Prevalence of inhibitory LPS within gut microflora could underlie hygiene hypothesis

This week prestigious journal Cell published seminal microbiome study that could explain hygiene hypothesis that suggests that "microbial cleanliness" in developed countries predisposes individuals to immunopathologies such as allergy, multiple sclerosis and other autoimmune diseases. 

This study found that gut microflora in babies from Finland and Estonia, but not from Russian Karelia, are enriched in gut bacteria with inhibitory LPS that failed to properly "educate" newborns innate immune system thus predisposing them to higher incidence of immunopathologies.

The first finding was that microbiota in babies from Finland, but not from Russian Karelia, were enriched in Bacteroides. Now this fact by itself isn't too surprising. 

The second finding was that lipid A component of LPS from these Bacteroides [unlike lipid A component of LPS from E. coli] species were non-stimulatory in TLR4 assay. In fact, molecular analysis showed that Bacteroides harbored tetra- and penta-acylated lipid A structures, as opposed to the hexa-acylated lipid A seen in E. coli. Such structural modification converted Bacteroides lipid A into totally non-stimulatory ligand.

In fact, Bacteroides derived lipid A was inhibitory when combined with stimulatory, E.coli derived lipid A [LPS tolerance assay, where primary exposure to E. coli LPS makes responding cell refractory to secondary exposure].

Finally, third finding was that presence of non-stimulatory lipid A had immunological consequences since it failed to reduce incidence of diabetes in NOD mice [whereas lipid A from E. coli could, as expected].  

In summary, this important study has provided one of the first definite molecular evidence underlying hygiene hypothesis and suggested the path for its prevention.

David Usharauli

     

Sialic acid decorated multi-valent nanoparticles prevent sepsis mortality in mice

Sepsis is an uncontrolled tissue-immune [hyper]response that leads to the organ failure and frequently patients' death. At present, there is no specific treatment for sepsis, just supporting care. Obviously, sepsis represents important unmet medical condition.

New study in Science Translational Medicine provided evidence that sialic acid coated nanoparticles could prevent sepsis-induced death in mice model of acute sepsis.  

This study is based on prior evidence that binding of sialic acid (type of sugar) to its receptors called Siglecs on macrophages or neutrophils could inhibit inflammatory pathways. Here, the authors showed that unlike bi-valent anti-Siglec antibodies, administration of multi-valent sialic acid coated nanoparticles (in a form of α2,8 sacharide linkage), at both T= 0h or T= +2h, prevented endotoxin induced animal death (i.p. injection).

Similar protection was observed in anesthetized mice undergoing cecal ligation and puncture (CLP) procedure (here too, α2,8-NANO was administered i.p.).

Additionally, intra-lung administration of α2,8-NANO showed significant (but not as impressive) protection against lung inflammation (a secondary complication following cecal ligation and puncture procedure).

The authors showed that protection provided by α2,8-NANO was IL-10 dependent in endotoxin injection model (though, for some reason, the authors had used 1mg α2,8-NANO here, not 2mg α2,8-NANO, as in other figures). Also, application of exogenous IL-10 in IL-10KO mice would have provided more clear role of IL-10 here. 

Finally, the authors tried to show that α2,8-NANO application could be useful for human condition as well. They showed that α2,8-NANO could reduce inflammatory cytokine secretion from primary monocytes stimulated with endotoxin in vitro (but for some reason the authors decided to show "processed", not raw data, implying that data were either not consistent or showed too much variability).

In summary, this study suggests that sialic acid decorated multi-valent nanoparticles may play beneficial role in sepsis management (I want to note here that I am against conducting such distressing experiments on lab animals as are i.p. injection of endotoxin or CLP. We must find better alternative models).

As mention earlier, this study comes with 2 major drawbacks (that make this study not easily "translational"):

(1) while animal studies, shown here, were quite impressive, primary human cells showed less sensitivity towards α2,8-NANO application.
(2) even in animal models, α2,8-NANO was applied through i.p. injection. This route of injection is completely non-applicable for human use and I am puzzled that labs still continuing using it. Why not just use injection via i.v. route?

David Usharauli

Dirt exposure could prevent asthma development in mouse

Allergic or exaggerated response to mostly innocuous antigens is a Terra Incognita in our understanding of how immune system really works. First question that comes to mind is why is there even such molecule as IgE that causes nothing but allergic reaction when combined with antigen? A hypothesis proposed by immunologist, Ruslan Medzhitov, suggests that the evolutionary role of IgE was to defend the body against toxins. But IgG could do that too? 

Another hypothesis commonly referred as hygiene hypothesis explains that allergic responses are, in fact, out-of-control type II immune reactions (highly skewed Th2-IgE-IL-13 axis) due to absence of "normal" education of immune system by reduced exposure to infections or infectious materials (bacteria, dirt, dust, fungi, dander, etc.) during early life (that supposed to induce Th1-IFNγ axis balancing Th2 response).

The new paper in journal Science is another example for such immune "balancing" act in support of hygiene hypothesis. Here, the authors reported that pre-exposure to low dose of LPS (endotoxin, a part of gram-negative bacteria) could raise the threshold for lung airway epithelial sensitivity to house dust mite (HDM) extract via modulation of NF-kb activity via up-regulation of A20 (Ubiquitin-editing protein).  

First, the authors showed that compared to control, mice pre-exposure to LPS before HDM application were protected from developing features of airway hypersensitivity (the most important figure is 1D).

Since LPS pre-exposure induced A20 up-regulation in lung epithelial cells, the authors tested mice with lung epithelial specific deletion of A20 (tnfaip3 KO). Indeed, beneficial effect of LPS exposure was abolished in these A20 deficient mice (tnfaip3 ^EC-KO).

Finally, the authors speculated that asthmatic individuals may be deficient for A20 in their lung epithelial cells. However, for some reason, contrary to the authors hypothesis, lung cells from severe asthmatics expressed more A20 molecules compared to lung epithelial cells from mild asthmatics (see figure below). The authors did not discuss this contradiction.

There are several weaknesses in this article. For one, most data are presented as bars rather than scatter plot with individual data points. 2nd, for some unknown reason, the authors decided not to show lung hypersensitivity assay with LPS pre-exposure from mutant mice, as it was done with wild-type mice in Fig. 1D. This assay is clinically the most relevant. This is a Science paper, to remind you.

So, in summary, while I do think that exposure to dirt during early life is relevant for human asthma development, based on results from this paper I am not convinced that expression of A20 in lung epithelial cells play a role here.  

David Usharauli

blind love: neuro-immune chemistry

Immune and nervous systems share two unique characteristics: (a) both systems undergo adaptive “education” to discriminate between self and nonself signals and (b) both systems have “memory” to it.

So far few research papers had been published that addressed the question how these two complex systems interact with each other. For example, one paper claimed that T cells can promote neuronal re-generation and hence contribute to the learning process (1). These study was based on evidence that T cell transfer into immunodeficient mice improved neuronal generation. In my opinion, this interpretation fails to take into account the fact that T cell transfer into immunodeficient mice affect not just brain function but for instance, gut permeability too. Why is it important? It is well known that endotoxin (LPS) level in the blood affects brain function. LPS level in the blood, in turn, is influenced by gut permeability, that in turn, is influenced by immune system status. T cell transfer into immunodeficient mice would allow differentiation of gut-homing T cells that may have reduced gut permeability thus indirectly affecting brain function.

If you are interested in neuro-immune research, then I will recommend to read the following two papers recently published in Science (2, 3).

1^st paper from Kevin Tracey's Lab provided the direct evidence that the presence of specialized, acetylcholine-secreting memory T cells were necessary and sufficient to relay signals from nervous system to the immune (2).

2^nd paper from Paul Kubes Lab showed that CD1d-deficient mice (that lack all NKT cells) were more susceptible to stroke-associated immunosuppression compared to wild-type mice (3). This immunosuppression could be prevented by stimulating NKT cells or blocking noradrenalin signaling in these cells because this protective effect of noradrenalin blockade was abolished in CD1d-deficient mice. It is of note that both NKT presence and simultaneous blockade of noradrenalin signaling in these cells was necessary for full protection. This is a kind of paradox. However, because there are two types of NKT cells in mice, there may be a simple explanation. So it will be interesting to compare CD1d-deficient mice to Jalpha18-deficient mice that lack only one type of NKT cells. In my opinion, NKT cells transfer into CD1d-deficient mice would have provided more direct evidence of protective role of this innate NKT cells.

David Usharauli

(1) http://www.nature.com/neuro/journal/v9/n2/full/nn1629.html

(2) http://www.sciencemag.org/content/early/2011/09/14/science.1209985.abstract

(3) http://www.sciencemag.org/content/early/2011/09/14/science.1210301.abstract