Identification and signature profiles for pro-resolving and inflammatory lipid mediators in human tissue

Abstract

Resolution of acute inflammation is an active process locally controlled by a novel genus of specialized pro-resolving mediators (SPM) that orchestrate key resolution responses. Hence, it is of general interest to identify individual bioactive mediators and profile their biosynthetic pathways with related isomers as well as their relation(s) to classic eicosanoids in mammalian tissues. Lipid mediator (LM)-SPM levels and signature profiles of their biosynthetic pathways were investigated using liquid chromatography-tandem mass spectrometry (LC-MS-MS)-based LM metabololipidomics. LM and SPM were identified using ≥6 diagnostic ions and chromatographic behavior matching with both authentic and synthetic materials. This approach was validated using the composite reference plasma (SRM1950) of 100 healthy individuals. Using targeted LM metabololipidomics, we profiled LM and SPM pathways in human peripheral blood (plasma and serum) and lymphoid organs. In these, we identified endogenous SPM metabolomes, namely, the potent lipoxins (LX), resolvins (Rv), protectins (PD), and maresins (MaR). These included RvD1, RvD2, RvD3, MaR1, and NPD1/PD1, which were identified in amounts within their bioactive ranges. In plasma and serum, principal component analysis (PCA) identified signature profiles of eicosanoids and SPM clusters. Plasma-SPM increased with omega-3 and acetylsalicylic acid intake that correlated with increased phagocytosis of Escherichia coli in whole blood. These findings demonstrate an approach for identification of SPM pathways (e.g., resolvins, protectins, and maresins) in human blood and lymphoid tissues that were in amounts commensurate with their pro-resolving, organ protective, and tissue regeneration functions. LM metabololipidomics coupled with calibration tissues and physiological changes documented herein provide a tool for functional phenotypic profiling.

Keywords: eicosanoids; inflammation; resolution; resolvins.

Copyright © 2014 the American Physiological Society.

Figures

Fig. 1.

Operationalization of lipid mediator (LM) metabololipidomics. Synthetic and authentic deuterated internal standards were added to samples before automated extraction by solid-phase extraction. LM were investigated by liquid chromatography-tandem mass spectrometry (LC-MS-MS) using targeted multiple reaction monitoring (MRM) coupled with information-dependent acquisition-enhanced product ion (EPI) data acquisition or data-dependent scan/enhance MS scan (EMS) or MSn. Identification of bioactive and biosynthetic pathway products was performed by matching retention time (RT) and at least 6 diagnostic ions from the spectrum with our LM database (see materials and methods for details). The LM and specialized pro-resolving mediator (SPM) database was constructed with synthetic and authentic standards, and quantitation was carried out using linear regression compared with standard curves from the LM database. Interrelationship(s) for identified mediators within each condition and between different conditions was further investigated by computational analysis that included principal component and cluster analyses.

Fig. 2.

Endogenous LM-SPM in human peripheral blood. MRM of signature ion pairs was obtained using the precursor ion (Q1) and a characteristic product ion (Q3) for each LM. Bioactive LM, isomers, and pathway markers were identified in human serum from docosahexaenoic acid (DHA) bioactive metabolome, including D-series resolvins, maresins, and protectins (A); eicosapentaenoic acid (EPA) bioactive metabolome, including EPA-derived lipoxins and E-series resolvins (B); and arachidonic acid (AA) bioactive metabolome, including AA-derived lipoxins, leukotrienes, and prostanoids (C). Data are representative of 3 different pooled sera, each from >100 individual healthy donors. See Glossary for definitions; m/z, mass-to-charge ratio.

Fig. 3.

LM-SPM MS-MS spectra from DHA bioactive metabolome identified in human serum. Characteristic MS-MS fragmentation patterns and diagnostic ions were employed for identification. MS-MS spectra are from the D-series resolvins, maresins, and protectins. M-H, molecular ion. Data are representative MS-MS spectra for SPM and products from Fig. 2.

Fig. 4.

LM-SPM MS-MS spectra from EPA bioactive metabolome identified in human serum. Characteristic MS-MS fragmentation patterns and diagnostic ions were employed for identification. MS-MS spectra are from the EPA-derived 5S,15S-diHEPE and E-series resolvins (RvE1, RvE2, and RvE3). Data are representative MS-MS spectra for SPM and products from Fig. 2.

Fig. 5.

LM-SPM MS-MS spectra from AA bioactive metabolome identified in human serum. Characteristic MS-MS fragmentation patterns and diagnostic ions were employed for identification. MS-MS spectra are from the lipoxins, prostanoids (PGD2, PGE2, and PGF2α), the thromboxane hydrolysis product TxB2, the leukotriene B4 metabolite 20-OH-LTB4, and 5S,12S-diHETE. Data are representative MS-MS spectra for products from Fig. 2.

Fig. 6.

Human LM-SPM signature profiles: principal component analysis. For human plasma from the National Institute of Standards and Technology (NIST) standard reference material (SRM 1950, a composite plasma from 100 healthy individuals), individual human plasma, commercial human serum (each a composite of ∼100 healthy individuals, ∼300 subjects total), and individual human serum, LM-SPM profiles were obtained using LM metabololipidomics. A: 3-dimensional score plot of human plasma with healthy donors from NIST SRM 1950 (100 subjects; blue circles, n = 22) compared with human serum (318 subjects, composites plus fresh serum; green circles, n = 20). B: 3-dimensional loading plot. Gray ellipse in the score plot (A) denotes 95% confidence regions (see materials and methods).

Fig. 7.

Human lymphoid tissues: LM-SPM profiles. LM obtained from human axillary lymph nodes and spleens were identified using LM metabololipidomics (as in Fig. 1). A and B: MRM chromatograms from human lymph nodes (A) and human spleens (B), obtained using the diagnostic ion pairs (Q1) and a characteristic product ion (Q3) for each SPM, e.g., lipoxins, E- and D-series resolvins, maresins, and protectins. Cumulative levels for individual LM families are depicted as a function of color intensity. Color scales (white to blue) are set from 0 to 200 pg/100 mg tissue. Each chromatogram is representative of n = 6 lymph nodes or n = 3 spleens. Patient demographics are reported in Table 2.

Fig. 8.

Human plasma LM-SPM signature profiles and increased phagocytosis. Healthy volunteers were given capsules containing essential fatty acid (EFA). After 2 h, 81 mg of acetylsalicylic acid (ASA) were taken, and heparinized blood was collected from each subject at 4 h. Plasma LM profiles were subject to LM metabololipidomics (0 h) and EFA and ASA intake (4 h; see materials and methods for details). A: 2-dimensional score plot of human plasma at 0 and 4 h. B: 2-dimensional loading plot. Gray ellipse in the score plots denotes 95% confidence regions; n = 10 healthy individual donors. C: phagocytosis in whole blood. Fresh heparinized whole blood samples were split and incubated (20 μl, 37°C, 60 min) with BacLight green-labeled Escherichia coli (∼2 × 107 colony-forming units). Phagocytosis was monitored by flow cytometry (see materials and methods). Results are expressed as %increase in mean fluorescence intensity (MFI); n = 10 healthy subjects. ***P = 0.001, 0 h vs. 4 h. Inset: representative histograms of phagocytosis in whole blood. D: positive correlation of specific SPM and phagocytosis in human peripheral blood using partial least squares-discriminant analysis from results in Table 3 (n = 10 healthy subjects). Healthy donor demographics are reported in Table 4.