Platelet-activating factor (PAF) is known to be an important mediator of anaphylaxis. However, there is a lack of information in the literature about the role of PAF in food allergy. The aim of this work was to elucidate the participation of PAF during food allergy development and the consequent adipose tissue inflammation along with its alterations. Our data demonstrated that, both before oral challenge and after 7 days receiving ovalbumin (OVA) diet, OVA-sensitized mice lacking the PAF receptor (PAFR) showed a decreased level of anti-OVA IgE associated with attenuated allergic markers in comparison to wild type (WT) mice. Moreover, there was less body weight and adipose tissue loss in PAFR-deficient mice. However, some features of inflamed adipose tissue presented by sensitized PAFR-deficient and WT mice after oral challenge were similar, such as a higher rate of rolling leukocytes in this tissue and lower circulating levels of adipokines (resistin and adiponectin) in comparison to nonsensitized mice. Therefore, PAF signaling through PAFR is important for the allergic response to OVA but not for the adipokine alterations caused by this inflammatory process. Our work clarifies some effects of PAF during food allergy along with its role on the metabolic consequences of this inflammatory process.
Food allergy is an immune-mediated response to food that affects approximately 5% of young children and 3% to 4% of adults in westernized countries [
PAF is a glycerophospholipid synthesized and secreted by mast cells, monocytes, and tissue macrophages. The binding of PAF to its receptor (PAFR) on platelets, monocytes, macrophages, and neutrophils results in many of the manifestations of anaphylaxis [
In order to understand the role of PAF in the development and consequences of food allergy, we used a murine model of food allergy to ovalbumin (OVA) developed by our group, in which sensitized mice are orally challenged for 7 days with an OVA-containing diet (OVA diet). In this model, many signs similar to those presented by patients with food allergy are developed, such as antigenic aversion, increased anti-OVA IgE production, and intestinal eosinophil infiltration, as well as a marked weight and adipose tissue loss [
Male BALB/c, six- to eight-week-old mice were obtained from the animal facility of Federal University of Minas Gerais (UFMG). PAF receptor deficient mice (PAFR−/−), with the same age as the wild type (WT) animals, were kindly provided by Professor Dr. Mauro Martins Teixeira, from the Immunology and Biochemistry department of UFMG. Mice received standard mouse chow diet (Purina) until the beginning of antigen challenge with OVA diet. The experiments were made according to the Ethical Principles in Animal Experimentation of our institution and the experimental protocol was approved by the Ethics Committee in Animal Experimentation of the University (protocol 85/2011-CETEA/UFMG).
OVA-sensitized mice (OVA+ group) received subcutaneous (sc) injection of 0.2 mL saline with 1 mg Al(OH)3, as adjuvant, and 10
Body weight was determined daily during the OVA challenge. During the oral challenge period, OVA diet consumption was assessed daily by weighting the remaining chow and comparing its weight with the previous day. Data were reported as amount of diet consumption per group during the challenge week and divided by the number of animals per cage. Before the challenge starts and after 7 days of continuous OVA oral challenge, mice were anesthetized by intraperitoneal (ip) injection of 10 mg/kg xylazine and 100 mg/kg ketamine hydrochloride and blood was obtained from brachial plexus. Later, under deep anesthesia, mice were euthanized in order to collect the small intestine and perigonadal adipose tissue.
In order to assess the DAI score [
Anti-OVA IgE ELISA was performed using plates coated with rat anti-mouse IgE (Southern Biotechnology Associates, Birmingham, USA), serum, and biotinylated OVA, as previously described [
The proximal portion of the jejunum was fixed in 4% phosphate-buffered formaldehyde for 24 h. After fixation, the tissue was dehydrated in ethanol, cleared in xylene, and embedded in paraffin. Sections were cut (5
After euthanasia, approximately 2 cm of the proximal jejunum was collected and frozen for posterior analyses. To measure eosinophilic peroxidase, 50 mg of samples were homogenized in 950
Samples from perigonadal adipose tissue were fixed in 4% phosphate-buffered formaldehyde for 24 h, dehydrated in absolute ethanol, cleared in xylene, and then embedded in paraffin. Histological sections (5
Mice were anesthetized with xylazine (10 mg/kg) and ketamine hydrochloride (100 mg/kg). The right jugular vein was cannulated and rhodamine 6G (Sigma, St. Louis, MO, USA) was injected intravenously (iv; 1.5 mg/kg) to label the leukocytes and endothelial cells. Then the perigonadal adipose tissue was collected by an abdominal incision. Rhodamine-epi-illumination was achieved with a 150 W variable HBO mercury lamp in conjunction with a Zeiss filter set 15 (546/12 nm band-pass filter, 580 nm Fourier transforms, 590 nm late potentials; Zeiss, Wetzlar, Germany). The microscopic images were captured using a Nikon eclipse 50i (Nikon Instruments Inc., Japan) microscope (×20 objective) with a video camera (5100 HS; Panasonic, Secaucus, NJ) and recorded digitally using both filter blocks consecutively. Data analysis was performed offline. Rolling leukocytes were considered as cells passing through a given point in the venule per minute. Adherent leukocytes were the cells that remained stationary for at least 30 s or longer within a 100
Levels of serum adiponectin, resistin, and leptin were determined with DuoSet ELISA kits and they were performed according to the instructions provided by the manufacturer (R&D System, Inc., Minneapolis, USA).
Before oral challenge, mice were euthanized and spleens were collected. Spleen cells were harvested in a RPMI 1640 medium (Sigma, USA) containing 10% fetal bovine serum and supplemented with 2 mM of L-glutamine, 10,000 of U penicillin, and 10 mg of streptomycin (Gibco, USA). Cells at 1 × 106/well were then incubated in 24-well plates and treated with OVA (five times crystallized hen’s egg albumin; Sigma, USA) (1000
Results were expressed as mean ± standard error of the mean (SEM) for six mice in each group (each experiment was repeated twice) and analyzed using GraphPad Prism version 4.0 (GraphPad Software, San Diego, CA). The variance homogeneity of the data was tested with Bartlett’s test. All data were analyzed for normality of distribution using the Kolmogorov-Smirnov test and were found to be normal. Parametric data were evaluated using one-way analysis of variance (ANOVA), followed by Newman-Keuls posttest. Differences were considered statistically significant at
As predicted by our experimental model of food allergy, sensitized mice after receiving OVA diet during one week (OVA+) showed higher serum anti-OVA IgE levels in comparison to nonsensitized animals that received the same diet (OVA−). Interestingly, levels of specific IgE in mice lacking PAFR from OVA+ group were significantly decreased in comparison to OVA+ WT mice (Figure
Markers of food allergy after 7 days of ovalbumin consumption by nonsensitized (OVA−) and sensitized (OVA+) mice. Serum anti-OVA IgE (a), DAI score (b), EPO activity in jejunum (c), and percentage of PAS (Periodic Acid-Schiff) by field also in jejunum histology (d). Representative photomicrographs of PAS stained (100x) intestine showing mucus production by goblet cells in evidence. Bars indicate 50
Besides positive specific IgE titers, sensitized mice after receiving OVA diet were characterized by eosinophil infiltration and increased mucus production in the small intestine. When these features were analyzed in sensitized PAFR−/− mice both parameters were significantly decreased in comparison to the OVA+ WT mice. OVA+ PAFR-deficient mice presented lower EPO activity (Figure
One feature of our experimental model of food allergy is that sensitized mice followed by an oral challenge present a marked body weight and adipose tissue loss [
Ingestion of ovalbumin diet by sensitized (OVA+) and nonsensitized (OVA−) mice, mean of all consumption of the week challenge per animal (a). Variation of the body weight between the first and last day of oral challenge (b). Perigonadal adipose tissue weight (c) and adipocyte area from this tissue (d) after 7 days of OVA diet challenge. Representative photomicrographs of H&E stained (100x) perigonadal adipose tissue used to determine adipocyte areas (
In order to assess the inflammation in the adipose tissue, intravital microscopy was performed. It was observed that sensitized (OVA+) mice, after 7 days of oral challenge, showed an increase in rolling and adherent leukocytes in the microvasculature of perigonadal adipose tissue in comparison to nonsensitized (OVA−) mice (Figures
Visualization of leukocyte-endothelium interaction in the microvasculature of perigonadal adipose tissue in sensitized (OVA+) and nonsensitized (OVA−) mice after 7 days of oral OVA challenge. Intravital microscopy was used to assess the rolling (a) and adherent leukocytes (b). Levels of serum leptin (c), resistin (d), and adiponectin (e) were determined. Data are expressed as mean ± SEM.
In order to analyze the systemic effects of adipose tissue inflammation, levels of adipokines in the serum were determined. Adipokines are cytokines produced mainly by this tissue in response to different stimuli such as inflammation. Sensitized (OVA+) mice after 7 days of OVA challenge showed a significant decrease in the serum levels of serum leptin, resistin, and adiponectin (Figures
In order to elucidate whether PAFR affects OVA-sensitization or operates during the effector phase of food allergy, the serum specific anti-OVA IgE was determined in sensitized animals before the oral challenge. As predicted by our experimental model of food allergy, sensitized mice (OVA+) showed higher serum anti-OVA IgE levels than nonsensitized animals (OVA−). Interestingly, levels of specific IgE of mice lacking OVA+ PAFR were significantly decreased in comparison to OVA+ WT mice (Figure
Serum anti-OVA IgE in nonsensitized (OVA−) and sensitized (OVA+) mice before oral ovalbumin challenge (a). Cytokine levels in supernatant of splenocyte culture from OVA-sensitized mice before oral challenge, IL-4 (b), IL-5 (c), and IL-10 (d). Data are expressed as mean ± SEM.
Our data demonstrated that, after 7 days of oral challenge with OVA diet, OVA-sensitized mice lacking the PAF receptor showed decreased levels of serum anti-OVA IgE associated with attenuated allergic markers such as eosinophil infiltration and mucus production in the small intestine in comparison to OVA+ WT mice. Moreover, the body weight and adipose tissue loss followed by adipocyte area reduction was less expressive in PAFR-deficient mice. However, some features of inflamed adipose tissue presented by sensitized PAFR-deficient mice and WT mice after oral challenge were similar. OVA+ PAFR-deficient mice, as WT animals, showed a higher rate of rolling leukocytes in the microvasculature of adipose tissue, indicating cell activation. Also, circulating levels of adipokines such as resistin and adiponectin were decreased in comparison to nonsensitized mice and reached levels comparable to OVA+ WT mice.
IgE is an immunoglobulin responsible for type I hypersensitivity reactions and in our experimental model of food allergy we observe an increase in the levels of specific anti-OVA IgE in sensitized mice that is exacerbated after the oral challenge. In an interesting way, OVA+ PAFR-deficient mice showed significantly lower levels of circulating specific IgE when compared to OVA+ WT mice. This same profile of immunoglobulin was also observed after oral challenge by sensitized PAFR-deficient mice. Although it was already shown that the treatment with PAF antagonists is not able to change levels of IgE antibodies after OVA-sensitization [
Analyzing the clinical score of the disease (DAI score), OVA+ PAFR-deficient mice showed a lower score than OVA+ WT mice. This lower score was due to a not so intense weight loss from the PAFR-deficient mice, even though both groups showed the same level of antigen aversion, and to the diminished diarrheal response presented by these mice. Attenuated diarrheal response in the PAFR−/− OVA+ group was expected because mast cells are required for experimental oral allergen-induced diarrhea due to the release and synergistic signaling induced by serotonin and PAF, but not histamine [
PAF is one of the most potent and versatile immune mediators found in mammals. Among its recognized functions are its ability to aggregate platelets and dilate blood vessels. Furthemore, among the nonregular functions exerted by PAF, we can mention its eosinophil chemotactic and activation capability [
Recent studies on PAF have shown that PAFR−/− mice fed with either a high-fat diet or a high-refined carbohydrate-containing diet presented higher adiposity than wild type mice. These animals also presented a different inflammatory profile on adipose tissue along with its consequent metabolic alterations, demonstrating that PAF has an important role in these processes [
Furthermore, allergic PAFR−/− mice levels of adipokines such as adiponectin and resistin, but not leptin, were reduced at the same levels of allergic WT mice. Leptin is an adipokine responsible for controlling the energetic homeostasis of the organism [
In summary, our data showed that the absence of PAF signaling through PAFR results in a decrease of allergic response to OVA. However, PAF was not important to the altered profile of adipokines caused by oral ingestion of OVA after previous sensitization. Our work contributes towards clarifying some effects of PAF during food allergy development along with the role of this molecule in the metabolic consequences of this inflammatory process. We suggest that anti-PAF therapies could be an interesting way to treat food allergy but its consequences should be followed since some associated metabolic alterations remain unchanged with this approach.
The authors declare that there are no competing interests regarding the publication of this paper.
Nathália Vieira Batista and Roberta Cristelli Fonseca contributed equally to this work.
The authors are grateful to the Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq/Brazil), Fundação de Amparo a Pesquisa do Estado de Minas Gerais (FAPEMIG/Brazil), and Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES/Brazil) for financial support. Some of the authors are also recipients of CNPq research fellowships (Nathália Vieira Batista, Rafaela Vaz Sousa Pereira, Ana Maria Caetano de Faria, Vanessa Pinho, and Denise Carmona Cara) and CAPES research fellowships (Denise Perez and Juliana de Lima Alves).