Saturday, August 30, 2014

Ancient defense mechanism is not color-blind

Here is a very simple paper from Nature. Sometimes simple equals great. Not this time, though. No wonder it took 1.5 year to have it published.

This is a study about Aryl Hydrocarbon receptor (AhR). As I have discussed in earlier posts, AhR senses environmental pollutants and toxins, like p-dioxin. More importantly, however, AhR senses natural, endogenously-generated molecules like amino acid tryptophan degradation products. Upon ligand engagement, AhR activates detoxifying enzymes.

Here (1), the authors tested the hypothesis the AhR can detect pathogen-associated molecules. Using in silico modeling they identify several pigmented virulence factors derived from P. aeruginosa (phenazine) and M. tuberculosis (phthiocol) that could bind AhR.

To verify this finding in vitro the authors used luciferase reporter assay with human macrophage cell line (THP-1) transfected with AhR responsive elements. Indeed, physiological concentration of phenazine or phthiocol could activate AhR

To test this observation in vivo, the authors infected AhR-KO mice with P. aeruginosa. Compared to wild-type control, AhR-KO mice succumbed to infection more rapidly. This was associated with more tissue damage and less neutrophil infiltration. 

Bone marrow chimera experiment indicated that both haematopoietic and non-haematopoietic cell expression of AhR were necessary for resistance to P. aeruginosa infection. 

In addition, the authors showed that resistance to M. tuberculosis (both at low or high dose) was also influenced by absence of AhR signaling. 

In summary, the authors postulate that AhR can sense infection-derived molecules and contributes to host defense.

While this model is very simple, there are few weaknesses in this paper. Since AhR function is quite broad, to show its relevance in infection-derived molecule detection, the authors should have used AhR-ligand deficient P. aeruginosa or M. tuberculosis, for example, mutant strain PA14 phz ½. However, the authors did not used it. In absence of such experiment, however, the authors conclusion regarding pattern-recognition receptor function of AhR is questionable.

Sunday, August 10, 2014

Interleukin-22's messy Nature

One way to determine if a research article published in prestigious journal comes with many “holes” is to look at its submission and acceptance time. If it is more than 6 months, it is highly likely an article is of low quality.

Here is one example of such article (1). It was published, after more than 1 year of review process, in most prestigious journal Nature and came from famous biotech company Genentech

You may wonder why Nature has published it? No idea.

This study focused on IL-22 and its role in metabolic syndrome, like obesity or defense against intestinal inflammation.

First, the authors showed that obese mice (ob/ob, db/db, or HDF) express less IL-22 compared to wild-type or lean mice.

Second, obese mice due to leptin signaling defect (ob/ob and db/db) were shown to be more susceptible for intestinal inflammation caused by C. rodentium that could be reversed by IL-22-Fc or IL-23 injections. The authors speculate that IL-23 was upstream of IL-22 signaling in this model, without providing any evidence.

Third, the authors switch to metabolic syndrome and observed that IL-22-Fc injection improves insulin sensitivity and blood glucose level.

Fourth, the authors suggest that IL-22-Fc injection improved symptoms of metabolic syndrome (weight loss, less food intake, increased gut barrier integrity, increased secretion of anorexic hormone PYY).

Finally, the authors showed that IL-22-Fc improved fat metabolism through its effect on liver cells and adipocytes.

However, experiments were complicated by the fact that IL-22KO and IL-22R1KO mice had different phenotype due to the fact that IL-22R1 can pair with other receptors (IL-10R2 and IL-20R2) and signal with IL-20 and IL-24.

In summary, the authors suggest that reduction of IL-23 driven IL-22 expression leads to alteration in proper fat metabolism in the liver cells and adipocytes, weakening of gut barrier integrity and chronic LPS presence in the blood stream, all resulting in obesity and metabolic syndrome development (insulin resistance).

Now, some thoughts to consider. First, are IL-22KO mice obese? Second, does injection of IL-23 rescue ob/ob mice on IL-22KO background?


Saturday, August 9, 2014

Eosinophils preference for peanut can lead to allergy

Peanut allergy is a very serious medical condition. Usually, gut immune system tolerates antigens derived from orally consumed food. In rare situations, however, immune system mistakes food antigens for noxious stimuli and mounts exaggerating IgE response. IgE response is basically a highly skewed Th2 response. No one knows how or why it happens.

I am going to review one paper published in Journal of Experimental Medicine (JEM) that studied animal model of peanut allergy (1). I would like to point out here that this mouse model of peanut allergy is an experimental model and may not fully or even partially recapitulate the mechanism of initiation of peanut allergy in humans.

First, analysis of gut lymphoid tissue showed that small intestine contained higher percentage of eosinophils (~ 20%). Interestingly, there was inverse relationship between number of eosinophils and level of gut microbiome.

Second, the authors showed that 4 consecutive, weekly, intra-gastric application of peanut (P) + cholera toxin (CT) induced peanut allergy. This allergic response was dependent on the presence of eosinophils since it was absent in GATA1 mutant mice with low GATA1 promoter activity lacking eosinophils. Adoptive transfer of wild-type eosinophils rescued peanut allergy development in GATA1 mutant mice.

Third, the authors found that IL-4 secretion by eosinophils was not necessary to induce peanut allergy in this model.

Fourth, it turned out that gut CD11c+ /CD103+ dendritic cells were also required to induce peanut allergy in this model since it was abolished in CD11c-DTR bone marrow chimera mice upon DT (diphtheria toxin) application.

Next, the authors found peanut + cholera toxin combination induced eosinophil degranulation releasing eosinophil peroxidase (EPO), similar to positive control reagent, L-PAF.

Incubation of mouse BMDCs or human in vitro derived DCs with native EPO, but not heat-inactivated EPO, induced their activation and IL-6 secretion without compromising their viability.

To confirm that EPO played a role in mouse model of peanut allergy, the authors tested EPO-deficient mice. As shown below, EPO-deficient mice failed to generate peanut specific IgE response similar to eosinophil-deficient GATA1 mutant mice.

In summary, the authors proposed that mechanism of peanut sensitization as following: food peanut antigen activates gut eosinophils to release EPO to drive local DCs migration into local LN to prime Th2 response.

Some consideration: (1) do eosinophils sense peanut? The authors do not show if peanut alone can activate eosinophils. (2) while data from EPO-deficient mice is impressive, the authors did not explain whether EPO-deficient mice selectively lacks EPO in eosinophils as the name would suggest (BM chimera would have been a good experiment).


Friday, August 1, 2014

Gut flora helps macrophages to flex the muscles

Though I am a little bit tired reviewing papers about gut microbiome, this new paper in Cell is so cute I could not ignore it.

GI tract's proper motility is obviously very important for a good health. It is also quite obvious that food and gut neuronal network regulate GI tract motility and of course, gut muscles. Now this gut muscles are specialized type of muscles called smooth muscles.

This short and very simple paper published in Cell (1) has made very unusual observation. One of the senior authors on this paper is Miriam Merad. She is well known for her work on macrophages.

First, the authors found that gut outer muscular layer harbored Cx3CR1+ macrophages. The development of these muscle associated-macrophages was dependent on CSF-1 (M-CSF) receptor as shown by their absence in CSF receptor-1 or CSF-1 KO mice. 

In addition, intra-peritoneal injection of anti-CSF1 antibody preferentially depleted muscle-associated macrophages.

The parallel set of experiments showed that ex vivo gut motility (peristaltic movements) were modified (accelerated) by anti-CSF1 antibody mediated muscle-associated macrophages depletion.

This was a peculiar observation since this muscle macrophages are sitting deep in the outer muscles layer of the gut and do no directly interact with gut flora, for example.

So, how their depletion affected GI tract motility?

To find the answer, the authors run microarray analysis on purified muscle-associated macrophages and found that molecule, called BMP2, was highly expressed by these macrophages

On the other hand, macrophages depletion affected neurons located in close proximity to the muscle macrophages. These neurons were found to express receptor for BMP2 and to express CSF-1.

It is well described that germ-free mice have GI tract motility issues. Anyone who worked with mice and seen germ-free mice intestine can ascertain that their colon is of enormous size. To further understand the relationship between neurons and macrophages and gut flora, the authors conducted series of experiments with antibiotic-treated mice. Antibiotic treatment diminished BMP2 expression in muscle macrophages and reduced GI peristaltic movement that was reversible with the addition of LPS in drinking water. Interestingly in vitro gut neurons produce more of CSF-1 in response to LPS.

In summary, the authors proposed the following model of physiological gut motility: products of gut microbiota induce neuronal cells (or other cells) in the gut to secrete CSF-1. CSF-1 stimulates development and/or maintenance of muscle-associated macrophages, which in turn secrete BMP2 required for normal gut neuronal activity.

This study represents another example supporting growing evidence that macrophages play more fundamental role in physiology, including already described role in thermoregulation.