{"paper_id":"cfdbce87-45d5-4f97-b557-d6df89759a3f","body_text":"1 \n12/15-Lipoxygenase orchestrates murine wound healing via PPARg-activating \noxylipins acting holistically to dampen inflammation. \n \n \nChristopher P Thomas*1, Victoria J Tyrrell2, James J Burston2, Sam R C Johnson2, Maceler \nAldrovandi2, Jorge Alvarez -Jarreta2, Rossa Inglis 2, Adam Leonard 2, Lydia Fice 2, Jeremie \nCostales2, Stefania Carobbio 3, Antonio Vidal -Puig4, Majd Protty 2, Carol Guy 2, Robert \nAndrews2, Barbara Szomolay 2, Ben C Cossins 2, Ana Cardus Figueras2, Simon A Jones2 \nand Valerie B O'Donnell*2  \n \n \n1School of Pharmacy and Pharmaceutical Sciences, Cardiff University, Cardiff, CF10 3NB, \nUK. 2Systems Immunity Research Institute and Division of Infection and Immunity,  School \nof Medicine, Cardiff University, Cardiff, CF14 4XN, UK, 3University of Cambridge Metabolic \nResearch Laboratories Institute of Metabolic Science, Addenbrookes Hospital, Cambridge, \nUK. 4Centro de Investigacion Principe Felipe, Valencia, 46012, Spain.  \n \n*Address correspondence to:  Valerie O’Donnell, o-donnellvb@cardiff.ac.uk, or \nChristopher Thomas, thomascp@cardiff.ac.uk \n \nClassification: Biological sciences, Immunology and Inflammation \nKeywords: lipid, wound, oxylipin, lipoxygenase,  \n \n \n \nSignificance statement. \nDefective wound healing is a significant global clinical problem. Macrophage 12/15-\nlipoxygenase (12/15 -LOX, Alox15) generates abundant lipid med iators termed oxylipins  \nduring inflammation . However, its  physiological role  during resolving wound healing  is \nunclear, with studies so far assessing the bioactivity of individual lipids pharmacologically, \nrather than holistically in physiological amounts. Here, we report that Alox15 deficiency in \nmice caused a fibrotic response with failure to dampen inflammation, due to a dysregulated \nPPARg/adiponectin axis. Treatment of Alox15-/- wounds with physiological mixtures of \nPPARg-activating 12/15-LOX primary  monohydroxy products  restored the phenotype . \nSeveral transcriptional networks ( Elf4, Cebpb  and Tcf3) controlled by Alox15 were \nuncovered, identifying new targets for promoting physiological wound healing.  \n \n \n \n  \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted March 27, 2025. ; https://doi.org/10.1101/2025.03.26.645456doi: bioRxiv preprint \n\n 2 \nAbstract \n12/15-lipoxygenase (12/15-LOX, Alox15) generates  bioactive oxygenated lipids during \ninflammation, however its homeostatic role(s) in normal healing are unclear. Here, the role \nof 12/15 -LOX in resolving skin wounds was elucidated , focusing on  how its lipids act \ntogether in physiologically relevant amounts. In mice, wounding caused acute appearance \nof 12/15-LOX-expressing macrophages and stem cells, coupled to early generation of ~12 \nmonohydroxy-oxylipins and enzymatically oxygenated  phospholipids (eoxPL) . Alox15 \ndeletion increased a-smooth muscle actin , collagen deposition , stem cell/fibroblast \nproliferation, IL6/pSTAT3, pSMAD3, and IFN-γ levels. Conversely, CD206 expression, \nF480+ cell s, MMP9 and MMP2 activities were reduced . Alox15-/- skin was deficient in  \nPPARg/adiponectin activity. Furthermore, while pro-inflammatory genes were upregulated \nas normal during wounding, many including Il6, Il1b, ccl4, Cd14, Cd274, Clec4d, Clec4e, \nCsf3, and Cxcl2 failed to revert to baseline during healing, indicating disruption of an anti-\ninflammatory brake. Reconstituting Alox15-/- wounds with a physiological mixture of Alox15-\nderived primary oxylipins generated by  healing wounds restored MMP and dampened \ncollagen deposition. The oxylipin mixture activated PPARg in vitro, while in vivo, the PPARg \nco-activator, Helz2, was significantly upregulated. Additional inflammatory and proliferative \ngene networks impacted by  Alox15-/- included Elf4, Cebpb  and Tcf3, with many of their \nassociated genes significantly dysregulated. In summary, the impact of 12/15-LOX is \nascribed to the deficiency of  abundantly generated monohydroxy oxylipins acting together \nvia PPARg/adiponectin. The identification of multiple gene alterations reveals several new \ntargets for treatment of non-healing wounds. Our studies demonstrate that abundant 12/15-\nLOX oxylipins act together, dampening inflammation in vivo, revealing a need to consider \nlipid signaling holistically. \n \n \n \n \n \n \n \n \n \n \n \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted March 27, 2025. ; https://doi.org/10.1101/2025.03.26.645456doi: bioRxiv preprint \n\n 3 \n \nIntroduction \n12/15-LOX (Alox15) is a leukocyte enzyme highly expressed in murine resident peritoneal \nmacrophages. The human homolog, 15 -LOX1 ( ALOX15) is inducible in peripheral \nmonocytes in response to Th2 cytokines, and expressed basally in reticulocytes, eosinophils \nand airway epithelium(1, 2). Alox15-/- mice are protected against atherosclerosis, diabetes, \nhypertension and abdominal aortic aneurysm, and show reduced thrombosis, while \nconversely, they develop worse arthritis(3-7). This indicates that the pathway is a significant \nplayer in inflammatory vascular disease. However, while central roles in disease are \nestablished, the function of 12/15-LOX in normal healing is less clear.   \n \n12/15-LOX generates families of structurally related lipid mediators through oxidation of \nunsaturated fatty acids (FA) and complex lipids, including phospholipids (PL) and cholesteryl \nesters (CEs)(8-11). The monohydroxy forms of oxidized FAs are first generated by LOXs , \nwith the most abundant being usually derived from arachidonate (AA) . These 12/15-LOX \nderived lipids can independently mediate bioactions relevant to inflammation, such as \nactivation of PPAR g (which dampens cytokines such as IL6 and TNF a)(7, 12-20). Studies \nup to now generally focused on their bioactions when added individually,  for example(21-\n24). However, in vivo they are not generated in isolation but in mixtures  comprising large \nnumbers of species at varying amounts. This is particularly relevant to PPAR g, which \nrecognizes overall ligand “tone” at relatively low affinity, rather than specific lipid structures \nat high affinity via GPCRs. The most quantitively abundan t free acid  12/15-LOX products \nare monohydroxy FAs, from arachidonic acid (AA) and other polyunsaturated fatty acids.  \nAdditionally, “specialized pro-resolving mediators” (SPM), such as resolvins, protectins, and \nmaresins are described as rarer products of the pathway (25). Here, the primary \nmonohydroxy FAs are further metabolized, generating oxygenated di- and tri-hydroxy FAs \nreported to signal via activation of GPCRs that include ALX/FPR2, DRV1/GPR32, \nDRV2/GPR18, and ERV1/ChemR23 (26, 27) . However, while SPM can dampen \ninflammation pharmacologically, their endogenous generation and GPCR binding were \nrecently queried(28-34).  \n \nSkin wounding (punch biopsy) represents a tractable model of  physiological inflammation \nresolution that includes  four phases: hemostasis, inflammation, proliferation, and \nremodeling, representing an ideal model in which to test the impact of Alox15. Throughout \nthis, lymphoid, myeloid, and tissue-resident cells interact , producing signaling molecules \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted March 27, 2025. ; https://doi.org/10.1101/2025.03.26.645456doi: bioRxiv preprint \n\n 4 \nwhich work in an orchestrated manner. During hemostasis, clotting factors and angiogenic \nfactors decrease bleeding and stimulate formation of new blood vessels (35). During the \ninflammatory phase, neutrophil and macrophage infiltration supports release of chemokines \nand cytokines, inflammatory agents and antigen control factors (36). Later, the proliferation \nphase is characterized by fibroblast and keratinocyte migration from the wound edge, \nmediating contraction and closure (37). Last, during remodeling, increased deposition and \ncross-linking of collagen takes place, balanced with removal of excess extracellular matrix \nby myofibroblast -derived collagenases called matrix metalloproteinases  (MMPs)(38). \nHerein, we used genetic, transcriptomic and lipidomic approaches to determine the role of \nAlox15 and its lipids i n physiological skin wound healing . We found that  the gene plays a \ncritical role in ensuring that the healing response is finely tuned, to enable effective healing. \nWithout 12/15-LOX, cellular and tissue responses proceed at accelerated rates suggestive \nof fibrosis. Abundant monohydroxy FA s, many already known PPAR g ligands, appear  \nresponsible for the phenotype when applied in physiological amounts. Our study highlights \na central role for Alox15 in normal healing, defines several new potential targets for \npromoting healing, and shows the need to consider lipid biology in a holistic manner when \ndelineating cellular signaling roles that drive health and disease.  \n \n  \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted March 27, 2025. ; https://doi.org/10.1101/2025.03.26.645456doi: bioRxiv preprint \n\n 5 \nMethods \n \nAnimal model  \nMice (8-12 weeks old C57/B6/J ) were purchased from Charles River UK (Margate, UK) , \nwhile Alox15-/- mice were bred in house (F11, C57BL/6J) in isolators. All animal experiments \nwere performed in accordance with the United Kingdom Home Office Animals (Scientific \nProcedures) Act of 1986, under License (PPL 30/3334) . Generating of healing wounds is \ndescribed in Supplementary Methods. \n \nGeneration of histological tissue sections and staining protocols   \nAt various time points up to 14 days, wounds were harvested, processed and stained either \nusing DAB or fluorescence immunohistochemistry, as described in Supplementary Methods. \nCollagen was stained using Masson Trichrome, and images acquired and analyzed using \nmicroscopy as described in Supplementary Methods.  \n \nRNASeq.   \nWound tissue dissected from 2 mice (8 wounds in total, 4 wounds per mouse) to generate \neach sample (n=4/condition) were snap-frozen in liquid N 2 before being stored at -80 0C. \nRNA was isolated using RNeasy MinElute Cleanup Kit (Catalogue number 74204 Qiagen, \nMD, USA), as described in Supplementary Methods. Total RNA was depleted of ribosomal \nRNA and sequencing libraries prepared with the Illumina®TruSeq Stranded Total RNA \nLibrary Prep Gold (Illumina, Inc) kit using TruSeq CD Index Adapters1 (Illumina, Inc) . RNA \nwas sequenced using a 75 -base paired-end (2x75bp PE) dual index read format on the \nHiSeq4000 (Illumina, Inc) according to the manufacturer’s instructions , as described in \nSupplementary Methods. \n \nLipid extraction   \nWounds were harvested, homogenized as outlined in Supplementary Methods. Internal \nstandards (5 ng each of PC  14:0_14:0 and PE 14:0_14:0) and 5 ul of eicosanoid internal \nstandards were added. Samples were extracted using a solvent extraction (eoxPL) and solid \nphase extraction (oxylipins) as outlined in Supplementary Methods. Lipids were \nreconstituted using methanol and stored at -80 °C until LC/MS/MS. \n \nLC/MS/MS analysis of oxylipins and eoxPL.  \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted March 27, 2025. ; https://doi.org/10.1101/2025.03.26.645456doi: bioRxiv preprint \n\n 6 \nLipids were quantified using reverse phase LC/MS/MS as described in Supplementary \nMethods. Assay parameters are  provided in Supplementary Table 3 and (39) for oxylipins \nand Supplementary Table 5 for eoxPL. For chiral analysis, lipids  were separated using a \nChiralpak IA-U column (50×3.0 mm, Diacel) in reverse phase mode, with assay parameters \nas for oxylipins.  \n \nGel zymography for MMP activity   \nWounds were snap frozen then homogenized and analyzed using Novex™ 10% Zymogram \nPlus (Gelatin)) gels (Thermo Fisher), as described in Supplementary Methods.  \n \nCell transfection and reporter assays.  \nHEK293 cells were transfected with mouse PPARγ and the Firefly luciferase under the \ncontrol of 3x Ppar Responsive Element (PPRE) (40), as described in Supplementary \nMethods. \n \n \n \nResults \nTissue 12/15-LOX is acutely induced by wounding and associated with higher macrophage \nnumbers.  \nThe typical architecture of a wild -type mouse punch wound is shown, showing the dermis, \nwound bed, scab and wound edge (Figure 1 A). Wounding caused a significant increase in \n12/15-LOX+ve cells in the skin at 24 hrs (Figure 1 B-D). Most expression was associated with \ntissue localized F480+ve macrophages (Figure 1 C). 12/15-LOX was also induced in stem \ncells located at the base of hair follicles adjacent to the wound, but not distal from it (Figure \n1 D). This expression pattern suggests that a soluble mediator signaling in response to \nwounding maybe responsible for induction and that the enzyme is upregulated early post-\nwounding in both cell types . The total number of F480 +ve monocytes/macrophages in the \nwound at Day 1 was not impacted by Alox15 deletion, although there was some reduction \nlater, on Days 4 and 7 (Supplementary Figure 1 A,B). In contrast, neutrophil numbers in the \nsub-endothelial compartment were unaffected by Alox15 deletion at any timepoint  \n(Supplementary Figure 1 C).  \n \nAlox15 deletion alters the phenotype of the healing wound, promoting fibroblast stem cell \nproliferation and differentiation.  \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted March 27, 2025. ; https://doi.org/10.1101/2025.03.26.645456doi: bioRxiv preprint \n\n 7 \nHealing is characterized by stem cell and fibroblast proliferation , collagen deposition, \nreconstituting the underlying tissue, as well as epithelial migration and differentiation to form \na new covering. Here, smooth muscle actin (myofibroblast marker) and collagen deposition \nwere elevated on Alox15 deletion (Supplementary Figure 1 D,E, Figure 1 E -G). This was \nmainly noted during the remodeling phase (day 14), where the majority of the dermal layer \nin Alox15-/- wounds was collagen dense (Figure 1 G). Next, we profiled the stem cell marker \nSSEA3 and the nuclear protein Ki-67, a marker for proliferating cells, in wound beds at day \n4. Both were increased in Alox15-/- with SSEA3 being significantly higher, s uggesting that \nthe healing wound at this early stage has a higher number of actively proliferating stem cells \n(Supplementary Figure 1 F-H). We next determined re-epithelialization of the wound during \nthe inflammatory (day 4) stage using cytokeratins 10 (C10, pink) and 14 (C14, green), which \ndetermine epithelial (keratinocyte) cell migration from the wound edge into the bed. Basal \nkeratinocytes which are mitotically active express C14, but during differentiation, they lose \nC14 and upregulate C10(41). At day 4, the migratory distance of C14+ve and C10+ve epithelial \ncells into the wound edge was similar for both strains (Supplementary Figure 2 A,B). Overall, \nthe profiles suggest that wound bed keratinocyte differentiation isn’t impacted by Alox15-/-. \nFurthermore, the phenotype of non-wounded skin was similar, where in both wild-type and \nAlox15-/- mice, C14 is seen to be expressed lower in the epithelium and associated with hair \nbundle cells, with C10 mainly in terminally differentiated (dead) keratinocytes (corneocytes) \non the surface ( Supplementary Figure 2 C ). Overall, the data indicate that while fibroblast \nand stem cell differentiation and proliferation in the dermis is impacted, keratinocyte \ndifferentiation on the surface of the wound isn’t significantly affec ted by the absence of \n12/15-LOX.  \n \nElevated TGFb/IFNg/IL6 activity is seen in the absence of Alox15.  \nNext, a series of inflammatory pathways were profiled using immunohistochemistry. Protein \nexpression of IL6 was slightly but not significantly higher (Supplementary Figure 2 D) , but \nthere was  significantly elevated pSTAT3 and pSMAD3 (activated by TGF b) detected in \nAlox15-/- wounds (Figure 2 A,B). Increased IFNg was also detected, primarily on epithelial \ncells, while conversely CD206/mannose receptor (a marker of M2 cells ), was somewhat \nreduced (Figure 2 C,D, day 4). Taken together with the collagen, fibroblast and proliferation \ndata, a pro-inflammatory/pro-fibrotic phenotype is suggested for Alox15 deficiency. \n \nAlox15-/- wounds show  reduced matrix metalloprotease (MMP) activities during wound \nhealing.  \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted March 27, 2025. ; https://doi.org/10.1101/2025.03.26.645456doi: bioRxiv preprint \n\n 8 \nMMPs are collagenases that play a crucial role in regulating the extracellular matrix \narchitecture during wound healing by removing and recycling collagen . Since Alox15-/- \nwounds showed elevated collagen deposition, zymography was used to evaluate the activity \nof critical isoforms, MMP2 (active and pro -forms) and MMP9 on day 7 post-wounding. For \nall three, collagenase activity was significantly reduced in Alox15-/-  wounds (Figure 2 E).  \n \nThe temporal profile of oxylipins is altered by Alox15-/- with many lipids reduced/absent \nUsing reverse phase LC/MS/MS, ~100 oxidized fatty acids were profiled  in wound tissue , \nincluding well-known prostaglandins, thromboxane, eicosanoids, docosanoids  and several \nSPMs. A summary of the 68 lipids detected is shown in a heatmap (Supplementary Figure \n3 A). Several monohydroxy lipids were strongly elevated at day 1 but absent in Alox15-/- \nwounds (15-HEPE, 14-HDOHE, 17-HDOHE, 13-HOTrE) (Figure 3 A). 12-HETE/12-HEPE, \nand 15-HETE/15-HETrE which are generated by 12/15-LOX but also by  platelet 12-LOX \nand COXs, were also highly increased and were reduced 50% in Alox15-/- wounds (Figure \n3 B). All these peaked at day 1, then declined subsequently, paralleling the early transient \nexpression of 12/15 -LOX in the wound . This indicates that free oxylipin generation from \n12/15-LOX is acute and transient, peaking during the inflammatory phase, with lipids being \nreduced back towards basal levels during the healing phase.   \n \nSeveral prostaglandin dehydrogenase (PGDH) metabolites generated from the oxidation of \nHODEs and HETEs (9-, 13 -oxoODE, and 15 -oxo-ETE) were similar in both strains, but \npeaked at day 4 with higher levels in Alox15-/- (Supplementary Figure 3 B ). Lipids from 5-\nLOX (5-HETE, LTB4) and COX (PGE2, PGD2, 11-HETE, 11-HEPE, TXB2 and other PGE2 \nisomers) were strongly elevated at day 1  and declined after but  were not impacted by \nAlox15-/- (Supplementary Figure 3 B). The LA products 9- and 13-HODE were elevated early \nand were slightly lower on day 1 in Alox15-/- (Supplementary Figure 3). Several \ncytochromeP450/soluble epoxide hydrolase (sEH) metabolites were detected at very low \namounts with small increases around day 4 which fell by days 7 and 14 (5,6-diHETrE, 8,9-\ndiHETrE, 11,12 -diHETrE, 14,15 -diHETrE, L TB4, 5,6 -EET, 7,8 -EpDPA, 13,14 -EpDPA) \n(Supplementary Figure 4). Last, 9,10-EpOME and 12,13 -EpOME elevated beyond day 4, \nalthough levels fluctuated significantly, similar to their sEH metabolites 9,10-diHOME and \n12,13-diHOME (Supplementary Figure 4). None were reduced by Alox15-/- indicating they \noriginated from other biochemical or non-enzymatic pathways. Indeed, many from CYP/sEH \nwere significantly higher at day 4 .  In relation to SPM, out of several monitored, only trace \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted March 27, 2025. ; https://doi.org/10.1101/2025.03.26.645456doi: bioRxiv preprint \n\n 9 \namounts of resolvinD5 were detected, with detailed structural analysis including chiral \nchromatography shown in Supplementary Methods and Data.   \n \nEnzymatically-oxidized phospholipids (eoxPL) are generated during wound healing, peaking \nduring the proliferative stage, with 12 -HETE-PEs significantly impacted by Alox15 \ndeficiency.  \nFree oxylipins generated by COXs or LOXs are also formed as complex lipids attached to \nmembrane phospholipids (PL), termed eoxPL. The most abundant are HETE -containing \nphosphatidylethanolamine (PE), either generated by direct attack on PE by 12/15-LOX, or \nby esterification of newly formed HETEs to lysoPE (10). 12/15-LOX generates 12 -HETE-\ncontaining eoxPL, while 5-HETE-PE arise via 5 -LOX (neutrophils)(8, 42). Platelet 12-LOX \nis a source of 12 -HETE-PEs(43). 15-HETE-PEs form either via 12/15-LOX, or through \nesterification of 15 -HETE by CO X, which is also a source of 11 -HETE-PEs. Following \nwounding, sustained elevations of 5, 11, 12, and 15 -HETE-PEs occurred, peaking on day \n4, then declining  (Figure 6 A , Supplementary Figure 5  A). Each represents a series of \nisomers differing by sn1 fatty acid, with 3-4 per HETE isomer. (Figure 3 C, Supplementary \nFigure 5 A). 8-HETE-PE were below LOQ indicating that there is little/no non -enzymatic \noxidation and confirming that the others are from LOX and COX. Overall, esterified HETEs \nwere less abundant than their corresponding free acid species. Consistent with generation \nby 12/15-LOX, 12-HETE-PE was reduced by >50% in Alox15-/- wounds, while others were \nunaffected (Figure 3 C). Individual 12-HETE-PE isomers were also all significantly lower in \nAlox15-/- wounds ( Supplementary Figure 5 B ). These data confirm  enzymatic origin, but \nsimilar to free HETE, a significant amount is from other sources , for example platelet 12S- \nor skin 12 R-LOXs. All HETE-PEs peaked around days 5 -10, later than free acid HETEs  \n(Figure 3 C). Thus, as free HETEs declined, the esterified forms were elevating. This may \nreflect onset of esterification processes driven by Lands cycle.  \n \nMMP activities and collagen deposition are restored to wild-type levels in Alox15-/- wounds \nby high abundance oxylipins, but not eoxPL.  \nOxylipins are not generated in isolation but as complex mixtures in vivo. Here, lipidomics \ndata informed formulation of  relevant mixtures to  add to healing  Alox15-/- wounds \n(Supplementary Table 1) . High oxylipins contained lipids generated acutely in high er \namounts that were relatively deficient in Alox15-/- wounds, with amounts added via topical \ndermal delivery aiming  to match the maximum levels detected post-wounding (12-HETE, \n17-HDOHE, 13-HOTrE, 15-HETE, 12-HEPE, 12-oxo-ETE, 15-HEPE, 12(13)-EpOME, 13-\n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted March 27, 2025. ; https://doi.org/10.1101/2025.03.26.645456doi: bioRxiv preprint \n\n 10 \nHODE, 14-HDOHE). Wounds treated with high oxylipins demonstrated increased levels of \nMMP9, aMMP2 and pMMP2 ( Figure 3 D,E, Supplementary Figure 6 A ). Significantly, this \ntreatment reversed the phenotype so that MMP activities were significantly elevated to \nbetween WT and Alox15-/- levels (Figure 3 F, Supplementary Figure 6 B ). No impact was \nseen with a pharmacological dose of PE 18:0a/12-HETE, an eoxPL which was 50 % reduced \nby Alox15 deletion (Figure 3 D,E, Supplementary Figure 6 A ). Next, the ability of lipid s to \nreduce the accelerated  collagen deposition of Alox15-/- was tested. Here, treatment with \nvehicle alone caused a non-significant increase in collagen, but this completely suppressed \nby the high oxylipin preparation, but as for MMPs, there was no impact of eoxPL (Figure 4 \nA,B).   \n \nTranscriptional analysis reveals reduced anti -inflammatory lipid metabolism  in Alox15 -/- \nhealthy skin, but a relatively normal acute response to wounding.  \nTo identify transcriptional networks modulated by Alox15-/-, RNASeq was performed on day \n0 (healthy tissue), and days 4 and 7 post-wounding. At Day 0, 143 genes were significantly \ndifferent between the  two strains (adjusted p -value < 0.05 ) (Supplementary Table 6). \nAnalysis of these using Ingenuity Pathway Analysis  (IPA) identified lipid metabolism as \nhighly represented, and a subset of relevant genes in that network is shown  (Figure 4 C). \nStrong downregulation of Adipoq (adiponectin) and Pparg (PPARg) was seen, associated \nwith a reduction in a series of genes that either control or are controlled by these (Figure 4 \nC) (44-55). PPARg is a transcription factor that induce s adiponectin(56), and it responds \ndirectly to oxylipin  ligands generated by Alox15 including several  HODEs, HETEs and \nHDOHEs(13, 16 -19). Adiponectin is a hormone and adipokine centrally involved in \nmetabolism, that is  protective against a number of inflammatory conditions such as \natherosclerosis and type 2 diabetes(57, 58). Both are crucial in mediating anti-inflammatory \nactions such as inhibition of pro-inflammatory NFkB signaling and NLRP3(59, 60). Additional \ndown-regulated genes in the network include regulators of lipid metabolism such as Slc27a1 \n(import of long -chain fatty acids), Acsm5 (Acyl-CoA synthetase medium -chain family \nmember 5), Gpd1 (regulates lipid metabolism), and two genes that regulate adipose tissue \ndevelopment, Adig (adipogenin) and Lgals12. Importantly, reduced  basal expression of \nAdipoq and Pparg indicates that Alox15-/- tissues would struggle to mount the anti-\ninflammatory response  required to counterbalance inflammation during the  later wound \nhealing phase in which PPARg signaling is known to play a role (61). At day 4, comparison \nbetween WT and Alox15-/- wounds showed that around 42 genes were significantly different \n(Supplementary Table 7) . These genes didn’t appear obviously functionally related. \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted March 27, 2025. ; https://doi.org/10.1101/2025.03.26.645456doi: bioRxiv preprint \n\n 11 \nHowever, the classic inflammatory response to injury, as measured by genes that include \nTnfa, Il1b, IFNg, Nlrp3, Cxcl2, Ccl4, and Il6 was preserved in both strains indicating a normal \nonset of inflammation (Figure 4 D).  \n \nLate wound healing in Alox15 -/- wounds shows a failure of inflammation to reduce to basal \nlevels, with many pro-inflammatory genes remaining upregulated. \nAt day 7, 79 genes were significantly different between the strains, with 60 higher in Alox15-\n/- wounds than WT (Supplementary Table 8). A Cytoscape analysis was performed using the \nwhole-time course dataset for the 79 genes (day 0, 4, 7 and both strains), and a sub-group \nof 45 were seen to strongly correlate, suggesting their behavior was co-ordinated during the \nentire wounding and healing process (Figure 4 E). These included several pro-inflammatory \ngenes such as Ccl4, Cd14, Cd274, Clec4d, Clec4e, Csf3, Cxcl2,Cxcl3, Fpr2, Il1b, Il6, Irg1, \nNfkbiz, Nlrp3, Ptgs2, Retnlg, Trem1  and Osm. Many are involved in  macrophage-driven \ninflammation, and all were highly upregulated in both WT and Alox15-/- wounds on day 4. \nHowever, in Alox15-/- wounds these all failed to reduce back to basal levels at day 7, with \ntheir transcription remaining around 50% of the day 4 levels  (Figure 5 A, Supplementary \nFigure 7). This contrasts with WT wounds where these fully returned to basal levels by day \n7 (Figure 5 A , Supplementary Figure 7). IPA analysis of this sub-group of genes \ndemonstrated that many are upregulated through common mechanisms, such as NFkB. For \nexample, within this group, an IPA sub-network predicted higher activity of the IL1, IFNb and \nInflammasome pathways in the Alox15-/- wounds. Importantly, IL1, IFNb and Inflammasome \nare all well known to be downregulated by PPARg (Figure 5 B). Indeed, several genes in \nthis network are also known to be down-regulated by activation/induction of PPARg either \ndirectly or via inhibition of NF-kb, including Il6, Nlrp3 and Il1b (60, 62, 63). Last, it was seen \nthat while Pparg expression was not induced by wounding, its expression fell in WT wounds \nto levels similar to those seen in Alox15-/- (Figure 5 C). This suggests that the lack of PPARg \nsignalling during the healing response results from a relative deficiency in 12/15 -LOX-\nderived ligands, explaining why their supplementation could reduce the fibrotic phenotype \nin Alox15-/- wounds.  \n \n“High oxylipins” generated during the wound response induce transcriptional activity of \nPPARg in a reporter assay.   \nTo determine whether the more abundant oxylipins generated by 12/15 -LOX during \nwounding activate PPAR g, the high oxylipin mixture was tested in a reporter assay, with \nHEK293 cells expressing mouse PPAR g and Firefly luciferase under control of 3x Ppar \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted March 27, 2025. ; https://doi.org/10.1101/2025.03.26.645456doi: bioRxiv preprint \n\n 12 \nResponsive Element (PPRE) (40) exposed to the lipids for 24 hrs. Doses tested  were \ninformed by oxylipin amounts detected in wounds  in this study and studies which show in \nvitro oxylipin activation of PPAR g in the 10 -100 µM range (13, 15 -18).  First, amounts \ndetected in an individual wound on Day 1 were tested (Supplementary Table 1). These were \ndiluted into 50 µl media giving a final concentration of 0.4 – 1.2 µM total oxylipins. However, \nat these doses, PPARg was not reliably activated (not shown). Thus, we next tested amounts \nof oxylipins previously shown to bind and activate PPARg in vitro.  The most abundant lipid \nin our mixture was 12-HETE, which ranged from 12 -36 µM at the three doses tested , with \ntotal oxylipins at 40-120 µM (Supplementary Table 1).  Activation of PPARg was seen for all \noxylipin doses, with 80 µM showing a significant increase ( Supplementary Figure 8 A ). \nThese data are in line with previous reports that many individual 12/15-LOX products can \nact as low-affinity PPARg ligands at these concentrations, and that Alox15-deficiency leads \nto loss of PPAR g activation in macrophages (13, 15-19).  Overall, our data shows that the \nmixture of oxylipins can act activate PPARg in a complex mixture, in full agreement with \nprevious literature. \n \nWounding induces significant increases in gene expression of the PPAR g co-activator \nPdip1/Helz2/Pric285.   \nWe noted that the amounts of oxylipins used in vitro were higher than we detected in vivo. \nHowever, exact wound oxylipin concentrations are not possible to determine, and it is also \nnot known how much oxylipin enters the cells to bind and activate PPAR g in vitro . \nFurthermore, the ability of oxylipins to activate PPAR g may be regulated by known co -\nactivators present in vivo , including free fatty acids and protein co -activators(64, 65) .  \nPrompted by this, the expression of known protein co -activators was next interrogated in \ntranscriptional data(65).  One showed consistent and significant upregulation in both WT \nand Alox15-/- wounds on both days 4 and 7, versus day 0 (Figure 5 D ), with highest \nexpression seen on day 4.  This protein, PPAR g-DBD-interacting protein 1a , \nHELZ2/PRIC285 (Helz2) is a helicase that binds DNA binding domains of PPAR g through \nits C -terminal region, and can  enhance PPARg activation by troglitazone directly(66).  \nWhether HELZ2/PRIC285 can similarly enhance the ability of oxylipins to bind and activate \nPPARg is unknown and remains to be tested.  \n \nComparison of temporal changes in gene expression indicates additional transcription \nactivators regulated by Alox15 beyond PPARg, including Elf4, Cebpb and Tcf3.  \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted March 27, 2025. ; https://doi.org/10.1101/2025.03.26.645456doi: bioRxiv preprint \n\n 13 \nLast, RNASeq was deeper interrogated for additional Alox15-modulated transcriptional \nregulators, using a temporal analysis.  Here, we characterized individual strains separately \nand found three new candidates (see Supplementary Methods and Data  for full analysis).   \nElf4 is a known anti -inflammatory transcription regulator of inflammation, which targets \nseveral genes in the list, including Anln, Asf1b, Ccnb2, Cdca3, Cenpa, Cenpe, Cks2, E2f8, \nHmmr, Kif4a, Mcm10, Ndc80, Oip5, Rrm2, Tpx2 (67). Tcf3 promotes cell migration and \nwound repair (68), and Cebpb is involved in macrophage repair responses and inflammation \n(69, 70) . Many genes mapping to networks that regulate cytokines were also identified, \nfurther evidencing the impact of Alox15 on inflammatory signaling and identifying a large \nnumber of novel targets for further study (see Supplementary Methods and Data).  \n \n \nDiscussion \nIn this study , lipidomic and transcriptomic analyses  of Alox15-/- mice together reveal \nsignificant pathways which are impacted by deletion of the enzyme during wound healing. \nOverall, many pro-inflammatory genes highly induced by wounding fail to return to basal \nexpression in Alox15-/- wounds. Taken with our phenotypic data and the response to known \nPPARg ligands from this pathway, an intrinsic anti-inflammatory action of 12/15-LOX, driven \nby the PPARg/adiponectin axis is lost in Alox15-/- mice. This allows an uncontrolled fibrosis \nresponse, which is almost identical to that previously reported  in PPAR g-deficient mice  \n(described below)(71). Transcription factors, including Elf4, Tcf3 and Cebpb may also play \nimportant roles, but their precise functions remain to be established. \n \n12/15-LOX generates abundant mono -oxygenated oxylipins, eoxPL and in concert with \nother LOXs is proposed to generate rarer multiply oxygenated SPM via transcellular \nbiosynthesis. However, which lipids are primarily responsible for the effects of 12/15 -LOX \nduring physiological healing/resolution have been unclear. Its lipids are not generated in \nisolation but in complex mixtures of varying abundance  but studies testing their role in \ninflammation have usually added them singly(21-24). To test the role of the enzyme under \nphysiological conditions, we adopted a well-characterized model of wounding that fully heals \nwithin 14-days. Transcriptomic, phenotypic and lipidomic approaches revealed that Alox15 \ndeletion leads to an accelerated healing response resulting in a “fibrotic” phenotype. During \nthis, higher collagen deposition, increased stem cell proliferation and differentiation are \nseen, as well as higher levels of inflammatory markers such as IL6/pSTAT3, pSMAD3 and \nIFNg. A failure of pro -inflammatory gene expression to reduce back to baseline during the \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted March 27, 2025. ; https://doi.org/10.1101/2025.03.26.645456doi: bioRxiv preprint \n\n 14 \nlater remodeling phase was found. Our transcriptional data identified a basal deficiency of \nPPARg expression and activity and also showed that many genes which are upregulated by \ntranscription factors such as NLRP3 or NFkB (both antagonized by PPARg) did not revert to \nbasal levels post-wounding. Linked with this, wild-type wounds generated large amounts of \nseveral known PPAR g ligands via 12/15 -LOX ( e.g. 12 -HETE, 15 -HETE, 12 -HEPE, 13 -\nHOTrE) during the early inflammatory phase, and critical features of normal wound healing \ncould be restored in Alox15-/- by supplementing with physiological levels of a mixture of \nthese(13, 15-20).  Furthermore, several oxylipins that were deficient in Alox15-/- wounds are \nknown to dampen IL-6 (12-HETE, 15-HETE, 14-HDOHE, 15-HEPE(7, 15, 72)) and NLRP3 \n(13-HOTrE(73)), with these effects also likely  mediated by PPAR g. In vitro testing \nestablished that the mixture of oxylipins could activate PPAR g in vitro . A caveat is that \nconcentrations needed, although in line with many other studies on PPARg, appeared to be \naround 100-fold higher than those found in vivo.  This suggests that the lipids alone are not \nsufficient and protein co-activators known to sensitize PPARg to agonists may be involved, \nsuch as HELZ2/PRIC285 (Helz2) (66), which we also found to be upregulated significantly \nduring the wounding response.  Based on our staining data, the most likely cellular sources \nof the lipids will be F480+ve macrophages and stem cells at the base of hair follicles.   \n \nNotably, the phenotype seen in our study is almost identical to that of PPARg-deficient mice, \nwhere increased actin, collagen pSMAD3 and an accelerated healing/fibrotic phenotype  in \nskin were described(71). Also, loss of PPARg in skin fibroblasts is associated with elevated \npSMAD3, while PPAR g agonists directly reduce actin and collagen expression (74, 75) . \nFurthermore, PPARg blockade elevates MMP1 and MMP9 in fibroblast-like synoviocytes(76) \nwhile its activation dampens fibroblast proliferation and differentiation (77). PPAR g also \ninhibits expression of IL6, IFN g and pSTAT3, and prevents pSMAD3 dependent collagen \nsynthesis and deposition in fibroblasts(78-81). Thus overall, the Alox15-/- fibrotic phenotype \nmost likely results from a simple failure to generate mixtures of abundant PPARg ligands \nduring the acute response to injury. In line with this, an anti -inflammatory and pro-healing \naction of pharmacological PPAR g agonists such as rosiglitazone in mouse wound models \nof diabetes and obesity has been described previously(82, 83).  \n \nAs well as free oxylipins, eoxPL were detected, elevating significantly during the remodeling \nstage, with 12-HETE-containing isomers reduced almost to basal levels in Alox15-/- wounds. \neoxPL are also known as PPAR g ligands, although added herein, they did not restore \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted March 27, 2025. ; https://doi.org/10.1101/2025.03.26.645456doi: bioRxiv preprint \n\n 15 \nMMP/collagen, most likely due to insufficient concentration (84, 85). Apart from very low \namounts of RvD5, SPM were not generally detected. A recent study on murine cutaneous \nwounds reported several RvDs in mouse wounds, including RvDs1,2,3,4,5,6 and two 17R-\nisomers and proposed a central role for these lipids in repair through reconstitution studies \nwith exogenous lipids (86). There, a dministration of individual RvDs  at around 100 \nng/wound/day per isomer demonstrated a pharmacological effect on healing(86). Oxylipin \namounts administered in our study (6.5 ng total dose/wound/day) were considerably lower. \nThus, in the previous study, the mechanism could involve SPM acting via low affinity PPARg \nbinding and activation, similar to other oxylipins. In this regard, the closely related RvD1 \n(7S,8R,17S-triHDOHE) was previously reported to activate PPAR g in a mouse model of \nacute lung injury (87).  Alternatively, a  recent study showed that RvDs can allosterically \nactivate the PGE 2 receptor, EP4, with RvD5 sensitizing at nM concentrations  (34). In our \nstudy, 7,17-diHDOHE (RvD5) was detected at extremely low amounts (max amount 1.5 \npg/wound, equating to around 4  fmol/wound). Although it’s not possible to calculate local \nconcentrations in wound s, these amounts appear too low to mediate EP4 sensitization or \nPPARg activation. In our study, the doses of oxylipins that were bioactive in vivo appeared \nto be around 100-fold lower than required for PPARg activation in vitro. The upregulation of \nco-activators such as Helz2 may provide a partial explanation since it can sensitize PPARg \nto agonists(66).  This remains to be experimentally tested in relation to oxylipins.  \n \nPPARg binds and is activated by  many diverse lipid ligands with relatively low affinity and \nlittle differentiation of enantiomeric structure . Thus, the concerted action of many agonists \ngenerated in relatively high amounts during the healing process is consistent with the known \nrole of PPARg in mouse wound healing(71). Here, our studies using Alox15-/- mice support \nthe idea that abundant lipids generated by the 12/15-LOX pathway act in concert to promote \nthe well-known anti-inflammatory actions of this transcription factor, preventing uncontrolled \nfibrosis through dampening inflammation directly. This is in line with the long-known action \nof many  12/15-LOX monohydroxy ligands  as PPAR g ligands, and previous reports of \ndefective PPAR g signaling in Alox15-/- macrophages(13), and supports consideration of \ntherapies targeting this pathway in defective wound healing in patients.  \n \nAcknowledgements:  \n \nFunding is acknowledged to VOD and CPT  from the Medical Research Council \n(MR/M011445/1), and from European Research Council (LipidArrays). We acknowledge \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted March 27, 2025. ; https://doi.org/10.1101/2025.03.26.645456doi: bioRxiv preprint \n\n 16 \nSarah Edkins and Shelley Rundell for technical support. Wales Gene Park is a Health and \nCare Research Wales-funded infrastructure support group . VOD acknowledges the Royal \nSociety Wolfson Merit Award Scheme. Funding to AVP is acknowledged by the European \nResearch Council (ERC, grant agreement No 669879). MP was funded by Wellcome Trust \nGW4-CAT Fellowship ( 216278/Z/19/Z). SC acknowledges a grant PID2021 -125406B-100 \nfrom MCIN/AEI /10.13039/501100011033/ and by ERDF a way of making Europe. We \ngratefully thank Hartmut Kühn, Humboldt University Berlin for the gift of antibody directed \nagainst 12/15-LOX, and Paul Martin, University of Bristol for assistance with the murine \nwound model.  \n \nAuthorship contributions:   \nJJB, SRCJ, VJT, MA, JAJ, RI, AL, LF, JC, BCC, CG, AC, SC conducted experiments.  JJB, \nCPT, VBO designed experiments.  JJB, VBO, CPT, AJC supervised experiments.  SAJ, JJB, \nVBO, CPT, AVP, SC  interpreted findings and provided additional intellectual input.  JJB, \nVBO drafted the manuscript. RA, BS performed statistical and informatic analysis of \nRNASeq data.  All authors edited the manuscript.  \n \nFigure Legends.  \n \nFigure 1, Wounding increases macrophage and hair follicle 12/15-LOX expression, \nand collagen levels. Panel A. Representative image of wound architecture. Panel B.  \nInduction of macrophage 12 /15-LOX by wounding  12/15-LOX+ve(green)/F4-\n80+ve(red)/DAPI+ve(blue) cells were measured at day 1 post wounding ( n= 5/group), data \nwere analyzed using one way ANOVA with Tukey post-hoc test * p<0.05, ** p<0.01. Panel \nC. Representative images from Panel B. Panels D. Expression of 12/15-LOX in hair follicles \nnear the wound edge post-wounding. Hair shafts are shown on day 1, wild-type mice. The \nleft and center panels show 12/15-LOX DAB+ve staining, and the right panel shows 12/15-\nLOX+ve green fluorescent staining (with DAPI counterstain). Panel E. Representative image \nof wound at low magnification showing region stained for collagen. Panel F. Collagen is \nelevated in Alox15-/- on days 7 and 14. Wounds were harvested and analyzed for collagen \nusing Masson’s Trichrome staining (collagen: blue, epithelial cells: deep red, nuclei: black, \nnon-collagen structures: pink) and pixels counted (n = 10-14/group). Unpaired Students t-\ntest, comparing WT and Alox15-/- separately * p<0.05, *** p, 0.005.  Panel G. Representative \nimages from Panel G.  \n \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted March 27, 2025. ; https://doi.org/10.1101/2025.03.26.645456doi: bioRxiv preprint \n\n 17 \nFigure 2. Alox15-/- wounds show elevated pSTAT3, pSMAD3, and INFγ but decreased \nCD206 and MMP activities . Panel A. Alox15-/- wounds show elevated pSTAT3 . pSTAT3 \nwas measured using fluorescence immunohistochemistry. n = 5 - 6/group. Panel B. Alox15-\n/- wounds show elevated pSMAD3 . pSMAD3 was measured using fluorescence \nimmunohistochemistry. n = 10-11/group. Panel C. Alox15-/- wounds show elevated IFN g. \nIFNg was measured using DAB immunohistochemistry. n = 9/group (4-6 fields per wound). \nPanel D. Alox15-/- wounds show reduced CD206 expression . CD206 was measured using \nDAB immunohistochemistry. n = 5/group ( 3-6 fields per wound ). For all panels data was \nanalyzed an unpaired t-test, * p < 0.05, ** p<0.01. Right panels show representative images \nfor all the proteins analyzed. Panel E. MMP activities are reduced in Alox15-/- wounds. MMP \nactivities were measured using zymography. n= 6/group. The ladder shows proteins \ncorresponding to 250, 148, 98, 64 and 50 kDa. Image J was used to calculate the density \nof each band. The gel is shown (right panel). The impact of Alox12-/- was analyzed using an \nunpaired t-test, mean ± SEM, * p < 0.05, ** p<0.01, *** p < 0.005. \n \nFigure 3. Alox15-/- wounds show lower levels of many oxylipins  and 12-HETE-PEs, \nwhile physiological levels of “high oxylipins” restore MMP activity . Panels A,B. \nOxylipins are rapidly elevated post -wounding, but many are reduced in Alox15-/- wounds. \nOxylipins were measured using LC/MS/MS as outlined in Methods. n = 6 samples/time point, \nwith 4 wounds pooled/sample . For all panels differences between groups were analyzed \nusing two -way Anova (red stars), with Bonferroni post hoc test between individual time \npoints (black stars), mean ± SEM, * p < 0.05, ** p<0.01, *** p < 0.005. Panel C. HETE-PEs \nelevate during healing, peaking on day 7, with significant loss of 12 -HETE-PE isomers in \nAlox15-/- wounds. Oxidized phospholipids were measured using LC/MS/MS as outlined in \nMethods (n = 5 samples/time point, with 4 wounds pooled/sample) . Unpaired t -test, * p \n<0.05, ** p < 0.01, *** p < 0.005. Panel D. “High oxylipins” restored MMP activities in wounds \nfrom Alox15-/- mice. Post wounding, lipids were added to wounds as indicated in Methods \nevery second day . Wounds were harvested and analyzed for MMP activities using \nzymography as in Methods (n = 4 , 4 – 5 wounds pooled/sample). Panel E. A gel showing \nrepresentative data from Panel C . Panel F. “High oxylipins” restore MMP activities to wild-\ntype level. Wounds were harvested and analyzed for MMP activities using zymography as \nin Methods (n = 4, 4 – 5 wounds pooled/sample). For Panels C,E, ANOVA with Tukey post \nhoc test was used, * p <0.05, ** p < 0.01, *** p < 0.005. \n \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted March 27, 2025. ; https://doi.org/10.1101/2025.03.26.645456doi: bioRxiv preprint \n\n 18 \nFigure 4. “High oxylipins” dampen collagen deposition, while unwounded Alox15-/- \nskin shows reduced PPAR g activity but a normal pro -inflammatory response to \nwounding, while a series of highly correlated genes fail to revert to baseline in Alox15-\n/- wounds at Day 7 . Panel A. Collagen generation is dampened by “high oxylipins” . Post \nwounding, lipids were added to wounds as indicated in Methods every second day. Wounds \nwere harvested at Day 7 and analyzed for collagen using Masson’s Trichrome staining  \n(collagen: blue, epithelial cells: deep red, nuclei: black, non -collagen structures: pink) and \npixels counted (n = 5-6 wounds/group. Panel B. Representative images from Panel A. Panel \nC. Unwounded  Alox15-/- skin shows significantly reduced PPAR g/adiponectin \nexpression/activity. RNASeq was carried out as indicated in Methods on non-wounded skin \n(n = 3 – 4 samples per group, each sample was a pool of 4 wounds per animal with 2 animals \nper pool = 8 wounds per sample). All genes shown are significantly reduced in Alox15-/- skin, \nand are either controlled by or regulate PPAR g/adiponectin. Panel D. Upregulation of a \nseries of canonical “inflammatory” genes is preserved in Alox15-/- wounds. Gene expression \nwas normalized for each gene to its Day 0 mean value, then expressed as fold-change (n = \n3 – 4 per group) each sample was a pool of 4 wounds per animal with 2 animals per pool = \n8 wounds per sample ). For Panel A, ANOVA with Tukey post hoc text . Panel E. A large \nnumber of genes that are significantly different between wild-type and Alox15-/- wounds at \nDay 7 highly correlate across the whole timecourse, indicating coordinated regulation. \nGenes that were found to be significantly differentially expressed at Day 7 were analyzed in \nCytoscape, using their expression levels for the entire time-course, with correlation [r] > 0.8 \nshown.  \n \nFigure 5. Genes that fail to revert at day 7 in Alox15-/- wounds remain 50% elevated  \nabove baseline , and many are controlled through NLRP3, IFN b and IL -1, \nPPARg expression is not upregulated during wounding, and Elf4 and Helz2 are \nupregulated during wounding . Panel A. All genes from the highly correlated network in \nFigure 4 E typically remain 50% elevated . Data from gene expression of highly correlated \ngenes was averaged and normalized to day 4, wild-type mean (the inflammatory response \nlevel) (n = 3 – 4 per group) . Panel B. IPA network analysis of genes that are significantly \ndifferent at Day 7 reveals significantly higher levels of Inflammasome, IFNb and IL -1 \npathways. Gene expression data from Day 7 was analyzed using IPA. Panel C. Pparg \nexpression is not increased during wounding. Transcriptional data on Pparg was compared \nacross the timecourse (n = 3 – 4 per group). For all gene expression data, students t -test, \nfollowed by Benjamin Hochberg correction : * p <0.05, ** p < 0.01, *** p < 0.005.   Panel D. \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted March 27, 2025. ; https://doi.org/10.1101/2025.03.26.645456doi: bioRxiv preprint \n\n 19 \nData from gene expression is shown for WT and Alox15-/- wounds during the time course (n \n= 3 – 4 per group).  For all gene expression data, students t -test, followed by Benjamin \nHochberg correction: * p <0.05, ** p < 0.01, *** p < 0.005. \n \n \n \nREFERENCES \n \n1. D. Heydeck  et al. , Interleukin -4 and -13 induce upregulation of the murine macrophage 12/15 -\nlipoxygenase activity: evidence for the involvement of transcription factor STAT6. Blood 92, 2503-\n2510 (1998). \n2. H. Kuhn, V. B. O'Donnell, Inflammation and immune regulation by 12/15 -lipoxygenases. Prog Lipid \nRes 45, 334-356 (2006). \n3. T. Cyrus  et al. , Disruption of the 12/15 -lipoxygenase gene diminishes atherosclerosis in apo E -\ndeficient mice. The Journal of clinical investigation 103, 1597-1604 (1999). \n4. D. Bleich et al., Resistance to type 1 diabetes induction in 12-lipoxygenase knockout mice. The Journal \nof clinical investigation 103, 1431-1436 (1999). \n5. P. B. Anning  et al. , Elevated endothelial nitric oxide bioactivity and resistance to angiotensin -\ndependent hypertension in 12/15-lipoxygenase knockout mice. Am J Pathol 166, 653-662 (2005). \n6. K. Allen -Redpath et al. , Phospholipid membranes drive abdominal aortic aneurysm development \nthrough stimulating coagulation factor activity. Proc Natl Acad Sci U S A 116, 8038-8047 (2019). \n7. G. Krönke  et al. , 12/15 -lipoxygenase counteracts inflammation and tissue damage in arthritis. J \nImmunol 183, 3383-3389 (2009). \n8. S. R. Clark et al., Esterified eicosanoids are acutely generated by 5 -lipoxygenase in primary human \nneutrophils and in human and murine infection. Blood 117, 2033-2043 (2011). \n9. B. H. Maskrey  et al. , Activated platelets and monocytes generate four \nhydroxyphosphatidylethanolamines via lipoxygenase. J Biol Chem 282, 20151-20163 (2007). \n10. V. B. O'Donnell, R. C. Murphy, New families of bioactive oxidized phospholipids generated by immune \ncells: identification and signaling actions. Blood 120, 1985-1992 (2012). \n11. P. M. Hutchins, R. C. Murphy, Cholesteryl ester acyl oxidation and remodeling in murine \nmacrophages: formation of oxidized phosphatidylcholine. J Lipid Res 53, 1588-1597 (2012). \n12. J. A. Ackermann, K. Hofheinz, M. M. Zaiss, G. Krönke, The double -edged role of 12/15-lipoxygenase \nduring inflammation and immunity. Biochim Biophys Acta Mol Cell Biol Lipids 1862, 371-381 (2017). \n13. J. T. Huang  et al., Interleukin-4-dependent production of PPAR -gamma ligands in macrophages by \n12/15-lipoxygenase. Nature 400, 378-382 (1999). \n14. Y. Wen et al., The role of 12/15 -lipoxygenase in the expression of interleukin -6 and tumor necrosis \nfactor-alpha in macrophages. Endocrinology 148, 1313-1322 (2007). \n15. J. Ávila-Román, E. Talero, C. de Los Reyes, S. García-Mauriño, V. Motilva, Microalgae-derived oxylipins \ndecrease inflammatory mediators by regulating the subcellular location of NFκB and PPAR -γ. \nPharmacol Res 128, 220-230 (2018). \n16. H. Pham, T. Banerjee, G. M. Nalbandian, V. A. Ziboh, Activation of peroxisome proliferator-activated \nreceptor (PPAR) -gamma by 15S -hydroxyeicosatrienoic acid parallels growth suppression of \nandrogen-dependent prostatic adenocarcinoma cells. Cancer Lett 189, 17-25 (2003). \n17. L. Sun et al., 12/15-Lipoxygenase metabolites of arachidonic acid activate PPARgamma: a possible \nneuroprotective effect in ischemic brain. Journal of Lipid Research 56, 502-514 (2015). \n18. A. Umeno et al., Comprehensive analysis of PPARγ agonist activities of stereo -, regio-, and enantio-\nisomers of hydroxyoctadecadienoic acids. Biosci Rep 40 (2020). \n19. R. Xu  et al. , Activation of peroxisome proliferator -activated receptor –γ by a 12/15 -lipoxygenase \nproduct of arachidonic acid: a possible neuroprotective effect in the brain after experimental \nintracerebral hemorrhage. Journal of Neurosurgery JNS 127, 522-531 (2017). \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted March 27, 2025. ; https://doi.org/10.1101/2025.03.26.645456doi: bioRxiv preprint \n\n 20 \n20. H. Yamada et al., Hydroxyeicosapentaenoic acids from the Pacific krill show high ligand activities for \nPPARs [S]. Journal of Lipid Research 55, 895-904 (2014). \n21. K. R. Chava et al., CREB-mediated IL-6 expression is required for 15 (S)-hydroxyeicosatetraenoic acid–\ninduced vascular smooth muscle cell migration. Arteriosclerosis, thrombosis, and vascular biology 29, \n809-815 (2009). \n22. J. K. Hampel et al., Differential modulation of cell cycle, apoptosis and PPARγ2 gene expression by \nPPARγ agonists ciglitazone and 9 -hydroxyoctadecadienoic acid in monocytic cells. Prostaglandins, \nleukotrienes and essential fatty acids 74, 283-293 (2006). \n23. K. Ma et al., 12-Lipoxygenase products reduce insulin secretion and β -cell viability in human islets. \nThe Journal of Clinical Endocrinology & Metabolism 95, 887-893 (2010). \n24. M. Tloti, D. Moon, L. Weston, J. Kaplan, Effect of 13 -hydroxyoctadeca-9, 11-dienoic acid (13-HODE) \non thrombin induced platelet adherence to endothelial cells in vitro. Thrombosis research 62, 305-\n317 (1991). \n25. C. N. Serhan, Pro -resolving lipid mediators are leads for resolution physiology. Nature 510, 92-101 \n(2014). \n26. C. N. Serhan, B. D. Levy, Resolvins in inflammation: emergence of the pro -resolving superfamily of \nmediators. J Clin Invest 128, 2657-2669 (2018). \n27. J. Pirault, M. Back, Lipoxin and Resolvin Receptors Transducing the Resolution of Inflammation in \nCardiovascular Disease. Front Pharmacol 9, 1273 (2018). \n28. N. Z. Homer, R. Andrew, D. W. Gilroy, Re-analysis of lipidomic data reveals Specialised Pro-Resolution \nLipid Mediators (SPMs) to be lower than quantifiable limits of assay in a human model of resolving \ninflammation. bioRxiv 10.1101/2023.03.06.530669, 2023.2003.2006.530669 (2023). \n29. N. H. Schebb  et al. , Formation, Signaling and Occurrence of Specialized Pro -Resolving Lipid \nMediators—What is the Evidence so far? Frontiers in Pharmacology 13 (2022). \n30. V. B. O’Donnell et al., Failure to apply standard limit-of-detection or limit-of-quantitation criteria to \nspecialized pro -resolving mediator analysis incorrectly characterizes their presence in biological \nsamples. Nature Communications 14, 7172 (2023). \n31. L. Kutzner et al., Development of an Optimized LC-MS Method for the Detection of Specialized Pro-\nResolving Mediators in Biological Samples. Frontiers in Pharmacology 10 (2019). \n32. N. H. Schebb et al., Comparison of the effects of long-chain omega-3 fatty acid supplementation on \nplasma levels of free and esterified oxylipins. Prostaglandins Other Lipid Mediat  113-115, 21 -29 \n(2014). \n33. C. Skarke et al., Bioactive products formed in humans from fish oils1[S]. Journal of Lipid Research 56, \n1808-1820 (2015). \n34. M. W. Alnouri  et al. , SPMs exert anti -inflammatory and pro -resolving effects through positive \nallosteric modulation of the prostaglandin EP4 receptor. Proceedings of the National Academy of \nSciences 121, e2407130121 (2024). \n35. M. Waser, W. H. Ziegler, C. Beretta -Piccoli, Mechanism of action of ketanserin: studies on \ncardiovascular reactivity in essential and diabetes -associated hypertension. J Hypertens 6, 471-479 \n(1988). \n36. A. Shedoeva, D. Leavesley, Z. Upton, C. Fan, Wound Healing and the Use of Medicinal Plants. Evid \nBased Complement Alternat Med 2019, 2684108 (2019). \n37. N. X. Landen, D. Li, M. Stahle, Transition from inflammation to proliferation: a critical step during \nwound healing. Cell Mol Life Sci 73, 3861-3885 (2016). \n38. H. Nagase, R. Visse, G. Murphy, Structure and function of matrix metalloproteinases and TIMPs. \nCardiovasc Res 69, 562-573 (2006). \n39. M. Misheva et al., Oxylipin metabolism is controlled by mitochondrial beta-oxidation during bacterial \ninflammation. Nat Commun 13, 139 (2022). \n40. S. Jitrapakdee  et al. , The peroxisome proliferator -activated receptor -gamma regulates murine \npyruvate carboxylase gene expression in vivo and in vitro. J Biol Chem 280, 27466-27476 (2005). \n41. F. Wang, A. Zieman, P. A. Coulombe, Skin Keratins. Methods Enzymol 568, 303-350 (2016). \n42. A. H. Morgan  et al. , Phosphatidylethanolamine -esterified eicosanoids in the mouse: tissue \nlocalization and inflammation -dependent formation in Th -2 disease. The Journal of biological \nchemistry 284, 21185-21191 (2009). \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted March 27, 2025. ; https://doi.org/10.1101/2025.03.26.645456doi: bioRxiv preprint \n\n 21 \n43. C. P. Thomas  et al., Phospholipid-esterified eicosanoids are generated in agonist -activated human \nplatelets and enhance tissue factor -dependent thrombin generation. J Biol Chem  285, 6891-6903 \n(2010). \n44. A. Benrick et al., Adiponectin protects against development of metabolic disturbances in a PCOS \nmouse model. Proc Natl Acad Sci U S A 114, E7187-E7196 (2017). \n45. K. Bhalla  et al. , N -Acetylfarnesylcysteine is a novel class of peroxisome proliferator -activated \nreceptor gamma ligand with partial and full agonist activity in vitro and in vivo. J Biol Chem  286, \n41626-41635 (2011). \n46. C. Cheadle et al., Regulatory subunits of PKA define an axis of cellular proliferation/differentiation in \novarian cancer cells. BMC Med Genomics 1, 43 (2008). \n47. Q. Liu  et al. , Adiponectin regulates expression of hepatic genes critical for glucose and lipid \nmetabolism. Proc Natl Acad Sci U S A 109, 14568-14573 (2012). \n48. S. Lobo, B. M. Wiczer, A. J. Smith, A. M. Hall, D. A. Bernlohr, Fatty acid metabolism in adipocytes: \nfunctional analysis of fatty acid transport proteins 1 and 4. J Lipid Res 48, 609-620 (2007). \n49. N. Maeda et al., Diet-induced insulin resistance in mice lacking adiponectin/ACRP30. Nat Med 8, 731-\n737 (2002). \n50. T. Ogura  et al. , Reduction of phosphodiesterase 3B gene expression in peroxisome proliferator -\nactivated receptor gamma (+/-) mice independent of adipocyte size. FEBS Lett 542, 65-68 (2003). \n51. D. Patsouris et al., PPARalpha governs glycerol metabolism. J Clin Invest 114, 94-103 (2004). \n52. E. Peverelli  et al. , PKA regulatory subunit R2B is required for murine and human adipocyte \ndifferentiation. Endocr Connect 2, 196-207 (2013). \n53. J. M. Way  et al. , Comprehensive messenger ribonucleic acid profiling reveals that peroxisome \nproliferator-activated receptor gamma activation has coordinate effects on gene expression in \nmultiple insulin-sensitive tissues. Endocrinology 142, 1269-1277 (2001). \n54. R. Y. Yang, D. K. Hsu, L. Yu, H. Y. Chen, F. T. Liu, Galectin -12 is required for adipogenic signaling and \nadipocyte differentiation. J Biol Chem 279, 29761-29766 (2004). \n55. S. Yu et al., Adipocyte-specific gene expression and adipogenic steatosis in the mouse liver due to \nperoxisome proliferator -activated receptor gamma1 (PPARgamma1) overexpression. J Biol Chem  \n278, 498-505 (2003). \n56. O. Astapova, T. Leff, Adiponectin and PPARgamma: cooperative and interdependent actions of two \nkey regulators of metabolism. Vitam Horm 90, 143-162 (2012). \n57. J. J. Diez, P. Iglesias, The role of the novel adipocyte-derived hormone adiponectin in human disease. \nEur J Endocrinol 148, 293-300 (2003). \n58. O. Ukkola, M. Santaniemi, Adiponectin: a link between excess adiposity and associated \ncomorbidities? J Mol Med (Berl) 80, 696-702 (2002). \n59. N. Ouchi, K. Walsh, Adiponectin as an anti-inflammatory factor. Clin Chim Acta 380, 24-30 (2007). \n60. K. J. Weber  et al. , PPARgamma Deficiency Suppresses the Release of IL -1beta and IL -1alpha in \nMacrophages via a Type 1 IFN-Dependent Mechanism. J Immunol 201, 2054-2069 (2018). \n61. M. Kapoor, F. Kojima, L. Yang, L. J. Crofford, Sequential induction of pro - and anti -inflammatory \nprostaglandins and peroxisome proliferators -activated receptor -gamma during normal wound \nhealing: a time course study. Prostaglandins Leukot Essent Fatty Acids 76, 103-112 (2007). \n62. Y. Cheng, S. Li, M. Wang, C. Cheng, R. Liu, Peroxisome Proliferator Activated Receptor gamma \n(PPARgamma) Agonist Rosiglitazone Ameliorate Airway Inflammation by Inhibiting Toll-Like Receptor \n2 (TLR2)/Nod -Like Receptor with Pyrin Domain Containing 3 (NL RP3) Inflammatory Corpuscle \nActivation in Asthmatic Mice. Med Sci Monit 24, 9045-9053 (2018). \n63. M. Heming et al., Peroxisome Proliferator -Activated Receptor-gamma Modulates the Response of \nMacrophages to Lipopolysaccharide and Glucocorticoids. Front Immunol 9, 893 (2018). \n64. I. Dussault, B. M. Forman, Prostaglandins and fatty acids regulate transcriptional signaling via the \nperoxisome proliferator activated receptor nuclear receptors. Prostaglandins & Other Lipid \nMediators 62, 1-13 (2000). \n65. S. Yu, J. K. Reddy, Transcription coactivators for peroxisome proliferator -activated receptors. \nBiochimica et Biophysica Acta (BBA) - Molecular and Cell Biology of Lipids 1771, 936-951 (2007). \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted March 27, 2025. ; https://doi.org/10.1101/2025.03.26.645456doi: bioRxiv preprint \n\n 22 \n66. T. Tomaru et al., Isolation and Characterization of a Transcriptional Cofactor and Its Novel Isoform \nthat Bind the Deoxyribonucleic Acid-Binding Domain of Peroxisome Proliferator-Activated Receptor-\nγ. Endocrinology 147, 377-388 (2006). \n67. P. M. Tyler  et al., Human autoinflammatory disease reveals ELF4 as a transcriptional regulator of \ninflammation. Nature Immunology 22, 1118-1126 (2021). \n68. Q. Miao et al. , Tcf3 promotes cell migration and wound repair through regulation of lipocalin 2. \nNature Communications 5, 4088 (2014). \n69. C. Y. Ko, W. C. Chang, J. M. Wang, Biological roles of CCAAT/Enhancer -binding protein delta during \ninflammation. J Biomed Sci 22, 6 (2015). \n70. D. Ruffell et al., A CREB -C/EBPbeta cascade induces M2 macrophage -specific gene expression and \npromotes muscle injury repair. Proc Natl Acad Sci U S A 106, 17475-17480 (2009). \n71. W. Sha, K. Thompson, J. South, M. Baron, A. Leask, Loss of PPARgamma expression by fibroblasts \nenhances dermal wound closure. Fibrogenesis Tissue Repair 5, 5 (2012). \n72. S. G. Wendell  et al. , 15 -Hydroxyprostaglandin dehydrogenase generation of electrophilic lipid \nsignaling mediators from hydroxy ω-3 fatty acids. J Biol Chem 290, 5868-5880 (2015). \n73. N. Kumar  et al. , 15 -Lipoxygenase metabolites of alpha -linolenic acid, [13 -(S)-HPOTrE and 13 -(S)-\nHOTrE], mediate anti -inflammatory effects by inactivating NLRP3 inflammasome. Sci Rep 6, 31649 \n(2016). \n74. H. A. Burgess  et al. , PPARgamma agonists inhibit TGF -beta induced pulmonary myofibroblast \ndifferentiation and collagen production: implications for therapy of lung fibrosis. Am J Physiol Lung \nCell Mol Physiol 288, L1146-1153 (2005). \n75. M. Kapoor et al., Loss of peroxisome proliferator -activated receptor gamma in mouse fibroblasts \nresults in increased susceptibility to bleomycin-induced skin fibrosis. Arthritis Rheum 60, 2822-2829 \n(2009). \n76. X.-F. Li  et al. , Functional role of PPAR -γ on the proliferation and migration of fibroblast -like \nsynoviocytes in rheumatoid arthritis. Scientific Reports 7, 12671 (2017). \n77. J. E. Milam et al., PPAR-gamma agonists inhibit profibrotic phenotypes in human lung fibroblasts and \nbleomycin-induced pulmonary fibrosis. Am J Physiol Lung Cell Mol Physiol 294, L891-901 (2008). \n78. R. Cunard et al., Repression of IFN-gamma expression by peroxisome proliferator-activated receptor \ngamma. J Immunol 172, 7530-7536 (2004). \n79. A. K. Ghosh  et al., Peroxisome proliferator -activated receptor-gamma abrogates Smad -dependent \ncollagen stimulation by targeting the p300 transcriptional coactivator. FASEB J 23, 2968-2977 (2009). \n80. G. Kokeny et al., PPARgamma is a gatekeeper for extracellular matrix and vascular cell homeostasis: \nbeneficial role in pulmonary hypertension and renal/cardiac/pulmonary fibrosis. Curr Opin Nephrol \nHypertens 29, 171-179 (2020). \n81. L. H. Wang  et al., Transcriptional inactivation of STAT3 by PPARgamma suppresses IL -6-responsive \nmultiple myeloma cells. Immunity 20, 205-218 (2004). \n82. G. Zhou, X. Han, Z. Wu, Q. Shi, X. Bao, Rosiglitazone accelerates wound healing by improving \nendothelial precursor cell function and angiogenesis in db/db mice. PeerJ 7, e7815 (2019). \n83. A. Siebert, I. Goren, J. Pfeilschifter, S. Frank, Anti -Inflammatory Effects of Rosiglitazone in Obesity -\nImpaired Wound Healing Depend on Adipocyte Differentiation. PLoS One 11, e0168562 (2016). \n84. P. Delerive  et al. , Oxidized phospholipids activate PPARalpha in a phospholipase A2 -dependent \nmanner. FEBS Lett 471, 34-38 (2000). \n85. H. Lee  et al. , Role for peroxisome proliferator -activated receptor alpha in oxidized phospholipid -\ninduced synthesis of monocyte chemotactic protein-1 and interleukin-8 by endothelial cells. Circ Res \n87, 516-521 (2000). \n86. J. Hellmann et al., Biosynthesis of D-Series Resolvins in Skin Provides Insights into their Role in Tissue \nRepair. J Invest Dermatol 138, 2051-2060 (2018). \n87. Z. Liao et al., Resolvin D1 attenuates inflammation in lipopolysaccharide -induced acute lung injury \nthrough a process involving the PPARγ/NF-κB pathway. Respir Res 13, 110 (2012). \n \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted March 27, 2025. ; https://doi.org/10.1101/2025.03.26.645456doi: bioRxiv preprint \n\n0\n5\n10\n15\n20\n25\n30\nUninjured Day 1\ninjured\nA\nBlue: DAPI (nuclei), Green: 12/15-LOX, Red: F480\nWild Type (Day 1)\n100 m\n12/15-LOX+ve\nmacrophages\nDermal wound bed\nDistal to wound Proximal to wound\n100 m\n100 m\nDay 1 Day 1\n200 m\nProximal to wound\nHair follicle\nHair shaft\n12/15-LOX+ve\nmacrophages\nHair follicle\nHair shaft\nDay 1\nHair follicle\n*\nNumber of F4/80/LOX+ve cells\nFigure 1\nSchematic of a wound showing typical features\nC\nD\n100 m\nB\nWou nd \nedgeScab\nScab\nWou nd \nedge 100 µm\nWild Type (unwounded)\nDermal bed\n0\n20\n40\n60\n80\n100\nWT Alox15-/- WT Alox15-/-\nDay 7 Day 14\n***\nCollagen staining density (MGI)\nWild Type Day 7\nWild Type Day 14\nAlox15-/- Day 7\nAlox15-/- Day 14\nG\nF\n125 µm\n62.5 µm\n 62.5 µm\n62.5 µm 62.5 µm\nepidermis epidermis\nepidermisepidermis\ndermis\ndermis\ndermis dermis\nE\n*\n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted March 27, 2025. ; https://doi.org/10.1101/2025.03.26.645456doi: bioRxiv preprint \n\n0\n5\n10\n15\n20\n25\n30\nWT Alox15-/-\n0\n100000\n200000\n300000\n400000\n500000\nWT Alox15-/-\n0\n5\n10\n15\n20\n25\n30\nWT Alox15-/-\n0\n200000\n400000\n600000\n800000\n1000000\n1200000\nWT Alox15-/-\nWild Type (day 4)\nWild Type (day 7)\nWild Type (day 7)\nWild Type (day 7)\nAlox15-/- (day 7)\nAlox15-/- (day 7)\nRed: pSTAT3\nBlue: DAPI\nRed: pSTAT3\nBlue: DAPI\nRed: pSMAD3\nBlue: DAPI\nRed: pSMAD3\nBlue: DAPI\n*\npSTAT3 immunostaining\n*\npSMAD3 immunostaining\nA\nB\nBrown: IFN Blue: hematoxylin\nBrown: CD206 Blue: hematoxylin\n200 µm\n200 µm\n100 µm100 µm Figure 2\n*** Wild Type (day 4) Alox15-/- (day 4)\nAlox15-/- (day 4)\nInterferon- expression\nC\nD *\nCD206 expression 100 m\n100 m\n100 µm 100 µm\n200 µm\n100 m\n200 µm\nepidermis\ndermis\nepidermis\ndermis\nepidermis\ndermis\nepidermis\ndermis\nepidermis\ndermis\nepidermis\ndermis\nDermal wound bedDermal wound bed\n0\n1000\n2000\n3000\n4000\n5000\nMMP9 pMMP2 aMMP2\nBand density (au) day 7\n* ** ***\nWildtype Alox15-/-\nMMP9\npMMP2\naMMP2\naMMP2 control lane\nLadder (kDa)\nE\n250\n148\n98\n64\n50\n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted March 27, 2025. ; https://doi.org/10.1101/2025.03.26.645456doi: bioRxiv preprint \n\n0\n5\n10\n15\n0 5 10 15\n5-HETE-PE\nWT\n0\n2\n4\n6\n8\n10\n12\n14\nAlox15-/- Vehicle eoxPL high\noxylipin\n0\n2\n4\n6\n8\n10\n12\n0\n2000\n4000\n6000\n8000\n0\n3000\n6000\n9000\npg /mg wound tissue\nDays post wounding Days post wounding Days post wounding Days post wounding\nC\nD\nF\npMMP2 band density\n(au, normalized to Alox15 -/-)\naMMP2 band density\n(au, normalized to Alox15 -/-)\naMMP2\nFigure 3\nMMP9 band density \n(au, normalized to Alox15 -/-) \n** **\n**\n* **\n*\n** **\n**\nband density (au)\n0\n4000\n8000\n12000\nMMP9 pMMP2\n** ** *** *\n0\n2\n4\n6\nMMP9\npMMP2\naMMP2\nAlox15-/- control\nE\n0\n1\n2\n3\n4\n5\n6\n7\n0 5 10 15\n11-HETE-PE\n0\n2\n4\n6\n8\n10\n12\n0 5 10 15\n12-HETE-PE\n0\n2\n4\n6\n8\n10\n12\n0 5 10 15\n15-HETE-PE\n*\n***\nA 15-HEPE\n0 5 10 15\n0\n5\n10\n15\n20\nWT\nALOX15-/-\nDays post wounding\npg/mg wound tissue\n14-HDOHE\n0 5 10 15\n0\n100\n200\n300\nDays post wounding\npg/mg wound tissue\n17-HDOHE\n0 5 10 15\n0\n100\n200\n300\nDays post wounding\npg/mg wound tissue\n13-HOTrE\n0 5 10 15\n0\n50\n100\n150\n200\n250\nDays post wounding\npg/mg wound tissue\n*** ***\n*\n*** ***\n**\nB\n12-HETE\n0 5 10 15\n0\n100\n200\n300\n400\n500\nWT\nALOX15-/-\nDays post wounding\npg/mg wound tissue\n12-HEPE\n0 5 10 15\n0\n2\n4\n6\n8\n10\nDays post wounding\npg/mg wound tissue\n15-HETE\n0 5 10 15\n0\n20\n40\n60\n80\n100\nDays post wounding\npg/mg wound tissue\n15-HETrE\n0 5 10 15\n0\n5\n10\n15\n20\nDays post wounding\npg/mg wound tissue\n10-HDOHE\n0 5 10 15\n0\n2\n4\n6\nDays post wounding\npg/mg wound tissue\n** ** **\nAlox15-/-\nAlox15-/-\n*** *** ***\n***\n*** * *** *** ***\n***\n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted March 27, 2025. ; https://doi.org/10.1101/2025.03.26.645456doi: bioRxiv preprint \n\nA B\nAlox15-/- co ntr ol\nAlox15-/- + high oxylipins Alox15-/- + eoxPL\nFigure 4\nepidermis\ndermis\nAlox15-/- + vehicle\ndermis\nCollagen staining density (MGI)\nAlox15-/- wounds\n0\n50\n100\n150\n200\n250\ncontrol vehicle high\noxylipins\neoxPL\n***\nC\n D\n0\n20\n40\n60\n80\n100\nWT KO\nFold change expression, \nnormalized to WT day 0\nIl6 Il1b Ccl4 Cxcl2\nCxcl3\nNlrp3 Ptgs2 Tnf\nLog2fold-change (KO vs WT)\nE\n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted March 27, 2025. ; https://doi.org/10.1101/2025.03.26.645456doi: bioRxiv preprint \n\n0\n0.2\n0.4\n0.6\n0.8\n1\n1.2\n1.4\n4833407H14Rik\n9830144P21Rik\nAnkrd33b\nCcl4\nCd14\nCd274\nClec4d\nClec4e\nCsf3\nCsrnp1\nCxcl2\nCxcl3\nEldr\nFpr2\nGm37691\nGm5483\nHcar2\nHdc\nIer3\nIl1b\nIl1bos\nIl1rn\nIl6\nIrg1\nMcemp1\nNabp1\nNfkbiz\nNlrp3\nNr4a3\nOsm\nPde4b\nPlek\nPpp1r15a\nPtgs2\nRab20\nRetnlg\nSamsn1\nSlc2a3\nSlc7a11\nSpp1\nSrgn\nStfa2l1\nThbs1\nTiparp\nTrem1\nWT Day  0 KO Day  0\nWTDay  4 KO Day 4\nWT Day 7 KO Day 7\nAll ex pressio n is n orm alized  to WT D ay 4\nAlox15-/- Day 7\nAlox15-/- Day 4\nDay 0  WT, \nDay 0 Alox15-/-\nDay 7 WT\nExpression level normalized to WT Day 4\nA\nFigure 5\nB\n0\n5\n10\n15\n20\n25\nWT KO WT KO WT KO\nC\nDay:          0                             4                               7  \nPparg\nLog2fold-change (KO vs WT)\nFPKM expression data \n****\n******\nD\n0\n2\n4\n6\n8\n10\n12\n0 2 4 6 8\nWT\nKO\nFPKM expression data \nDay\nHelz2\n****** ******\n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted March 27, 2025. ; https://doi.org/10.1101/2025.03.26.645456doi: bioRxiv preprint \n\nSupplementary Methods and Data \n \nSupplementary Methods \n \nAnimal model  \nMice (8-12 weeks old C57/B6/J ) were purchased from Charles River UK (Margate, \nUK), while Alox15-/- mice were bred in house (F11, C57BL/6J) in isolators. All animal \nexperiments were performed in accordance with the United Kingdom Home Office \nAnimals (Scientific Procedures) Act of 1986, under License (PPL 30/3334). Male and \nfemale mice were used for all studies except RNASeq where only males were used to \nreduce variation. Mice were housed in scantainers on a 12-hour light/dark cycle at 20 \n- 22 ºC, with free access to regular chow and water. Mice were anaesthetized using 3 \n- 3.5 % isoflurane delivered in 2  L per min 100  % oxygen . Once areflexic , mice \nreceived a sub -cutaneous dorsal injection of 10 μl Temgesic/Buprenorphine (1  μg). \nThey were shaved and re -tested to confirm the areflexic state. Spinal midline was \ndrawn before rotating to one side. Skin was folded using the spinal midline and two \npunch biopsy needles (BD pharma, UK) used to create 4 wounds (2 wounds per 4 mm \nbiopsy needle) (1). Wounds were trimmed using clean scissors , and wounds \nphotographed for size . Mice were transferred to a warming box until regaining \nconsciousness, before transfer into cage s lined with paper towel s. Mice were \nmonitored at 1 - 3 hours post wounding , and before the end of the light cycle. At 24 \nhours, mice were transferred into their original cages. In some experiments, mixtures \nof lipids were applied to wounds to determine their impact on healing . Two \npreparations were used, either “high -abundance” or “oxPL”. Here, immediately post-\nwounding, lipids were added (as described in Supplementary Table 1) in a final volume \nof either 50 µl (days 0,2) or 25 µl (days 4,6), with amount of lipids consistent across \nall days. Lipids were added in ethanol:Tween80:sterile water (1:1:18) as vehicle(2-4). \nOn days 2, 4, 6, mice were briefly anesthetized as above to enable lipid application . \nVehicle or lipids were added topically to the wound and allowed to sit in the wound \nsurface for 15 min, before mice were allowed to recover consciousness. On various \ndays post wounding,  mice were euthanized and wounds collected (typical weights \nfrom 5-10 mg per wound) and either processed for histology as outlined below using \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted March 27, 2025. ; https://doi.org/10.1101/2025.03.26.645456doi: bioRxiv preprint \n\nparaformaldehyde fixation or snap frozen in liquid nitrogen and then stored at -80 0C. \nFor lipid supplementation studies, mice were euthanized at day 7. \n \nGeneration of histological tissue sections   \nAt various time points up to 14 days, mice were killed using CO 2 (Schedule 1). The \nwound site and a small area of surrounding tissue was dissected and placed in 4 % \nparaformaldehyde for 72 hours, then 70% ethanol (to prevent further crosslinking) , \nthen the tissue processed for paraffin wax embedding. Wounds were placed into \nplastic cassettes and in a Leica tissue processor using the following parameters : \nethanol (60 %) under ambient temperature under vacuum for 90 min with agitation. \nEthanol (70 %) under ambient temperature under vacuum for 90 min with agitation. \nSix separate steps of ethanol (100 %) under ambient temperature under vacuum for \n60 min with agitation. Xylene (100 %) at 37 °C under vacuum for 120 min with agitation. \nXylene (100 %) at 45 °C under vacuum for 120 min with agitation. Two separate steps \nof wax (100 %) at 60 °C under vacuum for 120 min with agitation. Wax (100 %) at 60 \n°C under vacuum for 60 min with agitation.  Wax (100 %) at 60 °C under vacuum for \n45 min with agitation. Post-processing, plastic cassettes were removed, and wounds \nwere cast side on in molten (60 °C) wax, which was allowed to harden to form tissue \nwax blocks. Wax blocks were fastened to a microtome stage and secured. 10  μm \nslices were taken from each revolution and transferred to a 40  oC water bath  and \nallowed to unfold and float for 30 - 60 seconds before being placed onto a glass slide. \nSections were left to dry at room temperature, then at 55 oC overnight.  \n \nDAB Immunohistochemistry (F480, LY6G) \nSlides were placed in Histoclear (Fisher Scientific) for 3 min at room temperature then \ninto ethanol at 100 %, 90 % and 70 % for 3 min each before being placed into running \ntap water for 5 min. Next, slides were dried with a paper towel, and a hydrophobic pen \nused to draw borders around each section. Antigen retrieval was performed by \naddition of 5μg/ml proteinase K in PBS for 7 mins at R.T.  Slides were then rinsed in \nrunning tap water and  H 2O2 (3 %) was added and the sections incubated for 10 min  \nat room temperature. Slides were washed in running tap water. Block solution (500 µl \nPBS + 1 % bovine serum albumin + 0.1 % fish gelatin + 0.5% Triton X-100) and avidin \nblock (Avidin/Biotin block kit vector labs)  were added and sections incubated for 30 \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted March 27, 2025. ; https://doi.org/10.1101/2025.03.26.645456doi: bioRxiv preprint \n\nmin at room temperature. Block solution was removed, and slides washed twice with \nPBS. Primary antibody ( Supplementary Table 2) or isotype controls made up in \nantibody solution (50 ml PBS + 1 % bovine serum albumin + 0.1% fish gelatin + 0.5% \nTriton X-100 and biotin block solution)  was then added F4/80 Ab at 1/400 and Ly6G \nat 1/200. Slides with antibody solution were left for an extended period (14 hours or \novernight) then washed with PBS 0.1% Tween 2 times for 5 mins. Biotin conjugated \nsecondary antibody solutions were made up at 1:500 in block solution and added after \nthe washes for 30mins. Secondary antibody was removed and sections washed 2 \ntimes as done before.  Avidin -Biotin Complex (ABC) kit solution (Vector Labs) was \nthen made up (as per manufacturers’ instructions) 30 min before use, this was added \nfor 30 min before being removed with one 5 minute wash.  Sections were treated with \na 3,3'Diaminobenzidine (DAB) solution (Vector Labs) for 1 -2 min depending on \nstaining intensity. The sections were washed in running tap water for 5 mins, then \nplaced in hematoxylin for 15 secs. Sections were washed for 5 min in running tap \nwater.  Slides were then dipped 3 times in acid alcohol and then washed in running \ntap water for 1 min.  Finally slides were placed in Scotts Tap water for 18 seconds, \nfollowed by a final wash in running tap water for 5 mins.  Slides were  then placed into \nincreasing concentrations of ethanol (70 %, 90 %, 100 %) for 30 secs each, then \nplaced into Histoclear for 2x 30 sec incubations. Sectio ns were removed from \nHistoclear and dried carefully using paper towels. Slides were treated with 2 drops (50 \nμl) of distyrene and xylene (DPX solution) and covered with a glass cover slip and put \nin an oven at 60°C overnight.  \n \nDAB Immunohistochemistry (all other antigens). \nSlides were placed in Histoclear (Fisher Scientific) for 2 min at room temperature then \ninto ethanol at 100 %, 90 % and 70 % for 5 min each before being placed into DDH2O \nfor 5 min. Slides were then placed in citrate buffer (10 mM sodium citrate, 0.05 % \nTween 20, pH 6.0)  and incubated at 96 oC for 1 hr . They were cooled to room \ntemperature over 30 min, then placed back into DDH2O for 30 min. Next, slides were \ndried with a paper towel , and a h ydrophobic pen used to draw borders around each \nsection. Phosphate buffered saline (PBS, 1 ml) was added to sections for 2 min before \nremoval using a Pasteur pipette . H2O2 (3 %) was added and the sections incubated \nfor 10 min at room temperature. Slides were washed twice using PBS. Block solution \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted March 27, 2025. ; https://doi.org/10.1101/2025.03.26.645456doi: bioRxiv preprint \n\n(500 µl PBS + 1  % bovine serum albumin + 0.1  % fish gelatin) and avidin block  \n(Avidin/Biotin block kit vector labs)  was added and sections incubated for 45 min at \nroom temperature. Block solution was removed , and slides washed twice with PBS . \nPrimary antibody (Supplementary Table 2) or isotype controls made up in antibody \nsolution (50 ml PBS + 1 % bovine serum albumin + 0.1% fish gelatin and biotin block \nsolution) was then added. Slides with antibody solution were left for an extended \nperiod (14 hours or overnight) then washed with PBS 5 times with 10 -minute section \nsubmersion in bet ween washes. Secondary (HRP conjugated) antibody solutions \nwere made up at desired concentration (Supplementary Table 2) in block solution and \nadded after the washes for two hours. Secondary antibody was removed and sections \nwashed 5 times as done before with 10 -minute gaps between wash steps, Avidin -\nBiotin Complex (ABC) kit solution (Vector Labs) was then made up (as per \nmanufacturers’ instructions) 15 min before use, this was added for 45 min before being \nremoved with 5 washes. Sections were treated with a  3,3'Diaminobenzidine (DAB) \nsolution (Vector Labs) for 5 - 7 min depending on staining intensity. The sections were \nwashed multiple times with DDH 2O, then placed in hematoxylin for 2.5 minutes. \nSections were washed for 5 min in DDH2O. then placed into increasing concentrations \nof ethanol (70 %, 90 %, 100 %) for 2 min each, then placed into xylene for 20 sec. \nSections were removed from xylene and dried carefully using paper towels. Slides \nwere treated with 2 drops (50 μl) of distyrene and xylene (DPX solution) and covered \nwith a glass cover slip and stored 24 hrs to dry.  \n  \nFluorescence Immunohistochemistry. \nSlides were placed in Histoclear (Fisher Scientific) for 2 min at room temperature then \ninto ethanol at 100 %, 90 % and 70 % made up in DDH20 for 5 min each before being \nplaced into DDH 2O for 5 min. Slides were placed in citrate buffer (10  mM sodium \ncitrate, 0.05 % Tween 20, pH 6.0) and incubated at 96  oC for 1 hr. They were cooled \nto room temperature during 30 min, then placed into DDH 2O for 30 min. Next, slides \nwere dried with a paper towel, and a hydrophobic pen used to draw borders around \neach section. PBS (1 ml) was added to sections  for 2 min before removal using a \nPasteur pipette. Block solution (500 µl PBS + 1 % bovine serum albumin + 0.1 % fish \ngelatin) was added and sections incubated for 45 min at room temperature. Block \nsolution was removed and primary antibody ( Supplementary Table 2 ) or isotype \ncontrols made up in antibody solution (50 ml PBS + 1  % bovine serum albumin + 0.1 \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted March 27, 2025. ; https://doi.org/10.1101/2025.03.26.645456doi: bioRxiv preprint \n\n% fish gelatin) were then added. Slides with antibody solution were left for an extended \nperiod (14 hours or overnight). The slides were then washed with PBS 5 times with \n10-minute section submersion in between washes. Secondary antibody solutions were \nmade up at 1:300 in block solution and add ed after the washes and left for 2 hrs at \nroom temperature. Secondary antibody was removed and sections washed 5 times as \ndone before with 10-minute gaps between wash steps. Sections were counterstained \nwith DAPI solu tion (Invitrogen, 10ng/ml) and True Black ( Cambridge BioSciences, \nCambridge UK) as per manufacturer’s instructions. Sections were washed 5 times in \nDDH20 and then allowed to dry overnight and then coverslipped with  ProLong™ \nDiamond Antifade mounting medium (Invitrogen). \n \nCollagen staining using Masson trichrome \nCollagenous fibres were visualized using Masson trichrome stain (Abcam, ab150686). \nParaffin embedded sections were re-hydrated in decreasing concentration of ethanol \n(100 %, 96  %, 70  % in DDH 2O, t hen manufacturer’s instructions followed : placing \nsections in Bouin's fluid, Weigert's iron hematoxylin solution, Biebrich scarlet/acid \nfuchsin solution, phosphomolybdic/phosphotungstic acid solution, aniline blue solution \nand acetic acid solution . Sections were dehydrated in increasing concentrations of \nethanol (70 %, 96 %, 100 %), before being rinsed in xylene, dried and coverslipped \nwith DPX mounting media. Sections were visualized under a light microscope  Leica \nDM 2000 with a Leica DMC 2900 camera. \n \nImage acquisition and analysis  \nImages were acquired with ether a light microscope  (as detailed above)  or \nepifluorescent microscope  (EVOS M5000) . Images were transferred to ImageJ \nsoftware for post-acquisition analysis.  \n(i) For DAB imag es, contrast was enhanced by 5  % before isolation of the brown \n(DAB) staining channel after a color devolution set to ‘H DDAB’ (hematoxylin and \nDAB staining) . Brown (dab staining) channels were then inverted and set to \n‘rainbow smooth’ allowing localization and intensity of the DAB staining to be \ndetermined. Pixel count of the image was determined using a live histogram with \ndata only from 70-255 (grey scale) to avoid background (hematoxylin) staining.  \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted March 27, 2025. ; https://doi.org/10.1101/2025.03.26.645456doi: bioRxiv preprint \n\n(ii) For fluorescent imaging pixel count or mean grey intensity  (for Texas red or CY5 \nchannels) was used. Isotype controls used are provided in Supplementary Figure \n9. \n \nRNASeq   \nWound tissue dissected from 2 mice (8 wounds in total , 4 wounds per mouse ) to \ngenerate one sample were snap-frozen in liquid N2 before being stored at -80 0C. For \nWT and Alox15-/- a total of four samples were generated at each timepoint . Wounds \nwere placed frozen into a pre-cooled pestle and mortar, then ground into a fine powder \nwith liquid N2 added. Tri reagent ( 1 ml, Sigma-Aldrich, UK) was added . The sample \nwas then transferred to RNAse -free tubes and bromochloropropane ( 200 µl, Sigma-\nAldrich, UK) added before vortexing. Samples were placed on ice for 5 min, before \nbeing centrifuged at 15,000 g for 15 min at 4 oC. The upper aqueous phase was \ntransferred to a new tube and 250  µl of 3 M sodium acetate (pH 5.5), 700 µl (100  %) \npropanol and 10 µl glycogen added, before incubation at -80 0C overnight. Samples \nwere centrifuged at 15,000 g for 15 min at 4 0 C to pellet the RNA . The pellet was \nwashed 3 times using 70 % ethanol, then allowed to air dry for 5 min . RNAse-free \nwater (50 µl) was added. Samples were cleaned using an RNeasy MinElute Cleanup \nKit (Catalogue number 74204 Qiagen, MD, USA), using a column based clean up step. \nSample (1 µl)  was analyzed using a NanoDrop ™ 2000/2000c Spectrophotometer \n(ThermoFisher Scientific, Newport UK), and to ensure samples were free of \ncontamination. All samples had absorbance 260/230 ratios of between  1.7 and 2.0, \nand 260/280 values between 1.8 to 2.1. Each RNA sample (5ul) was used to determine \nRNA integrity analysis (RIN). Total RNA quality was assessed using the Agilent 4200 \nTapeStation with RNA ScreenTape® (Agilent Technologies) and quantity with the \nInvitrogen™ Qubit-iT™ RNA HS Assay Kit (Fisher Scientific) according to the \nmanufacturer’s instructions. Libraries were prepared from 300ng of total RNA with a \nRIN value > 7. Total RNA was depleted of ribosomal RNA and sequencing libraries \nprepared with the Illumina®TruSeq Stranded Total RNA Library Prep Gold (Illumina, \nInc) kit using TruSeq CD Index Adapters 1 (Illumina, Inc) . The steps included the \ndepletion of cytoplasmic and mitochondrial rRNA, depleted RNA fragmentation, 1st \nstrand cDNA synthesis, 2nd strand cDNA synthesis, adenylation of 3’ ends, adapter \nligation, DNA fragment enrichment by PCR amplification (10 -cycles) and validation. \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted March 27, 2025. ; https://doi.org/10.1101/2025.03.26.645456doi: bioRxiv preprint \n\nThe manufacturer’s instructions were followed except for the cleanup after the ribozero \ndepletion step where Ampure®XP beads (Beckman Coulter) and 80% Ethanol were \nused. The libraries were validated using the Agilent 4200 TapeStation® with high -\nsensitivity D1000 ScreenTape® (Agilent Technologies) to ascertain the insert size, \nand the Invitrogen™ Qubit™ dsDNA HS Assay Kit (Fisher Scientific) used to perform \nthe fluorometric quantitation. Following validation, the libraries were normalized to 4  \nnM, pooled tog ether and clustered on the cBot ™2 (Illumina, Inc) following the \nmanufacturer’s recommendations. The pool was then sequenced using a 75 -base \npaired-end (2x75bp PE) dual index read format on the HiSeq4000 (Illumina, Inc) \naccording to the manufacturer’s instructions. These combinatorial dua l (CD) index \nadapters were formerly called TruSeq HT.  \n \nLipid extraction   \nAt time points < 14 days, mice were killed using CO 2 (Schedule 1). The wound site \nand a small area of surrounding tissue was dissected . Wounds were placed into 0. 3 \nml buffer containing 100 µM diethylenetriamine pentaacetate. Ceramic beads (15  – \n20, 2.8 mm) were added (approx. 10 /sample) and tubes placed into a bead \nhomogenizer (Bead Ruptor Elite v1.1, Omni International). Wounds were \nhomogenized using two 15 sec cycles (with a 30 sec delay) at 7.1  ms-1, 4 oC. The \ntissue and beads were transferred into a glass extraction vial containing 1.25 ml \nhexane/isopropanol/glacial acetic acid (30:20:2). Tubes were rinsed with an additional \n0.3 ml of buffer, then were vortexed and this was then added to the extraction vials . \nInternal standards (5 ng each of PC  14:0_14:0 and PE 14:0_14:0) and 5 ul of \neicosanoid internal standards was added to each sample  to give final concentration \n75 nM (Supplementary Table 3). Following vortexing (1 min), hexane (1.25 ml) was \nadded followed by vortexing, then centrifugation for 5 min at 1500 rpm, 4 °C. The upper \norganic phase was recovered into a clean tube. Hexane (1.25 ml) was added to the \nlower phase, which was again vortexed and centrifuged as above . The upper layer \nwas combined with the previous hexane extract. A mixture of chloroform:methanol (1.9 \nml, ratio 1:2) was added to the remaining aqueous phase. Following vortex, chloroform \n(0.625 ml) was added and samples again vortexed. HPLC grade water (0.625 ml) was \nadded, and samples vortexed. Last, samples were centrifuged as above. The lower \nphase was carefully harvested and added to the hexane extracts obtained above. \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted March 27, 2025. ; https://doi.org/10.1101/2025.03.26.645456doi: bioRxiv preprint \n\nSamples were dried using a RapidVap (RapidVap, Labconco®) then resolubilized in \nmethanol (200 µl), with careful vortexing. The samples were split in half (2 x  100 µl \nsamples, sample A and sample B), with half being stored for eoxPL analysis, and the \nremainder analyzed for oxylipins (see below) . Samples were stored at -80 ºC until \nLC/MS/MS as described later.  \n \nOxylipin extraction   \n100 µl methanol was added t o sample B (listed above) followed by 1.655 ml HPLC -\ngrade water. Glacial acetic acid (45 µl) was added to acidify and samples were mixed \ngently. Sep-Pak ( C18 Waters) columns were loaded into a positive pressure (N 2) \nmanifold, then pre-conditioned using 100 % MeOH (12 ml), followed by acidified water \n(6 ml with 0.4 % glacial acetic acid ). Samples were loaded onto the preconditioned \ncolumns and allowed to drip through using gravity. Acidified water (10 ml) was passed \nthrough the column followed by 6 ml hexane (under nitrogen pressure). Columns were \nallowed to dry for 30 min. Oxylipins were eluted from the column using methyl formate \n(8 ml) into glass extraction tubes then solvent was evaporated under vacuum. Lipids \nwere reconstituted using methanol (100 µl) and stored at -80 °C until LC/MS/MS as \ndescribed below. \n \nLC/MS/MS analysis of oxylipins   \nLipids were quantified using reverse phase LC/MS/MS. They were separated using a \ngradient of 30 – 100 % B over 20 min (A: water:mobile phase B 95:5 + 0.1% acetic \nacid, B: acetonitrile:methanol, 80:15 + 0.1% acetic acid) on an Eclipse Plus C18 \nColumn (Agilent), and analyzed on a Sciex QTRAP® 6500(5). Source conditions: TEM \n475 °C, IS -4500, GS1 60, GS2 60, CUR 35. Chromatographic peaks were integrated \nusing Multiquant 3.0.2 software (Sciex) . The criteria for LOQ was signal:noise of at \nleast 5:1 and with at least 5-6 points across a peak . Lipids were quantified using a \nstandard curve generated and run at the same time as the samples  and multiple \nreaction monitoring ( MRM) channels  and all assay parameters are  provided in \nSupplementary Table 3 and (5). Each oxylipin was expressed per mg of skin tissue. \nExample chromatograms are provided in Supplementary Figure 10. \n \nLC/MS/MS (reverse phase) analysis of oxidized phospholipids \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted March 27, 2025. ; https://doi.org/10.1101/2025.03.26.645456doi: bioRxiv preprint \n\nLipid extracts were separated using reverse-phase HPLC on a Luna 3 µm C18 150 × \n2 mm column (Phenomenex, Torrance, CA) with a gradient of 50 – 100 % B over 10 \nmin followed by 30 min at 100 % B (A : methanol:acetonitrile:water, 60:20:20 with 1 \nmM ammonium acetate,  B: methanol, 1 mM ammonium acetate) with a flow rate of \n200 µl min−1. Lipids were analyzed in MRM mode on a 6500 Q-Trap (Sciex, Cheshire, \nUnited Kingdom), monitoring transitions from the precursor to product ion (dwell 75 \nms) with TEM 500 °C, GS1 40, GS2 30, CUR 35, IS − 4500 V, DP − 50 V, EP − 10 V, \nCE − 38 V and CXP at − 11 V. The peak area was integrated and normalized to the \ninternal standard. For qua ntification of HETE -PEs, standard curves were generated \nwith PE 18:0a/5 -HETE, PE 18:0a/8 -HETE, PE 18:0a/11 -HETE, PE 18:0a/12 -HETE \nand PE 18:0a/15-HETE synthesized as described previously(6). Information on MRM \ntransitions and m/z values are presented in Supplementary Table 5. HETE-PEs were \nquantified using standard curves with DMPE used as internal standard , with LOQ at \nsignal:noise 5:1.  Due to the limited standards available, identifications for some lipids \nare putative, based on the presence of characteristic precursor and product ions, and \nretention times.  Example chromatograms are provided in Supplementary Figure 11. \n \nChiral analysis of oxylipins \nLipid extracts were separated using a Chiralpak IA -U column (50×3.0 mm, Diacel) in \nreverse phase mode, with flow rate 300 ml/min, at 40 °C, according to (7), on a 6500 \nQ-Trap (Sciex, Cheshire, United Kingdom).  Mobile phase A was water:0.1 % acetic \nacid, and B was acetonitrile:0.1% acetic acid, and the gradient was 10 %  B raised to \n100 % B over 20 min followed by a 2 min hold then decrease to starting conditions \nover 2 min.  MRM transitions and instrument parameters were as used for oxylipin \nreverse phase analysis.  \n \nGel zymography for MMP activity   \nOn day 7, wounds were harvested, snap frozen in liquid N2 and stored at -80 OC until \nprocessing. Tissue was homogenized using ceramic beads in a Bead Ruptor Elite v1.1 \n(Omni International) using two rounds at 8 m/sec for 15 sec with a 30 sec dwell time, \nin 0.3 ml ice cold lysis buffer with protease inhibitors (50 mm Tris-HCl, 150 mm NaCl, \n1 % Nonidet P -40, 0.1 % SDS, 0.1  % deoxycholic acid , 2 μg/ml leupeptin, 2 μg/ml \naprotinin, 1 mm PMSF pH 7.4). Samples were placed on a rotary carousel for 30 min \nat 4 oC. The homogenate was centrifuged at 15,000 × g for 5 min at 4°C. Protein was \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted March 27, 2025. ; https://doi.org/10.1101/2025.03.26.645456doi: bioRxiv preprint \n\nmeasured using the  Bradford assay (Thermo fisher) and samples diluted to 15 \nμg/sample. Samples were diluted in sample buffer (Zymogram Sample Buffer, \n#1610764, Biorad) and loaded into the wells of precast gels (Novex™ 10% Zymogram \nPlus (Gelatin) ) gels (Thermo Fisher)). Electrophoresis was performed with a Tris -\nglycine running buffer (LC-26754, Invitrogen), at 125 V for 140 min. The gel was \nincubated for 1 hr at room temperature in 3  % Triton X-100 on a rotary shaker , then \nincubated with development buffer (50 mm Tris base, 40 mm HCl, 200 mm  NaCl, 5 \nmm CaCl2, and 0.2 % Brij 35) at 37 °C for 18 – 30 hr (depending on experiment) on a \nrotary shaker. Gels were stained  using 0.5 % w/v Coomassie blue G -250 in 50 % \nDDH20, 40 % methanol and 10 % acetic acid for 2 hr, and then destained for 1 hr using \ndiluent (50 % DDH20, 30 % methanol, 10 % acetic acid).  Gelatinolytic activity was \nobserved as clear zones or bands at the appropriate molecular weights. Mouse MMP-\n9 and human MMP -2 (R &D Systems) were used to locate  bands. Bands were \nquantified using the gel analysis plugin on ImageJ. \n \nCell transfection and reporter assays.  \nHEK293 cells were cultured and transfected as described previously with plasmids \nexpressing mouse PPARγ and the Firefly luciferase under the control of 3x Ppar  \nResponsive Element (PPRE) (8). The Renilla luciferase plasmid pRL -TK (Promega) \nwas also included in the transfection as an internal control. At day 2, the cells \nunderwent 24 h incubation in 50 µl media, with 1 μM rosiglitazone or DMSO (vehicle), \nlipid mixtures or methanol (vehicle). For each experiment (day) there were 4 replicates \nper condition and the experiment was repeated 3 independent times. On day 3, lysates \nwere prepared, and a luciferase assay was performed using a Dual -Luciferase \nReporter Assay System (Promega).  \n \nStatistical Analysis   \nTo compare wounds using immunohistochemistry, Students T-test was used * p<0.05, \n** p<0.01, *** p<0.001 and **** p<0.0001 . For multi-time point analysis, data were \nanalyzed using one-way ANOVA, with p < 0.05 considered statistically significant. For \nRNASeq, paired-end reads from Illumina sequencing were trimmed with Trimmomatic \nand assessed for quality using FastQC with default parameters. Reads were mapped \nto the Mouse GRCm38 reference genome using STAR and counts were assigned to \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted March 27, 2025. ; https://doi.org/10.1101/2025.03.26.645456doi: bioRxiv preprint \n\ntranscripts using FeatureCounts with the GRCm38.84 Ensembl gene build GTF. Both \nthe reference genome and GTF were downloaded from the Ensembl FTP site (9-14). \nDifferential gene expression analyses used the DESeq2 package (14) to produce an \nexcel output listing adjusted p -value and log2 fold change between conditions . The \ndata were then filtered so that only gene s with adjusted p -value < 0.05 were taken \nforward for analysis (Benjamini -Hochberg adjustment). Downstream pathway \nanalyses and gene annotation were performed in ingenuity IPA (Qiagen IPA). \nCytoscape was used to cluster genes by expression (FPKM) over all samples and \ntimepoints(15). Pearson correlation coefficients were calculated for all possible gene \npairs, and only highly significant genes retained (|r| > 0.8).  During analysis, one wild-\ntype (baseline) RNASeq sample failed a quality control check (Alox15 expression data \nindicated it was a knockout)  and was removed from further analysis. For temporal \nanalysis of gene expression changes, lists of genes were generated using \nMATLAB_R2022a. A heatmap was generated using Clustergram in MATLAB, where \ndata was standardized for each gene, so that the mean is 0 and the standard deviation \nis 1, and hierarchical clustering for rows performed. Data were analyzed using \nIngenuity Pathway Analysis . For oxylipidomics, a two -way ANOVA was used to \ncalculate differences within groups, (p < 0.05 considered significant) , with Bonferroni \npost hoc test (https://statisty.app/two -way-anova-calculator). Heatmaps for oxylipins \nand oxPL were generated using an R script that processes the raw dataset, into log10 \nvalues, based on an average of 5 biological replicates, which then passes the resulting \ndata to pheatmap (https://cran.r-project.org/web/packages/Pheatmap) to produce the \nfinal image.  \n \nSupplementary Results. \n \nStructural analysis of resolvinD5.  \nWhile most SPM were not detected, a  peak co -eluting with the resolvinD5 (RvD5 , \n7S,17S-diHDOHE) standard was seen (Supplementary Figure 12). RvD5 represents \none stereoisomer of 4 possible 7,17 -diHDOHEs.  The lipid is described to originate \nfrom 12/15 -LOX dependent formation of 17 -HDOHE, followed by its further \noxygenation by 5 -LOX, following transcellular uptake of 17 -HDOHE into \nleukocytes(16). However reverse phase LC/MS/MS is unable to fully separate these \nisomers, and it was not possible to obtain an MS/MS spectrum to compare with the \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted March 27, 2025. ; https://doi.org/10.1101/2025.03.26.645456doi: bioRxiv preprint \n\nstandard, due to the low levels of the lipid present in wounds. To address this, \nsecondary MRMs were next analyzed, with two arising from fragmentation at C7 (m/z \n141, 199) and one at C17 (m/z 261, Supplementary Figure 12 A-D). For the synthetic \nRvD5, the three MRMs co-eluted at 10.06 min as expected (Supplementary Figure 12 \nC).  In day 1 wound extracts, a lipid was detected at 10.06 min showing co-eluting ions \nfor m/z 199 and 141, however the third MRM (m/z 261) eluted slightly earlier \n(Supplementary Figure 12 B,D). Both the putative RvD5, and other later eluting lipids \nthat were detected using these MRMs were absent from Alox15-/- wound extracts \n(Supplementary Figure 12 E).  Next, chiral analysis was undertaken, with synthetic \nRvD5 eluting at 7.23 min, with the expected MRMs also co -eluting (Supplementary \nFigure 12 F).  Chiral analysis of the wound lipid extract showed several peaks eluting \nbetween 6.8 – 8 min (Supplementary Figure 12 G).  Based on ion ratios, the large \npeak at 7.89 min is likely to be the same as the two seen around 11 min on reverse \nphase LC/MS/MS (Supplementary Figure 12 A,G). A very small peak at 7.22 min had \nthe same retention time as RvD5 standard (Supplementary Figure 12 F,G) and similar \nion ratios with the m/z 199 ion dominating.  A recent study using a similar chiral \nseparation method monitored RvD5 and its isomers after pre-isolating 7,17-diHDOHE, \nusing the MRM m/z 359 -141, and showed that RvD5 elutes slightly later th an its \nisomers, 7 R,17S, 7 R,17R and 7 S,17R-diHDOHEs(17). In mouse wounds earlier \neluting peaks that could represent these isomers are seen. These have the same \nMRMs as RvD5, in particular the peak at 6.84 min (Supplementary Figure 12 G, inset) \nindicating oxygenation at C7 and C17. Several other lipids eluted just after the putative \nRvD5 (7.3-7.6 min), and these may represent additional related structures such as \npositional isomers (Supplementary Figure 12 G).  Spiking the wound lipid extract with \nsynthetic standard showed co -elution of RvD5 with the peak at 7.2 min \n(Supplementary Figure 12 H,I).  Overall, the data suggest the wound may contain low \nlevels of RvD5, together with other related isomers of 7,17-diHDOHE.  As for reverse \nphase analysis, the putative RvD5 peak along with all the other lipids detected using \nchiral analysis were absent in Alox15-/- wounds indicating their dependence on the \nenzyme (Supplementary Figure 12 J). However, 7,17-diHDOHE (coeluting with RvD5) \nwas present at ~0.5 % the levels of 17-HDOHE. Considering this, and the presence of \nisomers, it is possible RvD5 may have originated from non -enzymatic secondary \noxidation of 12/15-LOX-derived 17S-HDOHE to form 7R,17S-diHDOHE and 7S,17S-\n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted March 27, 2025. ; https://doi.org/10.1101/2025.03.26.645456doi: bioRxiv preprint \n\ndiHDOHE(RvD5) eluting at 6.84 and 7.22 min, respectively.  Further studies are \nneeded to establish the origin of the lipid in wounds , for example \npharmacological/genetic inhibition of 5 -LOX, and MS/MS analysis of the 7 R,17S-\ndiHDOHE epimer under our chromatographic conditions.  \n \nComparison of temporal changes in gene expression suggest additional transcription \nactivators regulated by Alox15 beyond PPARg include elf4, Cebpb and Tcf3.  \nTo further interrogate Alox15-/- wounds for transcriptional regulators beyond PPARg, a \ntemporal analysis was performed  on the RNASeq data . Here, analysis of individual \nstrains separately allowed testing for genes behaving differently during progression of \nwound healing. In WT mice, 1705 transcripts significantly increased > 50 % on Days \n0 and 4, while by Day 7, they reduced by > 25 % compared to Day 4 (Supplementary \nFigure 13 A, List 1). Thus, these elevate on acute injury, then return close to normal \nafter one week (Supplementary Table 9). Interrogating these in Alox15-/- mice, 154 did \nnot increase on Day 4 by > 25 % compared to Day 0 (Supplementary Figure 13 B, \nSupplementary Table 10). Thus, these failed to elevate during acute inflammation in \nAlox15-/-. Using IPA analysis, several transcription factors were identified as possible \nupstream regulators. Elf4 is a known anti -inflammatory transcription regulator of \ninflammation, which targets several genes in the list, including Anln, Asf1b, Ccnb2, \nCdca3, Cenpa, Cenpe, Cks2, E2f8, Hmmr, Kif4a, Mcm10, Ndc80, Oip5, Rrm2, \nTpx2(18). In support of this idea, we found that expression of Elf4 was significantly \nincreased by wounding in both WT and Alox15-/- mice (Supplementary Figure 13 D). \nThis indicates that while Elf4 is upregulated by wounding, it may not be \ntranscriptionally active in the absence of 12/15 -LOX. Additional  transcription \nregulators strongly associated with the initial response to wounding included Tcf3 and \nCebpb. Tcf3 promotes cell migration and wound repair (19), and Cebpb is involved in \nmacrophage repair responses and inflammation (20, 21). Expression data for these \ngenes showed Cebpb is upregulated on Day 4, significantly in Alox15-/-, while reduced \nback to baseline expression by Day 7 .  However, Tcf3 expression was unaffected by \nwounding in either strain (Supplementary Figure 14 A,B).  \n \nLast, genes in List 1 were re -interrogated to identify transcripts that reduced < 25 % \non Day 7 compared to Day 4 in Alox15-/-. These represent genes that fail to resolve to \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted March 27, 2025. ; https://doi.org/10.1101/2025.03.26.645456doi: bioRxiv preprint \n\nbasal levels during inflammation in Alox15-/-. Here, 538 were identified (Supplementary \nFigure 14 C, Supplementary Table 11 ). IPA analysis of these proposed \n“lipopolysaccharide” as the top upstream regulator, consistent with the failure of many \nknown pro-inflammatory genes which respond to this bacterial product to reduce back \nto basal levels as shown in our earlier analysis, e.g., Il6, Ptgs2, and Il1b \n(Supplementary Figure 7). Similarly, IPA also proposed the top affected canonical \npathway for List 3 as “Pathogen Induced Cytokine Storm Signaling” which includes 33 \ngenes which failed to fully resolve. These are shown in a heatmap, comparing Days 4 \nor 7 with Day 0 in both WT and Alox15-/- mice. Three distinct groups are seen  \n(Supplementary Figure 14 C):  \n(i) Genes that elevate by Day 4 and then are reduced by Day 7 in WT. The y elevate \nto a similar level in Alox15-/- but do not fall back to baseline by Day 7 or elevate \nfurther by that time (Fos, Ddx58, Stat1, Ccr3, Nlrc4, Naip1, Faslg, Gsdmd, Clec7a, \nItb, Lif, Stat4) \n(ii) Genes that elevate by Day 4 then reduce by Day 7 in WT, while elevating higher in \nAlox15-/- at Day 4, and not falling back to baseline by Day 7 (Cxcl3, Aim2, Ccl4, \nCcl3l3, Cxcl10, Pgf, Nos2, Cxcl2, Tlr2, Mlkl, Sting1, Cxcr4, Csf2rb, Zbp1, Cklf)  \n(iii) Genes that elevate far higher in WT than Alox15-/- but reduce back by Day 7 in \nboth (Il1r1, Ccl7, Cxcl6, Col13a1, Ccr5, Irf7).  \nAnalysis using STRING 11.5 showed that group (i) genes are members of networks \nthat regulate cytokines, including IFN-I, IL-12, IL-21, IL-35, IL-20 family and IL-6 family \nsignaling networks. For groups (ii) and (iii) the main KEGG pathways were “cytosolic \nDNA-sensing” and “IL-17 signaling”, respectively. Notably this analysis confirms our \nearlier data which indicates that inflammatory signaling is strongly impacted by Alox15-\n/- deletion, while identifying a large number of novel targets for further study.  \n \n \n \nSupplementary Figure Legends \n \nSupplementary Figure 1. Alox15-/- wounds show reduced wound bed \nmacrophages, and increased smooth muscle actin , SSEA3 and Ki -67 \nexpression.  Panel A. Alox15 deletion leads to reduced macrophage influx. F480+ve \ncells were measured in wounds as described in Methods (n= 6–8/group, 4-9 fields per \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted March 27, 2025. ; https://doi.org/10.1101/2025.03.26.645456doi: bioRxiv preprint \n\nwound). Panel B. Representative images from Panel A. Panel C. Neutrophil cell \nnumbers (visualized with by Ly6g DAB positive staining) are similar in wildtype and \nAlox15-/- mice, (n = 7–8/group, 4 -9 fields per wound ). Panel D. The  myofibroblast \nmarker smooth muscle actin  alpha was increased in Alox15-/- wounds. Cells were \nstained with anti- alpha smooth muscle actin and visualized with DAB staining (n = 7-\n8/group). For panels B,H: unpaired t -test, * p < 0.05, *** p<0.00 5. Panel E. \nRepresentative images of smooth muscle actin . Panel F. SSEA3 and Ki -67 are \nelevated in SSEA3 and Ki -67 Alox15 -/- wounds. Expression was measured using \nimmunohistochemistry, followed by DAB visualization. n = 6/group (SSEA3), 10/group \n(Ki67). Panels G,H. Representative data from Panel F.  \n \nSupplementary Figure 2. Epithelial proliferation and terminal differentiation of \nkeratinocytes are not impacted in Alox15-/- either basally or post-wounding, and \nIL-6 is unaffected. Panel A. Cytokeratin 14 migration migrated from the wound edge \nin Alox15-/- wounds at day 4 is not impacted. Quantification of the migratory distance \nof highly proliferative non -differentiated cytokeratin 14 (green) and non -proliferative \nterminally differentiated cytokeratin 10 (pink) was quantified using fluorescence \nimmunohistochemistry. n = 4/group. Panel E. Representative images from Panel B. \nPanel C. Representative images showing cytokeratin 10 and 14 staining in non -\nwounded skin. All panels, unpaired Students t-test, * p<0.05, *** p, 0.005.  Panel D. \nAlox15-/- wounds show unchanged IL-6 expression . IL-6 was measured using \nfluorescence immunohistochemistry. n = 6/group. \n \nSupplementary Figure 3.  Heatmap and time course data for oxylipin levels \nduring wounding shows that Alox15-/- wounds generate lower levels of many \noxylipins. Panel A. A heatmap shows log10 of mean values for all lipids across all \ngroups tested. Oxylipins were measured using LC/MS/MS as outlined in Method s (n \n= 5 samples/time point, with 4 wounds pooled/sample ). Panel. Oxylipins are rapidly \nelevated post-wounding, but many are reduced in Alox15-/- wounds. Oxylipins were \nmeasured using LC/MS/MS as outlined in Methods.  n = 6 samples/time point, with 4 \nwounds pooled/sample . For all panels differences between groups were analyzed \nusing two-way Anova (red stars), with Bonferroni post hoc test between individual time \npoints (black stars), mean ± SEM, * p < 0.05, ** p<0.01, *** p < 0.005. \n \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted March 27, 2025. ; https://doi.org/10.1101/2025.03.26.645456doi: bioRxiv preprint \n\n \nSupplementary Figure 4. Time course data for oxylipins generated during \nwounding, as in Supplementary Figure 2 . Oxylipins were measured using \nLC/MS/MS as outlined in Methods.  n = 5 samples/time point, with 4 wounds \npooled/sample. For all panels differences between groups were analyzed using two -\nway Anova (red stars), with Bonferroni post hoc test between individual time points \n(black stars), mean ± SEM, * p < 0.05, ** p<0.01, *** p < 0.005. \n \nSupplementary Figure 5. Heatmap of eoxPL generation during wounding, and \nindividual timecourses of 12 -HETE-PEs. Panel A.  Heatmap shows expression \n(log10) of mean values for all lipids across all groups tested eoxPL were measured \nusing LC/MS/MS as outlined in Methods.  n = 5 samples/time point, with 4 wounds \npooled/sample. Panel B. 12-HETE-PE isomers are significantly reduced in Alox15-/- \nwounds. Oxidized phospholipids were measured using LC/MS/MS as outlined in \nMethods (n = 5 samples/time point, with 4 wounds pooled/sample) . Unpaired t-test, * \np <0.05, ** p < 0.01, *** p < 0.005. \n \nSupplementary Figure 6. Raw data for MMP quantitation in wounds Panels A,B. \nGel zymography showing data used to quantify MMP activities during wounding.  \n \nSupplementary Figure 7. Genes in the network show the same pattern of \nexpression throughout the time course, increasing on Day 4, but with a failure \nto revert to baseline at Day 7 in Alox15-/- wounds. Data for several affected genes \nare shown, all gene expression data was normalized to its Day 0 mean value, and \nthen expressed as fold -change (n = 3 – 4 per group) . For all gene expression data, \nstudents t-test, followed by Benjamin Hochberg correction: * p <0.05, ** p < 0.01, *** \np < 0.005. \n \nSupplementary Figure 8.  High oxylipins activate PPARg transcription. Oxylipins, \nRosiglitazone or vehicle controls were added to HEK293 cells expressing mouse \nPPRE, as described in Methods.  After 24 hrs, luciferase activity was analyzed. Data \nare shown normalized to the relevant vehicle control (n = 3 independent experiments, \nmean +/- SEM, students t-test, * p <0.05, *** p < 0.005.  \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted March 27, 2025. ; https://doi.org/10.1101/2025.03.26.645456doi: bioRxiv preprint \n\n \nSupplementary Figure 9. Isotype and no -primary controls for antibodies in the \nstudy. Left panels:  To confirm target-specific DAB staining both a no primary control \nand isotype control were run for the antibodies F4/80, INF γ and CD206 followed by \nhematoxylin counter staining. Images were taken at 20x magnification, scale bar \n=100μm. Right panels: To confirm target specific immunofluorescence staining both a \nno-primary control and isotype control was run for the antibodies, Cytokeratin 10 \n(K10), cytokeratin 14 (K14), IL6, Phospho SMAD3, Phospho STAT3, followed by a \nDAPI stain. Images were taken at 10x magnification, scale bar =200μm.   \n \nSupplementary Figure 10. Representative chromatographic peaks for oxylipins \ndetected during wounding. Screenshots were taken from Multiquant software, with \nthe shaded area indicat ing the peak integrated. Wound lipids were confirmed to co -\nelute with primary standards in the same analytical batch. \n \nSupplementary Figure 11. Representative chromatographic peaks for eoxPL \ngenerated during wounding . Identity was verified comparing retention time  with \nstandards as outlined in (6), based on comparison with PE 18:0a_HETE for the \nrelevant positional isomers. Note that standards for 16:0p, 18:0p and 18:1 forms are \nnot available , and so relative RT compared to standards is used along with MRM \ntransitions which use internal daughter ions for all HETE positional isomers, along with \nLOQ of > 5 for signal:noise for peaks . The order of elution is characteristic for the \ndifferent sn1 forms, as shown in (22-24). \n \nSupplementary Figure 12. Reverse phase and chiral phase analysis of 7,17 -\ndiHDOHE suggests RvD5 along with additional isomers are present in WT \nmouse wounds at day 1. Lipid extracts from Day 1 wounds were pooled and analyzed \nusing reverse and chiral phase LC/MS/MS as described in Methods. For reverse \nphase, the same method was used as for the oxylipin assay, but focusing on MRM \ntransitions for RvD5, and removing scheduling.  Panels A-E. Reverse phase analysis \nof the synthetic RvD5 standard along with wound lipid extract from WT and Alox15 -/- \nmice. Panels A,B,D show WT lipid extract, while Panel C shows the RvD5 standard.  \nPanels F-J. Chiral phase analysis of synthetic RvD5 standard along with wound lipid \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted March 27, 2025. ; https://doi.org/10.1101/2025.03.26.645456doi: bioRxiv preprint \n\nextract from WT and Alox15-/- mice. Panels G-J show wound extracts, while F shows \nthe RvD5 standard.  \n \nSupplementary Figure 13. IPA temporal analysis shows altered regulation of \nseveral inflammatory pathways and induction of Elf4. Panel A. List 1 represents \n1705 transcripts in the WT condition  whose expression on Day 4 is significantly \nincreased by > 50% compared to Day 0, i.e., with log2 -fold change at least log2 FC \n(1.5) and adjusted p-value < 0.05,  but where  log2FC values on Day 7 compared to \nDay 4 are reduced by > 25 %. Panel B. List 2 represents a subset of List 1 consisting \nof 154 transcripts, whose expression on Day 4 is not increased by > 25% compared \nto D0 in Alox15-/- wounds, i.e., with a maximum log2 -fold change log2FC (1.25). \nWilcoxon signed-rank test shows that the D4 data between the WT and Alox15-/- is \nsignificantly different . Panel C.  List 3 comprises 538 transcripts from List 1 with a \nminimum log2-fold change log2FC (1.5) and adjusted p-value < 0.05 on D4 in Alox15-\n/- wounds, whose log2FC values are reduced by < 25 % on Day 7 compared to Day 4 \nin Alox15-/- wounds. Wilcoxon signed -rank test shows that the D7 data between the \nWT and Alox15-/- conditions is significantly different . Panel D. Elf4 is induced on \nwounding.  Gene expression data was normalized to its Day 0 mean value, and then \nexpressed as fold-change (n = 3 – 4 per group). For all gene expression data, students \nt-test, followed by Benjamin Hochberg correction: * p <0.05, ** p < 0.01, *** p < 0.005. \n \nSupplementary Figure 14. Cebpb is upregulated during wounding, but not Tcf3, \nwhile IPA identifies groups of genes with common behaviour that don’t resolve \nfully post wounding. Panels A-C. Data from gene expression is shown for WT and \nAlox15-/- wounds during the time course (n = 3 – 4 per group).  For all gene expression \ndata, students t-test, followed by Benjamin Hochberg correction: * p <0.05, ** p < 0.01, \n*** p < 0.005.  Panel C. IPA analysis shows clusters of genes with similar behaviour, \nwhich don’t fully resolve.   Genes that do not fully resolve but behave in groups  are \nplotted in this heatmap. Plotted are log2fold change data for wounds of the same strain \ncomparing day 0 with either day 4 or day 7, within strain comparisons only.    \n \n \n \nREFERENCES \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted March 27, 2025. ; https://doi.org/10.1101/2025.03.26.645456doi: bioRxiv preprint \n\n \n1. A. V. Kostarnoy  et al. , Receptor Mincle promotes skin allergies and is capable of \nrecognizing cholesterol sulfate. Proceedings of the National Academy of Sciences of \nthe United States of America 114, E2758-E2765 (2017). \n2. C. A. Castro, J. B. Hogan, K. A. Benson, C. W. Shehata, M. R. Landauer, Behavioral \neffects of vehicles: DMSO, ethanol, Tween -20, Tween -80, and emulphor -620. \nPharmacol Biochem Behav 50, 521-526 (1995). \n3. J. L. Wiley, J. J. Burston, Sex differences in Delta(9)-tetrahydrocannabinol metabolism \nand in vivo pharmacology following acute and repeated dosing in adolescent rats. \nNeurosci Lett 576, 51-55 (2014). \n4. R. Elfiyani, A. Amalia, S. Pratama, Effect of Using the Combination of Tween 80 and \nEthanol on the Forming and Physical Stability of Microemulsion of Eucalyptus Oil as \nAntibacterial. Journal of Young Pharmacists 9, s1-s4 (2017). \n5. M. Misheva et al., Oxylipin metabolism is controlled by mitochondrial beta -oxidation \nduring bacterial inflammation. Nat Commun 13, 139 (2022). \n6. A. H. Morgan et al., Quantitative assays for esterified oxylipins generated by immune \ncells. Nat Protoc 5, 1919-1931 (2010). \n7. X. Fu, Z. Xu, M. Gawaz, M. Lämmerhofer, UHPLC -MS/MS method for chiral \nseparation of 3 -hydroxy fatty acids on amylose -based chiral stationary phase and its \napplication for the enantioselective analysis in plasma and platelets. Journal of \nPharmaceutical and Biomedical Analysis 223, 115151 (2023). \n8. S. Jitrapakdee et al., The peroxisome proliferator -activated receptor-gamma regulates \nmurine pyruvate carboxylase gene expression in vivo and in vitro. J Biol Chem  280, \n27466-27476 (2005). \n9. A. M. Bolger, M. Lohse, B. Usadel, Trimmomatic: a flexible trimmer for Illumina \nsequence data. Bioinformatics 30, 2114-2120 (2014). \n10. A. Dobin et al., STAR: ultrafast universal RNA-seq aligner. Bioinformatics 29, 15-21 \n(2013). \n11. Y. Liao, G. K. Smyth, W. Shi, featureCounts: an efficient general purpose program for \nassigning sequence reads to genomic features. Bioinformatics 30, 923-930 (2014). \n12. https://www.bioinformatics.babraham.ac.uk/projects/fastqc/ ( \n13. http://www.ensembl.org/info/data/ftp/index.html/ ( \n14. M. I. Love, W. Huber, S. Anders, Moderated estimation of fold change and dispersion \nfor RNA-seq data with DESeq2. Genome Biol 15, 550 (2014). \n15. P. Shannon  et al. , Cytoscape: a software environment for integrated models of \nbiomolecular interaction networks. Genome Res 13, 2498-2504 (2003). \n16. S. Hong, K. Gronert, P. R. Devchand, R. -L. Moussignac, C. N. Serhan, Novel \nDocosatrienes and 17<em>S</em>-Resolvins Generated from Docosahexaenoic Acid \nin Murine Brain, Human Blood, and Glial Cells: AUTACOIDS IN ANTI -\nINFLAMMATION *. Journal of Biological Chemistry 278, 14677-14687 (2003). \n17. K. R. Kampschulte N, Loewen A, Schebb NH, Deducing formation routes of oxylipins \nby quantitative multiple heart -cutting achiral -chiral 2D -LC-MS. ChemRxiv \n10.26434/chemrxiv-2024-tgv9t-v3 (2024). \n18. P. M. Tyler et al., Human autoinflammatory disease reveals ELF4 as a transcriptional \nregulator of inflammation. Nature Immunology 22, 1118-1126 (2021). \n19. Q. Miao et al., Tcf3 promotes cell migration and wound repair through regulation of \nlipocalin 2. Nature Communications 5, 4088 (2014). \n20. C. Y. Ko, W. C. Chang, J. M. Wang, Biological roles of CCAAT/Enhancer -binding \nprotein delta during inflammation. J Biomed Sci 22, 6 (2015). \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted March 27, 2025. ; https://doi.org/10.1101/2025.03.26.645456doi: bioRxiv preprint \n\n21. D. Ruffell et al., A CREB-C/EBPbeta cascade induces M2 macrophage -specific gene \nexpression and promotes muscle injury repair. Proc Natl Acad Sci U S A  106, 17475-\n17480 (2009). \n22. S. R. Clark  et al., Esterified eicosanoids are acutely generated by 5 -lipoxygenase in \nprimary human neutrophils and in human and murine infection. Blood 117, 2033-2043 \n(2011). \n23. B. H. Maskrey  et al. , Activated platelets and monocytes generate four \nhydroxyphosphatidylethanolamines via lipoxygenase. J Biol Chem 282, 20151-20163 \n(2007). \n24. C. P. Thomas  et al. , Phospholipid -esterified eicosanoids are generated in agonist -\nactivated human platelets and enhance tissue factor -dependent thrombin generation. J \nBiol Chem 285, 6891-6903 (2010). \n \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted March 27, 2025. ; https://doi.org/10.1101/2025.03.26.645456doi: bioRxiv preprint \n\n0\n200\n400\n600\n800\n1000\n0 1 4 7 14 0 1 4 7 14\nWild Type Alox15-/-\nF480+ cells (brown) \nWild Type Alox15-/-\nNumber of neutrophils\nWild Type (4 days post wounding) Alox15-/- (4 days post wounding)\nF480 (brown)\nC\nAlox15-/-\n***\nSmooth muscle actin\nD E\nDay 7\nBrown: Smooth muscle actin\n50 µm50 µm\n100 µm\n***\n*A B\nWildtype\n100 µm\n***\nepidermis\ndermis\nepidermis\ndermis\nWound bed dermis\nWound bed dermis\n0\n100\n200\n300\n400\n0 1 4 7 0 1 4 7\n0\n400000\n800000\n1200000\n1600000\nWildtype Alox15-/-\nBlue: collagen\nDeep red: epithelium\nPale red: non-collagen containing dermal cells\nBlack: nuclei\n0\n40000\n80000\n120000\n160000\nWT Alox15-/-\nscab\nscab\nWild Type Day 4\nAlox15-/- Day 4\nBrown: SSEA3\nDermal wound bed\nBrown: SSEA3 Brown: Ki-67\nBrown: Ki-67\n0\n50000\n100000\n150000\n200000\n250000\n300000\nWT Alox15-/-\nWound bed\n*\nSSEA3+ve pixelsKi-67 mean grey intensity at wound edge\nWound bed\nF H\nWild Type Day 4\nAlox15-/- Day 4\nG\n125  µM\n125  µM 100  µM\n100  µM\nDermal wound bed\nepidermis\nepidermis\nepiderm is\nepiderm is\nSupplementary Figure 1\n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted March 27, 2025. ; https://doi.org/10.1101/2025.03.26.645456doi: bioRxiv preprint \n\nBlue: nuclei\nGreen: cytokeratin 14\nPink: cytokeratin 10\nB\nWild Type Day 4\nDistance (m) from wound edge\nCytokeratin 10 Cytokeratin 14\nA\nAlox15-/- Day 4\nWound edge\nWound edge\nScab Scab\n400 µm\n400 µm\nC\nWild Type: basal\n150 µm\nAlox15-/-: basal\n150 µm\nDay 4\n0\n200\n400\n600\n800\n1000\n1200\nWT KO WT KO\nIL6-expression\nAlox15-/- (day 7)\nRed: IL6\nBlue: DAPI\nRed: IL6\nBlue: DAPI\nD\n0\n10\n20\n30\n40\n50\nWT Alox15-/-\n100 µm\n100 µm\nepidermis\ndermis\nepidermis\ndermis\nSupplementary Figure 2\n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted March 27, 2025. ; https://doi.org/10.1101/2025.03.26.645456doi: bioRxiv preprint \n\n5-HETE\n0 5 10 15\n0\n50\n100\n150\nWT\nALOX15-/-\nDays post wounding\npg/mg wound tissue\nLTB4\n0 5 10 15\n0\n5\n10\n15\n20\n25\nDays post wounding\npg/mg wound tissue\n5,6-diHETE\n0 5 10 15\n0\n2\n4\n6\n8\nDays post wounding\npg/mg wound tissue\nPGE2\n0 5 10 15\n0\n50\n100\n150\n200\n250\nDays post wounding\npg/mg wound tissue\nPGD2\n0 5 10 15\n0\n50\n100\n150\nDays post wounding\npg/mg wound tissue\n11-HETE\n0 5 10 15\n0\n20\n40\n60\nWT\nALOX15-/-\npg/mg wound tissue\nDays post wounding\n 11-HEPE\n0 5 10 15\n0.0\n0.2\n0.4\n0.6\n0.8\n1.0\nDays post wounding\npg/mg wound tissue\nTXB2\n0 5 10 15\n0\n10\n20\n30\n40\nDays post wounding\npg/mg wound tissue\n8-iso-PGE2\n0 5 10 15\n0\n10\n20\n30\n40\nDays post wounding\npg/mg wound tissue\nPGF2a\n0 5 10 15\n0\n10\n20\n30\n40\n50\nDays post wounding\npg/mg wound tissue\n*\n8-HDOHE\n0 5 10 15\n0\n2\n4\n6\n8\n10\nDays post wounding\npg/mg wound tissue\nSupplementary Figure 3\n0\n0. 5\n1\n1. 5\n2\n0 5 10 15\n7,17-diHDOHE\nWT\nAlox15-/-\n***\n** *\n12-oxoETE\n0 5 10 15\n0\n5\n10\n15\n20\n25\nWT\nALOX15-/-\nDays post wounding\npg/mg wound tissue\n9-oxoETE\n0 5 10 15\n0\n100\n200\n300\n400\nDays post wounding\npg/mg wound tissue\n13-oxoETE\n0 5 10 15\n0\n50\n100\n150\nDays post wounding\npg/mg wound tissue\n15-oxoETE\n0 5 10 15\n0\n5\n10\n15\nDays post wounding\npg/mg wound tissue\n*\n9-oxoODE 13-oxoODE\nAlox15-/-\n* *\n*\n*\n*\nB\nA\n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted March 27, 2025. ; https://doi.org/10.1101/2025.03.26.645456doi: bioRxiv preprint \n\n9-HODE\n0 5 10 15\n0\n100\n200\n300\n400\nWT\nALOX15-/-\nDays post wounding\npg/mg wound tissue\n13-HODE\n0 5 10 15\n0\n100\n200\n300\n400\n500\nDays post wounding\npg/mg wound tissue\n8-HETE\n0 5 10 15\n0\n10\n20\n30\n40\nDays post wounding\npg/mg wound tissue\n5,6-diHETrE\n0 5 10 15\n0\n2\n4\n6\n8\nWT\nALOX15-/-\nDays post wounding\npg/mg wound tissue\n8,9-DiHETrE\n0 5 10 15\n0\n1\n2\n3\nDays post wounding\npg/mg wound tissue\n 11,12-DiHETrE\n0 5 10 15\n0\n1\n2\n3\n4\nDays post wounding\npg/mg wound tissue\n14,15-DiHETrE\n0 5 10 15\n0\n1\n2\n3\nDays post wounding\npg/mg wound tissue\nLXB4\n0 5 10 15\n0\n5\n10\n15\n20\n25\nDays post wounding\npg/mg wound tissue\n5,6-EET\n0 5 10 15\n0\n10\n20\n30\n40\nWT\nALOX15-/-\nDays post wounding\npg/mg wound tissue\n7,8-EpDPA\n0 5 10 15\n0\n2\n4\n6\n8\nDays post wounding\npg/mg wound tissue\n13,14-EpDPA\n0 5 10 15\n0.0\n0.5\n1.0\n1.5\n2.0\n2.5\nDays post wounding\npg/mg wound tissue\n9,10-EpOME\n0 5 10 15\n0\n20\n40\n60\nDays post wounding\npg/mg wound tissue\n 12,13-EpOME\n0 5 10 15\n0\n50\n100\n150\nDays post wounding\npg/mg wound tissue\n*\n*\nLTB4\nSupplementary Figure 4\n*\n*\n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted March 27, 2025. ; https://doi.org/10.1101/2025.03.26.645456doi: bioRxiv preprint \n\n0\n1\n2\n3\n4\n5\n6\n7\n0 5 10 15\n0\n0.5\n1\n1.5\n2\n2.5\n3\n0 10 20\n0\n0.5\n1\n1.5\n2\n2.5\n0 5 10 15\nWT\nKO\nPE 16:0p_12-HETE PE 18:0p_12-HETEPE 18:0a_12-HETE\nDays post woundingDays post woundingDays post wounding\npg/mg  per wound \n*\n***\n***\n***\nSupplementary Figure 5\nA\nWT D0\nKO D0\nWT D1\nKO D1\nWT D4\nKO D4\nWT D7 \nKO D7\nWT D14\nKO D14\nB\n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted March 27, 2025. ; https://doi.org/10.1101/2025.03.26.645456doi: bioRxiv preprint \n\nA\nMMP9\npMMP2\naMMP2\nAlox15-/- Control\n Wildtype Alox15-/- Alox15-/- \n+ high oxylipins\nMMP9\npMMP2\naMMP2\nB\nSupplementary Figure 6\n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted March 27, 2025. ; https://doi.org/10.1101/2025.03.26.645456doi: bioRxiv preprint \n\n0\n40\n80\n120\n160\n200\nWT KO WT KO WT KO\nCxcl3\n***\nFold change \nexpression level\n0\n10\n20\n30\n40\nWT KO WT KO WT KO\nDay:          0                  4                    7  \nFpr2\n0\n20\n40\n60\nWT KO WT KO WT KO\nIrg1\n0\n1\n2\n3\n4\nWT KO WT KO WT KO\nFold change \nexpression level\nDay:          0                  4                    7  Day:          0                  4                    7  \nDay:           0                  4                    7  \nNfkbiz\n0\n10\n20\n30\n40\nWT KO WT KO WT KO\nNlrp3\n0\n10\n20\n30\nWT KO WT KO WT KO\nPtgs2\nDay:          0                  4                    7  Day:          0                  4                    7  \n0\n20\n40\n60\n80\nWT KO WT KO WT KO\nDay:           0                  4                    7  \nFold change \nexpression level\nRetnlg\n0\n5\n10\n15\n20\n25\n30\nWT KO WT KO WT KO\nTrem1\n0\n10\n20\n30\n40\n50\nWT KO WT KO WT KO\nOsm\nDay:          0                  4                    7  Day:          0                  4                    7  \n* ***\n***\n*** ***\n*** *** ***\nSupplementary Figure 7\n0\n10\n20\n30\n40\n50\nWT KO WT KO WT KO\nIl6\nDay:          0                  4                    7  \nFold change \nexpression level\n**\n0\n20\n40\n60\nWT KO WT KO WT KO\nDay:          0                  4                    7  \nIl1b\n***\n0\n10\n20\n30\n40\n50\nWT KO WT KO WT KO\nDay:          0                  4                    7  \nCcl4\n***\n0\n5\n10\n15\n20\n25\n30\nWT KO WT KO WT KO\nCd14\n***\nDay:          0                  4                    7  \nFold change \nexpression level\n0\n2\n4\n6\n8\nWT KO WT KO WT KO\nDay:          0                  4                    7  \nCd274\n*\n0\n10\n20\n30\n40\n50\nWT KO WT KO WT KO\nClec4d\n*\nDay:          0                  4                    7  \n0\n10\n20\n30\n40\n50\nWT KO WT KO WT KO\nDay:          0                  4                    7  \nFold change \nexpression level\nClec4e\n***\n0\n20\n40\n60\n80\nWT KO WT KO WT KO\nDay:          0                  4                    7  \nCsf3\n0\n10\n20\n30\n40\n50\nWT KO WT KO WT KO\nCxcl2\n*\n*\nDay:          0                  4                    7  \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted March 27, 2025. ; https://doi.org/10.1101/2025.03.26.645456doi: bioRxiv preprint \n\n0\n0.5\n1\n1.5\n2\n2.5\n3\n3.5\n0 20 40 60 80 100 120 140\nFold change versus control\nTotal oxylipins added ()\nRosiglitazone (pos control) 1 M\nOxylipins\nVehicle\n***\n*\nSupplementary Figure 8\n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted March 27, 2025. ; https://doi.org/10.1101/2025.03.26.645456doi: bioRxiv preprint \n\nF4/80 \nINFγ \nCD206\nIsotype control No primary\nK10 + K14 \nIL6 \nPSMAD3\nPSTAT3\nIsotype control No primary\nLY6G\nF4-80\nSupplementary Figure 9\n100 m 100 m\n100 m\n100 m 100 m\n100 m\n100 m\n100 m100 m\n100 m\n200 m 200 m\n200 m\n200 m200 m\n200 m\n200 m 200 m\n200 m 200 m\nA\n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted March 27, 2025. ; https://doi.org/10.1101/2025.03.26.645456doi: bioRxiv preprint \n\n5-HETE\n 8-HETE\n9-HETE\n 9-HETE (zoom)\n11-HETE\n 12-HETE\n15-HETE\n 20-HETE\n5-HEPE\n 8-HEPE\n11-HEPE\nSupplementary Figure 10\n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted March 27, 2025. ; https://doi.org/10.1101/2025.03.26.645456doi: bioRxiv preprint \n\n12-HEPE\n 15-HEPE\n18-HEPE 4-HDOHE\n7-HDOHE 8-HDOHE\n20-HDOHE\n 9-HODE\n13-HODE 9-HOTrE\n13-HOTrE 5-HETrE\nSupplementary Figure 10 (cont’d)\n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted March 27, 2025. ; https://doi.org/10.1101/2025.03.26.645456doi: bioRxiv preprint \n\n15-HETrE\n 9-OxoODE\n13-OxoODE\n12-OxoETE\n 15-OxoETE\n9,10-DiHOME\n 12,13-DiHOME\n5,6-DHET\n 8,9-DHET\n11,12-DHET\n 14,15-DHET\n5-OxoETE\nSupplementary Figure 10 (cont’d)\n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted March 27, 2025. ; https://doi.org/10.1101/2025.03.26.645456doi: bioRxiv preprint \n\n5,15-DiHETE\n 17,18-DiHETE\nLeukotriene B4\n9,10-EpOME\n 12,13-EpOME\n5,6-EET\n7,17-diHDOHE, co -elutes with resolvinD5 standard\nSupplementary Figure 10 (cont’d)\n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted March 27, 2025. ; https://doi.org/10.1101/2025.03.26.645456doi: bioRxiv preprint \n\n11,12-EET 14,15-EET\n7,8-EpDPA 10,11-EpDPA\n13,14-EpDPA 16,17-EpDPA\nPGD1\nPGD2 PGE1\nSupplementary Figure 10 (cont’d)\n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted March 27, 2025. ; https://doi.org/10.1101/2025.03.26.645456doi: bioRxiv preprint \n\nPGE2\n PGE3\nPGB2\n 13,14-dihydro-15-keto PGE2\n15-deoxy-PGJ2\n PGF2α\nThromboxane B2\n6-trans Leukotriene B4\n8-iso PGE2 \nSupplementary Figure 10 (cont’d)\n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted March 27, 2025. ; https://doi.org/10.1101/2025.03.26.645456doi: bioRxiv preprint \n\nPE 16:0p_5-HETE \nPE 18:0p_5-HETE\n \nPE 18:0a_5-HETE \nPE 16:0p_11-HETE\n \nPE 18:1p_11-HETE\n \nPE 18:0p_11-HETE \nPE 18:0a_11-HETE \nSupplementary Figure 11\n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted March 27, 2025. ; https://doi.org/10.1101/2025.03.26.645456doi: bioRxiv preprint \n\nPE 16:0p_12-HETE\nPE 18:0p_12-HETE\nPE 18:0a_12-HETE\nPE 16:0p_15-HETE\n \nPE 18:1p_15-HETE \nPE 18:0p_15-HETE\nPE 18:0a_15-HETE\nSupplementary Figure 11 (cont’d)\n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted March 27, 2025. ; https://doi.org/10.1101/2025.03.26.645456doi: bioRxiv preprint \n\n9.2 9.6 10.0 10.4 10.8\nTime, min\n0.0\n2.0e4\n4.0e4Intensity, cps\n10.06\nRvD5 standard 50 pg\nm/z 359.2-199.1\nm/z 359.2-141.1\nm/z 359.2-261.1\n9.0 10.0 11.0 12.0 13.0\nTime, min\n0.00\n3.00e4\n6.00e4\n9.00e4\n1.20e5\nIntensity, cps\n11.05\n10.92\n10.06\nDay 1 wound lipid extract\n9.2 9.6 10.0 10.4\nTime, min\n0.0\n5000.0\n1.0e4\nIntensity, cps\n10.06\nm/z 359.2-199.1\nm/z 359.2-141.1\nm/z 359.2-261.1\n9.8 10.0 10.2 10.4\nTime, min\n0.0\n5000.0\n1.0e4Intensity, cps\n10.06\n9.0 10.0 11.0 12.0\nTime, min\n3.0e4\n6.0e4Intensity, cps\n10.06\nWT \nAlox15 KO \nm/z 359.2-199.1\n7,17-diHDOHE\nZoomed in from A\nFurther zoomed in showing that /z 359.2-261.1 \ndoesn’t co-elute with the other MRM transitions. \nSeveral peaks present in WT extract are absent in \nAlox15-/-, including 7,17-diHDOHE (RvD5). \n7,17-diHDOHE\n7,17-diHDOHE\nA\nB C\nD E\nm/z 359.2-199.1\nm/z 359.2-141.1\nm/z 359.2-261.1\nm/z 359.2-199.1\nm/z 359.2-141.1\nm/z 359.2-261.1\nSupplementary Figure 12\n217\n199 [217-H20]\n141261\n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted March 27, 2025. ; https://doi.org/10.1101/2025.03.26.645456doi: bioRxiv preprint \n\n6.85.2 6.0 7.6 8.4\n5.48\n0.0\n3.0e4\n6.0e4\n7.26\n5.2 6.0 6.8 7.6 8.4\nTime, min\n0.0\n3.0e4\n6.0e4Intensity, cps\n7.23\nRvD5 standard 50 pg\nm/z 359.2-199.1\nm/z 359.2-141.1\nm/z 359.2-261.1\n5.2 6.0 6.8 7.6 8.4Time, min0.0\n5.0e4\n1.0e5\n1.5e5Intensity, cps\n7.89\n5.20\n7.22Intensity, cps\n6.6 8.2\n7.22\n0.0\n3.0e4\n6.0e4Intensity, cps\nUnspiked lipid extract from \nDay 1 showing peak that co-\nelutes with RvD standard\nSame lipid extract spiked \nwith 25 pg RVD5 \nstandard\nChiral analysis Chiral analysis\n5.4 6.2 7.0 7.8 8.6\nTime, min\n0.00\n3.00e4\n6.00e4\n9.00e4\nIntensity, cps\n7.89\nm/z 359.2-199.1\nm/z 359.2-141.1\nm/z 359.2-261.1\nm/z 359.2-199.1\nm/z 359.2-141.1\nm/z 359.2-261.1\nWT extract\nAlox15-/- extractThe peaks present in WT extract are absent in \nAlox15-/-, including 7,17-diHDOHE (RvD5). \nSupplementary Figure 12 (cont’d)\nF\nG\nH I\nJ\n7.22\n6.84\n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted March 27, 2025. ; https://doi.org/10.1101/2025.03.26.645456doi: bioRxiv preprint \n\nSupplementary Figure 13\nA\nC\nList 1, Wild Type, 1705 List 2, Wild Type, 154 List 2, Alox12-/-, 154B\nList 3, Wild Type, 538 List 3, Alox12-/-, 538\n0\n2\n4\n6\n8\n0 2 4 6 8\nWT\nKO\nFPKM expression data \nDayElf4\nD\n****** ******\n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted March 27, 2025. ; https://doi.org/10.1101/2025.03.26.645456doi: bioRxiv preprint \n\n0\n10\n20\n30\n40\n50\n0 2 4 6 8\nWT\nKO\n0\n20\n40\n60\n80\n100\n0 2 4 6 8\nWT\nKO\nFPKM expression data \nFPKM expression data \nDay Day\nTcf3 Cebpb\n*** (KO)\n* (KO)\nA B\nSupplementary Figure 14 \nGroup (i)\nGroup (ii)\nGroup (iii)\nC\n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted March 27, 2025. ; https://doi.org/10.1101/2025.03.26.645456doi: bioRxiv preprint","source_license":"CC-BY-4.0","license_restricted":false}