumor Necrosis Factor (TNF)-alpha, interleukin (IL)-1beta, IL-6, IL-8 and IL-12 are the major mediators. Before therapy (To) and at the end of 6-month interferon (IFN)-alpha/Ribavirin (RIB) treatment (T6), circulating endotoxin levels were measured in responder and non responder HCV+ patients. At T0, 57% of the non responders were endotoxin-positive and had, on average, 54 pg/ml of plasma LPS while in 50% of the responder patients endotoxin were found with an average of 29 pg/ml. At T6, in responders LPS were no longer detectable, while in 42% of the non responders LPS were found (average levels 45 pg/ml). In terms of serum cytokine concentration, at T6 IFN-gamma levels when compared to those detected at T0 were increased in both endotoxin-positive and endotoxin-negative patients. However, at T6 IL-10 concentration was significantly increased only in the group of endotoxin-negative subjects (responder patients), in comparison to T0 values. The origin of endotoxemia in HCV+ patients seems to be multifactorial, likely depending on impaired phagocytic functions and reduced T-cell mediated antibacterial activity. In these patients, however, one cannot exclude the passage of LPS from the gut flora to the blood stream, owing a condition of altered intestinal permeability. At the same time, a less efficient detoxification of enteric bacterial antigens at the hepatic level should be taken into consideration. Finally, novel therapeutic attempts aimed to neutralize LPS in the host are discussed.

Ulcerative colitis (UC) and Crohn's disease (CD) [inflammatory bowel disease (IBD)] are both characterized by an exaggerated immune response at the gut associated lymphoreticular tissue level. Such an abnormal and dysregulated immune response may be directed against luminal and/or enteric bacterial antigens, as also supported by murine models of inflammatory bowel disease (IBD) caused by organisms such as Citrobacter rodentium and Helicobacter hepaticus. Bacterial endotoxins or lipopolysaccharides (LPS) have been detected in the plasma of IBD patients and an abnormal microflora and/or an increased permeability of the intestinal mucosa have been invoked as cofactors responsible for endotoxemia. At the same time, the evidence that phagocytosis and killing exerted by polymorphonuclear cells and monocytes and the T-cell dependent antibacterial activity are decreased in IBD patients may also explain the origin of LPS in these diseases. In IBD, pro-inflammatory cytokines and chemokines have been detected in elevated amounts in mucosal tissue and/or in peripheral blood, thus suggesting a monocyte/macrophage stimulation by enteric bacteria and/or their constituents (e.g. LPS). On these grounds, in experimental models and in human IBD, anti-cytokine monoclonal antibodies and interleukin receptor antagonists are under investigation for their capacity to neutralize the noxious effects of immune mediators. Finally, the administration of lactobacilli is beneficial in human IBD and, in murine colitis, this treatment leads to a normalization of intestinal flora, reducing the number of colonic mucosal adherent and translocated bacteria.

Our data demonstrate an impairment of natural immunity exerted by peripheral blood phagocytes and lymphocytes in patients with UC and CD. ANCA and anti-LF antibodies were present mainly in UC patients but their presence did not affect PMN-mediated phagocytosis and killing. Finally, plasma endotoxins may contribute to the chronic inflammatory status, likely by inducing release of proinflammatory mediators.