Urinary Lipid Production Profile of Patients With Food Allergy

Abstract

Upon immediate allergic inflammation, activated mast cells produce abundant bioactive lipid mediator prostaglandin (PG) D₂. PGD₂ is metabolised and excreted in urine as tetranor-PGDM. We have previously reported urinary tetranor-PGDM as a sensitive biomarker for food allergy (FA) reactions in children [1]. However, the production profile of other lipid mediators in the urine of FA patients remains unknown. Bioactive lipids are produced by enzyme-dependent or independent metabolism of polyunsaturated fatty acids (PUFAs) such as arachidonic acid (AA), linoleic acid (LA), eicosapentaenoic acid (EPA), docosahexaenoic acid (DHA) and dihomo-gamma-linolenic acid (DGLA). There are three types of oxidative enzymes for PUFAs: cyclooxygenase (COX), lipoxygenase (LOX) and cytochrome P450 (CYP). The produced lipid mediators regulate inflammation and are extracted mainly in urine, thus their production profile can reflect the inflammatory status of our bodies. In this study, we comprehensively analysed the production profile of lipid mediators in the urine of subjects who received oral food challenge (OFC).

We collected the same urine samples from subjects as previously reported [1]. Children suspected of having FA who underwent OFC to milk, egg, peanut or sesame were recruited between December 2014 and August 2015, and urine samples were collected before (pre) and 4 h after OFC (post). Of a total of 42 children, 31 were assessed as having positive results (OFC-positive; FA). This study received ethical approval from the Committee, University of Tokyo School of Medicine (2017,10586). All participants provided informed consent.

Initially, we confirmed that OFC and/or its positivity did not influence urinary pH, or specific gravity, while they did influence the lipid content (Figure 1A). There was no difference in urinary lipid content in children who ingested offending food, suggesting that lipid content may reflect the inflammatory reaction in the body. Next, we analysed the levels of 196 types of lipid metabolites in urine using LC–MS/MS and detected 19 lipid metabolites. These methods are available in the following repository (https://zenodo.org/records/14207617).

Figure 1B shows the AA metabolites levels in subjects' urine. As reported previously, the urinary levels of tetranor-PGDM were increased in OFC-positive-post urine, and its levels were higher than those of OFC-negative-post urine [1]. The levels of other PGD₂ metabolites, 13,14-dihydro-15-keto-tetranor-PGD₂ and tetranor-PGJM (precursor and non-enzymatic metabolite of tetranor-PGDM, respectively), were also higher in the OFC-positive-post urine. It is well known that mast cells express H-PGDS and produce PGD₂ [2]. During the progression of FA, mast cells infiltrate into the relatively large area of intestinal mucosa and release PGD₂. This phenomenon presumably results in the detection of amounts of PGD₂ metabolites in the urine of FA patients.

The urinary levels of metabolites of major inflammatory mediators PGE₂ and thromboxane (TX) A₂, specifically tetranor-PGEM and 11-dehydro-TXB₂, were increased in the OFC-positive-post urine compared with those of OFC-positive-pre urine. Furthermore, 11-dehydro-TXB₂ in OFC-positive-post urine was higher than in OFC-negative-post urine. 11-dehydro-TXB₂ has been reported to increase during symptom induction in patients with atopic asthma [3]. In contrast, the levels of 11-dehydro-2,3-dinor-TXB₂; β-oxide of 11-dehydro-TXB₂, in OFC-positive-post urine were decreased.

Direct oxidised products of PUFAs, known as isoprostanes, serve as indicators of inflammatory responses [4]. 8-iso-PGA₂, an isoprostane, was increased in both OFC-positive-post and OFC-negative-post urine (Figure 1B). 8-iso-PGA₂ activates transient receptor potential cation channel subfamily A member 1 (TRPA1), which triggers allergies [5]. In contrast, the increase in OFC-negative-post urine seems to be unrelated to FA. Another isoprostane, 8-iso-15(R)-PGF_2α and its isomer, 8-iso-PGF_2α, were decreased, while its metabolite, 2,3-dinor-8-iso-PGF_2α, increased in OFC-positive-post urine. 2,3-dinor-8-iso-PGF_2α is known to be increased by oxidative stress, suggesting the increasing oxidative stress in the body of OFC-positive patients.

Upon allergic inflammation, activated mast cells are reported to produce leukotriene (LT) E₄ in a 5-LOX-dependent manner. Consistently, OFC-induced allergic inflammation increased the urinary level of LTE₄ (Figure 1B). Of interest, OFC-induced allergic inflammation decreased the urinary levels of LA-15-LOX metabolites, specifically 9-KODE, 13-HODE and 13-KODE (Figure 1C). In early inflammation, 5-LOX expresses strongly and 15-LOX decreases [6]. The present results suggest that urinary lipids 4 h after OFC may reflect the early inflammatory response.

As shown in Figure 1C, the urinary level of EPA was decreased while the level of its CYP metabolite 5,6-DiHETE was increased in OFC-positive-post urine. We previously reported that 5,6-DiHETE is metabolised from EPA during the healing phase of mouse colitis and represents anti-inflammatory effects [7]. Its increase in urine may reflect the onset and recovery from FA colitis.

In addition, OFC-induced allergic inflammation increased the urinary levels of inflammatory-related lipid mediators, such as DGLA-derived isoprostane 8-iso PGA₁, oleic acid-derived oleoylethanolamide (OEA) and platelet-activating factor (PAF)-derived Lyso-PAF (Figure 1C). OEA is one of the N-acylethanolamines and is involved in eosinophilic inflammation in asthma [8]. PAF is a potent inflammatory lipid mediator metabolised to Lyso-PAF [9].

Here, we revealed the lipid mediator production profile in FA patients' urine and demonstrated that 19 lipid metabolites, including tetranor-PGDM, in the urine of FA patients were indicative of allergic inflammation. Although urinary levels of tetranor-PGDM reflect mast cell activity in intestine, other lipids need to be clarified their production rationales. A limited number of urine samples were used in this study. In addition, 10 years have passed since the samples were collected, although no problem in this respect as the relative amounts of lipids were measured. In the future, the absolute amount of lipids should be measured using fresh samples. The discovery of the behaviour of tetranor-PGDM and other lipids will lead to a better understanding of the pathogenesis of food allergy and the search for new diagnostic markers.

T.M., Y.O., M.N. and S.I. designed the experiments and supervised the project. S.M., T.H., T.I. and N.N. performed experiments and analysed the data. S.I., K.Y.H., T.F., M.N., T.S. and Y.O. provided patients' samples and clinical information. S.M., N.N. and T.M. wrote the manuscript. All authors provided feedback on the manuscript.

The authors declare no conflicts of interest.