Fig 2: Administration of RvD1 to WD-fed Ldlr-/- mice with established atherosclerosis restores lipid medicator balance.(a) Quantification of lesional RvD1 and LTB4 levels (left and middle panels) and RvD1:LTB4 ratio (right panel) in aortic lesions of 17-week WD-fed Ldlr-/- mice that were administered vehicle control (Veh) or 100 ng RvD1/mouse (3 × /week i.p.) during weeks 12–17; the ratio data in 3a right panel are also shown for mice fed the WD for 8 weeks (Early lesions). For a, data are shown as mean±s.e.m., one-way ANOVA with Fisher's least significant difference post-hoc analysis, *P<0.05, n=8 for early lesions, n=11 for advanced lesions/Veh group, and n=10 for advanced lesions/RvD1 group. (b) Representative images and quantification of lesional oxCaMKII by immunofluorescence microscopic analysis. MFI data are shown as mean±s.e.m., one-way ANOVA with Tukey's multiple comparison test, *P<0.05 and ***P<0.001 early lesions versus other groups. ^P<0.05 Veh versus RvD1 groups, n=8 for early lesions, n=11 for advanced lesions/Veh group, and n=10 for advanced lesions/RvD1 group. Scale bar, 100 µm. (c) Representative confocal immunofluorescence images and quantification of 5-LOX localization in aortic lesional macrophages of 17-week WD-fed Ldlr-/- mice given vehicle control (Veh) or RvD1 during weeks 12–17. Macrophage F4/80 is red, 5-LOX is green, and Hoechst (nuclei) is blue. Arrows indicate nuclear region. In these images, nuclear 5-LOX appears as blotchy light blue staining within the dark blue Hoechst-stained nuclei. The quantitative data are shown as mean±s.e.m., t-test, *P<0.05, n=11 for Veh group and n=10 for RvD1 group. Scale bar, 5 µm.

Fig 3: SAXS-TT results for mouse brain.a Sagittal, axial, and coronal virtual slices of tensor-reconstructed reciprocal space-map intensity at myelin peak q-values. Grayscale colormap was used for the unspecific scattering signal, akin to unspecific tomographic contrasts such as MRI or CT. b Same virtual slices depicting the tensor-reconstructed myelin-specific signal. “Hot” colormap was used for the myelin-specific signal, akin to molecular imaging outcomes, e.g. using fluorescent tags. c 3D myelin distribution map of highly myelinated areas (red). d Coronal slice of the 3D fODF map, with tensors represented by ellipsoids, colored by tensor trace. e Side-view of 3D map of fiber orientations, represented by lines. Color and orientation correspond to the largest tensor eigenvalue. f Same coronal slice as in a, b, d from map in e. g Tractogram of same section based on main fiber orientation. h Box plots displaying myelin levels for different regions of the mouse brain, for right and left hemisphere, segmented using the DSURQE mouse brain atlas48. Notches represent median values, boxplot upper/lower ends represent 25/75 percentiles, and whiskers extend to ±2.7 standard deviations from the mean. Outliers are in red dots. All white matter regions (CIN, AC, CC, CP, CST, IC, CBP, OPT) displayed statistically similar myelin levels between left and right hemispheres using unpaired two-tailed t-test (no multiple comparison correction). [VEN: lateral ventricles, nright/nleft = 319/334, p value: 4 × 10-9; CRUS: crus 1&2, cerebellar lobule 7 (gray matter), nright/nleft = 7844/8591, p value: 2 × 10-6; OBGL: olfactory bulb, granule layer (gray matter), nright/nleft = 5846/5889, p value: 0.02; CIC: cingulate cortex, nright/nleft = 6273/6004, p value: 4 × 10-3; AMY: amygdala, nright/nleft = 11,191/12,069, p value: 0; SSC: secondary somatosensory cortex, nright/nleft = 4381/4318, p value:3 × 10-9; PVC: primary visual cortex, nright/nleft = 887/938, p value: 0.06; BF: barrel field, nright/nleft = 5695/6516, p value: 2 × 10-14; HP: hippocampus, nright/nleft = 29368/29811, p value: 10-25; HTH: hypothalamus, nright/nleft = 13,423/13,413, p value: 5 × 10-14; STR: striatum, nright/nleft = 28218/28766, p value: 3 × 10-13; TH: thalamus, nright/nleft = 22,630/22,812, p value:10-20; GLP: globus pallidus, nright/nleft = 3456/3723, p value: 0.28; CIN: cingulum, nright/nleft = 162/114, p value: 0.58; AC: anterior commissure, nright/nleft = 360/426, p value: 0.43; CC: corpus callosum, nright/nleft = 15,467/15,441, p value: 0.81; CP: cerebral peduncle, nright/nleft = 2115/2178, p value: 0.17; CST: cerebrospinal tract, nright/nleft = 2081/1977, p value: 0.54; IC: internal capsule, nright/nleft = 3304/3221, p value: 0.84; CBP: cerebellar peduncle, nright/nleft = 3513/3748, p value: 0.21; OPT: optic tract, nright/nleft = 1862/2056, p value: 0.70]. Source data are provided as a Source Data file.

Fig 4: Human vulnerable atherosclerotic plaques have decreased 5-LOX SPM levels and a lower SPM:LT ratio than stable lesions.(a) Flash-frozen human plaques were separated into vulnerable (V) and stable (S) regions (left panel), which were quantified for fibrous cap thickness:lesion area ratio (top right panel) and percent plaque necrosis (bottom right panel). (t-test, **P<0.01, n=8 donors). (b) Quantification of key DHA-derived lipid mediators, with their respective biosynthetic pathways indicated (LOX, lipoxygenase). (c) Quantification of key AA-derived lipid mediators. (d) Comparison of 5-LOX-derived SPMs (RvD1 and LXA4) and the SPM:LT ratio in vulnerable versus stable regions. For (b–d), t-test, *P<0.05, n=15 donors. (e) Quantification of RvD1 by ELISA in human macrophages that were incubated for 5 h with vehicle control, or 35 µM of 7KC, followed by Veh or 10 µM DHA, for an additional 40 min, and then the media were assayed for RvD1 with or without a 1 h pre-incubation with 10 µM NAC. Data are shown as mean±s.e.m. (n=4 separate donors). Statistical analysis was conducted using one-way ANOVA with the Kruskal–Wallis test and Dunn's multiple comparison post-hoc analysis, *P<0.05 of n=4 separate donors. (f) Quantification nuclear:non-nuclear 5-LOX ratio by confocal microscopy in macrophages incubated as in e, with two additional groups, NAC alone and 7KC+NAC. Images were acquired on a Leica confocal microscope, and nuclear:non-nuclear 5-LOX MFI was quantified using Image J. Data are shown as mean±s.e.m. (n=4 separate donors). Statistical analysis was conducted using one-way ANOVA with the Kruskal–Wallis test with the Dunn's multiple comparison post-hoc analysis, *P<0.05.