Maresin-2 promotes mucosal repair and has therapeutic potential when encapsulated in thermostable nanoparticles

Significance

Maresin-2 (MaR2) is a bioactive lipid with potent anti-inflammatory properties. However, its role as a pro-repair mediator during epithelial wound healing remains undefined. Here we show that MaR2 drives epithelial migration and mucosal repair following DSS-induced colitis or mechanical biopsy-induced injury of the colonic mucosa. A limitation for the therapeutic use of lipids is the requirement for ultracold specialized freezers for storage. To overcome the labile nature of MaR2, we generated polylactic acid thermostable nanoparticles. Such encapsulation preserved pro-repair effects of MaR2 in vivo even following storage at temperatures of 4 °C and above, suggesting increased bio- and thermostability. Therefore, MaR2 nanoparticles have great translational therapeutic potential as most medical facilities would meet conditions required for their storage.

Abstract

Resolution of inflammation and mucosal wound healing are crucial processes required to re-establish homeostasis following injury of mucosal tissues. Maresin-2 (MaR2), a lipid specialized pro-resolving mediator derived from omega-3 polyunsaturated fatty acid, has been reported to promote resolution of inflammation. However, a potential role for MaR2 in regulating mucosal repair remains undefined. Using lipidomic analyses, we demonstrate biosynthesis of MaR2 in healing intestinal mucosal wounds in vivo. Importantly, administration of exogenous MaR2 promoted mucosal repair following dextran sulfate sodium-induced colitis or biopsy-induced colonic mucosal injury. Functional analyses revealed that MaR2 promotes mucosal wound repair by driving intestinal epithelial migration through activation of focal cell-matrix adhesion signaling in primary human intestinal epithelial cells. Because of its labile nature, MaR2 is easily degradable and requires ultracold storage to maintain functionality. Thus, we created thermostable polylactic acid MaR2 nanoparticles that retain biological activity following extended storage at 4 °C or above. Taken together, these results establish MaR2 as a potent pro-repair lipid mediator with broad therapeutic potential for use in promoting mucosal repair in inflammatory diseases.

Epithelial cells in mucosal surfaces reside at the interface of distinct environments and serve as protective barriers from environmental antigens and pathogens and play an important role in regulating tissue homeostasis (1–3). For example, intestinal epithelial cells (IECs) form a selectively permeable barrier that separates mucosal tissues from a luminal environment rich in microbes and antigens (4).

Injury to the epithelial barrier resulting in erosions and wounds is characteristic of several pathologic conditions that include acute and chronic inflammatory diseases. Epithelial injury results in exposure of host tissues to environmental antigens and pathogens and contributes to the pathogenesis of inflammatory disorders (1, 5, 6). As such, disease activity in chronic inflammatory bowel disease (IBD) is associated with persistent epithelial damage and impaired repair resulting in mucosal erosions/ulceration, increased exposure to luminal antigens, and exacerbation of disease symptoms (7, 8). Anti-inflammatory and immunosuppressive approaches have long been a cornerstone of IBD therapeutics (9, 10). However, given that a considerable percentage of patients with chronic inflammatory disorders fail to improve under these treatments, there is an unmet need for therapeutic approaches that drive resolution of inflammation and mucosal repair in order to restore the critical epithelial barrier and mucosal homeostasis.

In response to injury, epithelial cells migrate as a collective sheet to cover denuded surfaces and restore the critical epithelial barrier. Migrating epithelial cells remodel focal cell-matrix adhesions in order to achieve coordinated migration into wounds. Additionally, immune cells are recruited to sites of injury. Crosstalk of infiltrating immune cells, mesenchymal, and epithelial cells has been proposed to play an important role in repair. Infiltrating immune cells in mucosal organs such as the gut, skin, lung, and eye synthetize specialized pro-resolving mediators (SPMs) and release immunoregulatory factors that induce expression of epithelial G-protein-coupled receptors (GPCRs) (11–15). SPMs are bioactive pro-repair molecules that signal to promote resolution of inflammation and drive tissue repair and regeneration (16–19). Maresins (macrophage mediators in resolving inflammation, MaRs) represent a family of lipid SPMs that are derived from omega-3 polyunsaturated fatty acid (20). Two MaR family members (MaR1 and MaR2) have previously been shown to reduce peritoneal inflammation by limiting neutrophil (PMN) infiltration and enhancing macrophage uptake of apoptotic PMN (21). While a previous report demonstrates that MaR1 has protective effects during experimental colitis in vivo (22), a potential role for MaR2 in regulating mucosal repair has yet to be reported.

Here we demonstrate that MaR2 promotes colonic mucosal repair and that MaR2 encapsulated in thermostable nanoparticles has therapeutic potential for use in chronic inflammatory diseases such as IBD.

Results

MaR2 Is Released in Healing Colonic Mucosal Wounds In Vivo.

Previous work has demonstrated that the specialized pro-resolving lipid mediator Resolvin (Rv) E1 plays an important role in promoting intestinal mucosal repair (23). Multiple-reaction monitoring liquid chromatography-tandem mass spectrometry (MRM-LCMS) analyses of intact colonic tissue and healing colonic mucosal wounds harvested between 48 h and 96 h post-injury (Fig. 1A) revealed significantly increased biosynthesis of the lipid mediator MaR2 in injured tissue (11.1 ± 5.0 pg, non-wounded tissue control; 351.1 ± 116.8 pg, 48 h, *P < 0.05; 180.1 ± 44.8 pg, 72 h; and 262.1 ± 84.9 pg, 96 h, *P < 0.05; Fig. 1B). In contrast to MaR2, levels of a related lipid family member MaR1 did not show significant changes following colonic mucosal injury (23).

Fig. 1.

MaR2 is generated in the colon in response to mucosal injury. (A) Consecutive wounds were created on the dorsal aspect of the distal colon. Punch biopsies of wounds and control (non-wounded tissue) were harvested for analysis of lipid mediators by MRM-LCMS. (B) MaR2 molecular structure. Results are expressed as mean ± SEM. N = 3 samples each composed of biopsies from three mice with punch biopsies obtained from non-wounded (Control) and healing wounded tissue from distal colon wounds. Values are expressed as pg/100 mg tissue. *P < 0.05.

MaR2 Promotes Colonic Mucosal Wound Healing In Vivo.

Given the observed increase in MaR2 biosynthesis in healing wounds, we next evaluated the effect of exogenous MaR2 administration on intestinal mucosal repair using a well-characterized biopsy wounding model (24, 25). As can be seen in Fig. 2A, intraperitoneal (i.p.) injection of MaR2 (2 ng/g body weight) resulted in a significant increase in colonic mucosal wound healing at 72 h relative to mice administered PBS/vehicle control (34.58 ± 2.98%, Vehicle; 47.67 ± 2.36%, MaR2; **P < 0.01). To verify effects of MaR2 on mucosal wound repair, we utilized a second model of colonic injury and repair that analyzes recovery from acute dextran sulfate sodium (DSS)-induced colitis. Importantly, administration of MaR2 (2 ng/g body weight) significantly improved recovery from DSS-induced colitis (**P < 0.01, ***P < 0.001, and ****P < 0.0001; Fig. 2B) as determined by the disease activity index (DAI), a composite score encompassing weight loss, stool consistency, and the presence of occult blood. Additionally, histological analyses of colonic tissues revealed that mice treated with MaR2 had significantly reduced mucosal ulceration/erosion and reduced numbers of infiltrating immune cells compared to mice treated with vehicle control (5.474 ± 0.472 HCS, Vehicle; 3.951 ± 0.493 HCS, MaR2, *P < 0.05; Fig. 2C). Taken together, these findings suggest that MaR2 plays an important role in driving colonic mucosal repair.

Fig. 2.

MaR2 promotes colonic mucosal repair. (A) Biopsy wounds were made in the dorsal aspect of the descending colon. 2 ng/g of MaR2 or vehicle control was administered by IP injection 24 h post-wounding. Digital measurement of wound surface area was performed at 24 and 72 h post-wounding. White-dashed lines show wound perimeter. Quantification of colonic mucosal wound healing was determined by measuring wound areas at 24 and 72 h and normalizing against vehicle control. Data are representative of two independent experiments with three and four mice per group and are expressed as means ± SEM. **P < 0.01. (B) Mice were subjected to 5% DSS for 5 d followed by 4 d of water to promote recovery. Disease activity index (DAI) scores are represented as average scores of 0 to 4 for percent weight loss, stool consistency, and presence of occult blood. 2 ng/g of MaR2 or vehicle control was administered by IP injection on day 0, 2, 4, 6, and 8 (red stars). Data are represented as means ± SEM. N = 4 independent experiments with 4 to 6 mice per group. **P < 0.01, ***P < 0.001, and ****P < 0.0001. (C) Representative images of H&E-stained Swiss roll colon sections. Black regions are magnified in insets. Histological scoring of hematoxylin and eosin (H&E)-stained tissue sections of colonic mucosa: Histological colitis score (HCS) represents a ratio of the length of injured/ulcerated areas relative to the entire colon length, as assessed in Swiss roll mounts of the entire colon. Points represent individual mice. Data are representative of three independent experiments with 4 to 5 mice per group and are expressed as means ± SEM. *P < 0.05.

MaR2 Drives Intestinal Epithelial Wound Repair In Vitro.

Efficient repair of epithelial wounds is critical to restore mucosal barrier function in the intestine and dampen inflammatory responses. Given that MaR2 increased colonic mucosal wound repair in mice, we next determined whether MaR2 promotes intestinal epithelial repair in vitro. Intestinal epithelial monolayers were scratch-wounded, and repair was monitored by time-lapse imaging. Dose–response and time course studies using the model human IEC line HT29/B6 revealed that exposure to MaR2 (at a range of concentrations from 100 to 800 nM) did not affect rates of IEC wound closure (SI Appendix, Fig. S1A). Given that pro-repair functions of SPMs such as MaR2 are mediated via binding to receptors, which are themselves up-regulated by cytokines in an inflammatory environment (26), we next examined if pro-repair functions of MaR2 are observed upon co-incubation with the pro-inflammatory cytokines TNFα and IFNγ. Dose–response scratch wound healing assays revealed that 10 ng/mL of TNFα with IFNγ induced a relatively small, but significant increase in IEC wound closure at 24 h, while higher cytokine concentrations resulted in almost 100% wound closure at this time point (**P < 0.01; SI Appendix, Fig. S1B). Therefore, to better detect the additive effects of combining cytokine treatment with MaR2, we exposed scratch-wounded HT29/B6 IEC monolayers to 10 ng/mL TNFα and IFNγ in combination with MaR2 (50 to 200 nM). Importantly, incubation of IECs with 200 nM MaR2 in combination with 10 ng/mL TNFα/IFNγ resulted in significantly increased wound closure at 24 h post-wounding relative to treatment with cytokines alone (***P < 0.001; SI Appendix, Fig. S1C). Further analysis revealed that exposure of HT29/B6 IECs to MaR2 in combination with TNFα/IFNγ significantly increased epithelial repair (relative to treatment with TNFα/IFNγ alone) at all time points measured between 16 and 24 h (**P < 0.01, and ***P < 0.001; Fig. 3A). Taken together, data suggest that that TNFα/IFNγ potentiate pro-repair effects of MaR2.

Fig. 3.

MaR2 promotes intestinal epithelial wound repair during inflammation. IEC monolayers scratched and treated with vehicle control (Vehicle), MaR2 (200 nM), or TNFα (10 ng/mL)/IFNγ (10 ng/mL) ± 200 nM of MaR2. Wound areas from 0 h to 24 h post-scratching/treatment were used to calculate the percentage of wound closure. (A) Percentage of wound closure in HT29/B6 IECs at 3, 6, 12, 16, 20, and 24 h post-wounding. Results are expressed as mean ± SEM. N = 5. * TNFα/IFNγ vs. TNFα/IFNγ/MaR2: **P < 0.01 and ***P < 0.001. (B) Cell migration analyzed by mean-squared displacement was calculated following the trajectories of 20 cells per wound in images obtained every hour from scratch-wounded HT29/B6 IECs. Results are expressed as mean ± SEM. N = 3. * TNFα/IFNγ vs. TNFα/IFNγ/MaR2: *P < 0.05, **P < 0.01, and ****P < 0.0001. (C) Percentage of wound closure in human colonoids at 3, 6, 16, 20, and 24 h post-wounding. Results are expressed as mean ± SEM. Data from two experiments of three monolayers. * TNFα/IFNγ vs. TNFα/IFNγ/MaR2: *P < 0.05, and **P < 0.01. (D) Cell migration analyzed by mean-squared displacement was calculated following the trajectories of 20 cells/wound in images obtained every hour from scratch-wounded human colonoids. Results are expressed as mean ± SEM. Data from one experiment of three monolayers. * TNFα/IFNγ vs. TNFα/IFNγ/MaR2: *P < 0.05 and **P < 0.01.

Since epithelial repair is mediated by collective migration of cells, we next assessed effects of MaR2 on IEC migration efficiency following in vitro scratch wounding and time-lapse imaging. Mean-squared cell displacement (MSD) tracks the surface area explored by cells over time and is a common metric used to assess cell migration speed, distance traveled, and migration directionality. Calculation of MSD for IECs at the leading edge of healing epithelial wounds was calculated according to the method described by Gorelik and Gautreau (27). These analyses revealed that IECs exposed to 200 nM MaR2 in combination with 10 ng/mL TNFα/IFNγ had significantly higher MSD values compared to cells treated with TNFα/IFNγ alone (*P < 0.05, and **P < 0.01; Fig. 3B), suggesting MaR2 enhances IEC migration efficiency. To further analyze effects of MaR2 on epithelial repair, we utilized primary human colonic epithelial cells referred to as colonoids. Human colonoids were cultured and differentiated into two-dimensional (2D) monolayers. Analogous to results obtained with the model intestinal epithelial cell line, HT29/B6 cells, 200 nM MaR2 in combination with 10 ng/mL TNFα/IFNγ significantly increased primary colonoid wound repair relative to treatment with TNFα/IFNγ alone (*P < 0.05; and **P < 0.01; Fig. 3C). In addition, treatment with 200 nM MaR2 and 10 ng/mL TNFα/IFNγ resulted in higher MSD values compared to colonoids treated with cytokines alone demonstrating that MaR2 results in increased efficiency of primary IEC migration and repair (**P < 0.01 and ****P < 0.0001; Fig. 3D).

In addition to collective cell migration, proliferation can also contribute to epithelial repair processes (28). Therefore, we next analyzed effects of MaR2 on IEC proliferation following injury. Treatment with MaR2 (alone or in combination with TNFα/IFNγ) did not have any significant effect on rates of IEC proliferation following scratch wounding as assessed by EdU incorporation. Taken together, these data demonstrate that MaR2 drives wound healing by promoting collective IEC migration under the influence of inflammatory stimuli.

MaR2 Activates Pro-Migration Signaling Pathways during IEC Wound Healing.

To investigate the mechanism(s) by which MaR2 promotes epithelial repair, we analyzed focal adhesion (FA) proteins that regulate cell migration and therefore wound repair. 16 h post-injury, treatment of IECs with MaR2 (in combination with TNFα/IFNγ) resulted in increased phosphorylation of focal adhesion kinase (FAK) at the activation site Tyr³⁹⁷ which has been shown to regulate focal adhesion dynamics and cell movement. Importantly, activation of FAK^Y397 following treatment with MaR2 with TNFα/IFNγ was significantly increased relative to treatment with cytokines alone (**P < 0.01; Fig. 4A). Previous studies have demonstrated that Src signals downstream of FAK to promote epithelial cell migration and wound closure (29, 30). Here we report increased activation of Src (Src^Y416) as well as decreased phosphorylation at the inactivation site (Src^Y527) in healing IECs exposed to MaR2 with TNFα/IFNγ (**P < 0.01; Fig. 4A). In addition to signaling mediators, vinculin and talin are key structural proteins that couple the actin cytoskeleton to integrins and the extracellular matrix to facilitate epithelial migration. Importantly, significantly increased levels of vinculin and talin were observed in IEC exposed to MaR2 with TNFα/IFNγ compared to cells exposed to cytokines alone (**P < 0.01; Fig. 4A).

Fig. 4.

MaR2 promotes phosphorylation signals that drive migration. HT29/B6 IEC monolayers were scratch-wounded and treated with vehicle control (Vehicle), MaR2 (200 nM), or TNFα (10 ng/mL)/IFNγ (10 ng/mL) ± 200 nM MaR2 for 16 h. (A) Levels of vinculin, talin, FAK, and Src and their phosphorylated versions FAK^Y397, Src^{Y527 and Y416} were compared by immunoblotting. Densitometry values of phosphorylated protein blots were normalized to total protein levels. Levels of vinculin and talin were normalized to loading control. Results are expressed as mean ± SEM. N = 6. * TNFα/IFNγ vs TNFα/IFNγ/MaR2: **P < 0.01. (B) Representative thermal Math Lookup Table (LUT) micrographs and corresponding surface plot of focal contacts at the leading edge of wounded IEC monolayers (white arrows). IECs were treated with vehicle control (Vehicle), MaR2 (200 nM), or TNFα (10 ng/mL)/IFNγ (10 ng/mL) ± MaR2 (200 nM) for 16 h post-wounding. Immunofluorescence staining for pPaxillin^Y118 and pFAK^{Y397 and Y861} is shown in Green Fire Blue LUT. (Scale bar: 100 μm.) Results are expressed as mean ± SEM. N = 3 to 5. * TNFα/IFNγ vs. TNFα/IFNγ/MaR2: *P < 0.05, and **P < 0.01. (C) Spreading HT29/B6 IEC were treated with vehicle control (Vehicle), MaR2 (200 nM), or TNFα (10 ng/mL)/IFNγ (10 ng/mL) ± 200 nM MaR2 for 16 h. Representative confocal micrographs of focal contacts at the leading edge of IECs (white arrowheads). Immunofluorescence staining for vinculin is shown in white with nuclei shown in blue. (Scale bar: 100 μm.) Results are expressed as mean ± SEM. N = 5. * TNFα/IFNγ vs TNFα/IFNγ/MaR2: **P < 0.01.

Next, we determined effects of MaR2 on cellular localization of mediators that promote IEC migration. For these analyses, immunofluorescence (IF) labeling of IEC wounds 16 h after injury was performed. Such analyses revealed that treatment with MaR2 in combination with TNFα/IFNγ significantly increased localization and activation of the FA adaptor protein paxillin (paxillin^Y118) at the leading edge of cells migrating to heal wounds (**P < 0.01; Fig. 4B). Importantly, a significant increase in FAK activation (FAK^Y397 and FAK^Y861) at the leading edge of migrating cells was observed after treatment with MaR2 with TNFα/IFNγ compared to IECs incubated with TNFα/IFNγ alone (*P < 0.05; Fig. 4B). Given these observed changes in FAK and paxillin localization, we next quantified changes in the focal adhesion protein vinculin. MaR2 in combination with TNFα/IFNγ resulted in significantly increased numbers of vinculin containing FAs at the leading edge of healing IEC wounds relative to IECs exposed only to TNFα/IFNγ (**P < 0.01; Fig. 4C). Taken together, these results suggest that MaR2 activates FAK-Src-paxillin and talin-vinculin signaling axes to drive collective IEC migration and mucosal wound repair in the intestine.

Polylactic Acid (PLA) Nanoparticles increase Thermostability of MaR2.

Nanoparticles (NP) encapsulation has been shown to stabilize lipid SPMs and enhance repair of epithelial skin wounds (31) and corneal abrasions (13). In the intestine, NP-encapsulated RvE1 was shown to increase mucosal wound repair by ∼20% relative to un-encapsulated RvE1 (23). Therefore, we generated nanoparticle encapsulated MaR2 using PLA, a biocompatible polymer (32) that can be used to produce PLA nanoparticles, the breakdown products of which are normal metabolites (lactic acid) (33). We synthetized Empty NPs and MaR2 NPs through the “water-in-oil-in-water” emulsion method and imaged the MaR2 NPs by scanning electron microscopy (SEM) as shown in Fig. 5A. To determine if NP encapsulation of MaR2 potentiated increased mucosal healing, effects of MaR2 NPs on repair following DSS-induced colitis were determined. As can be seen in SI Appendix, Fig. S2A, while administration of 2 ng/g of body weight MaR2 or MaR2 NPs (1 ng MaR2 / 2 μg Empty NPs) significantly improved recovery from DSS-induced colitis relative to vehicle controls (*P < 0.05, ***P < 0.001, and ****P < 0.0001), no significant difference in mucosal recovery was observed between MaR2 and MaR2 NPs.

Fig. 5.

Nanoparticle encapsulation of MaR2 extends its shelf life and preserves pro-repair effects in vivo. (A) Polylactic acid in oil phase (PLA) is mixed with MaR2 dissolved in ethanol and sonicated in water. Dried PLA nanoparticles (NP) containing MaR2 (MaR2 NP) are generated and as observed by scanning electron microscopy (SEM) micrography (black square). (B) Mice were subjected to 3% DSS for 5 d followed by 4 d of water recovery. Disease activity index scores are represented as average scores of 0 to 4 for percent weight loss, stool consistency, and presence of occult blood. Nanoparticles containing 50 ng/g of MaR2 (MaR2 NP), or empty control nanoparticles (Empty NP), stored at 4 °C or room temperature (RT) were administered by IP injection on days 2, 4, 6, and 8 (red stars). Data are represented as means ± SEM. N = 2 independent experiments with seven mice per group. 4 °C Empty NP vs. 4 °C MaR2 NP: black *P < 0.05, ***P < 0.001, and ****P < 0.0001. RT Empty NP vs RT MaR2 NP: blue *, *P < 0.05, **P < 0.01, and ***P < 0.001. (C) Histological scoring of hematoxylin and eosin (H&E)-stained tissue sections of colonic mucosa. Histological colitis score (HCS) represents a ratio of the length of injured/ulcerated areas relative to the entire colon length, as assessed in Swiss roll mounts of the entire colon. Data points represent individual mice. Data are representative of three independent experiments with five mice per group and are expressed as means ± SEM. *P < 0.05. (D) Representative images of H&E-stained Swiss roll colon sections. Black-boxed regions are magnified in Insets.

A second considerable advantage of NP encapsulation is that it extends shelf life, expands thermostability, and mitigates extreme temperatures required for effective storage of labile reagents including lipids (34). Similar to other thermolabile lipids, it is recommended that MaR2 be stored at minus 80 °C in order to retain its biological activity. Therefore, we determined effects of one month of storage at 4 °C or room temperature (RT) on bioactivity of MaR2 and MaR2 NPs. As expected, storage of MaR2 at 4 °C or RT resulted in a loss of pro-repair effects following DSS-induced colitis (SI Appendix, Fig. S2 B and C). These findings suggest that MaR2 loses its potent pro-repair activity when stored at a temperature of 4 °C or above. In contrast, MaR2 NPs stored at 4 °C or RT retained biological function and significantly improved recovery from DSS-induced colitis (*P < 0.05; ^**P < 0.01; ^***P < 0.001, and ^****P < 0.0001; Fig. 5B). Importantly, histological analysis of colonic sections from mice treated with MaR2 NPs that had been stored at 4 °C or above revealed significantly less mucosal ulceration compared to mice treated with un-encapsulated MaR2 stored at the same temperatures (*P < 0.05; Fig. 5 C and D). These results demonstrate that encapsulation of MaR2 in NPs results in lipid stabilization and retention of potent mucosal repair properties without the need for stringent low-temperature storage, highlighting the increased thermostability and potential clinical utility of NP-encapsulated MaR2.

Discussion

Epithelial barriers at mucosal and dermal surfaces act as shields that separate internal and external environments protecting underlying tissues from toxins and infectious pathogens. Epithelial barrier disruption is a hallmark of numerous pathologic conditions that include acute and chronic inflammatory diseases. In response to epithelial damage, overlapping stages of wound healing consisting of inflammation, proliferation/re-epithelization, and remodeling (35) must occur in order to restore critical epithelial barrier function. Indeed, impaired epithelial repair contributes to the pathogenesis of chronic inflammatory diseases including (36–38) IBD (7).

Epithelial wound healing is regulated by inflammatory and resolving mediators, including protein and lipid SPMs, that are released into wound environments by epithelial cells as well as infiltrating immune cells (1, 39). Maresins are SPMs that have been proposed to have anti-inflammatory and pro-resolving functions (20). Here we show increased biosynthesis of MaR2, but not the related SPM MaR1, in the healing colonic mucosa. Furthermore, temporal analysis of MaR2 levels in biopsy-induced colonic wounds revealed maximal increases within 2 d of injury, a period that represents a transition between the pro-inflammatory and restorative phases of mucosal repair when inflammatory responses need to be dampened and restorative pathways are activated. In keeping with the importance of SPM biosynthesis during intestinal inflammation and repair, increased synthesis of SPMs including Rvs and protectins (Pds) has previously been reported in colonic mucosal biopsies from IBD patients (40, 41) and in inflamed and healing murine colonic mucosa (23, 40). MaR2 is generated by macrophages, which in the inflamed colonic mucosa could be from resident or infiltrating macrophages (21, 42). Our previous study using the punch-biopsy model of colonic mucosal injury demonstrated that infiltrating monocytes/macrophages peak 24 to 48 h after injury (43). This corresponds to the highest levels of MaR2 that was detected in healing colonic mucosa after biopsy-induced injury. These findings suggest that macrophages are likely the main source of MaR2. Nevertheless, during DSS-induced colitis immune cell infiltration varies depending on the mucosal damage, and therefore, it is more difficult to determine the specific cell source of SPMs (44). It is not known if resident macrophages are capable of synthesizing MaR2 or if other cell types, such as epithelial cells, other immune cells or even support cells as fibroblasts or mesenchymal cells, could secrete MaR2. The therapeutic potential of SPMs is further highlighted by our data showing that MaR2 promotes mucosal repair following DSS-induced colitis or following mechanical induced colonic mucosal injury. Data showing that MaR2 drives mucosal repair in the gut is supported by previous studies demonstrating that lipids including RvD1, RvD2, RvE1, and PdD1n-3 docosapentaenoic acid (n-3 DPA), RvD5n-3 DPA, and MaR1 reduce clinical injury in TNBS- and DSS-induced colitis (22, 40, 45, 46). SPMs (including RvE1, and RvD1) have also been shown to promote dermal repair (47–49) and drive mucosal repair following acute lung injury (49–51). Moreover, RvE1, RvD1, PdD1, and MaR1 have been shown to induce intestinal epithelial wound repair in the in vitro model of IECs Caco-2 (52). However, our study is the first to demonstrate that MaR2 directly promotes intestinal repair.

Mechanistically, we show that MaR2 drives IEC migration and repair in an inflammatory environment. This is consistent with MaR2 binding to an epithelial GPCR receptor, several of which are known to be up-regulated by IFNγ and TNFα during mucosal inflammation (11, 53, 54). Furthermore, we have previously reported that TNFα up-regulated in healing biopsy-induced wounds and that antibody mediated neutralization of this cytokine down-regulates expression of the GPCR, platelet activating factor receptor and, impairs colonic mucosal wound repair (11). Critically, IFNγ and TNFα are important mediators in the inflammatory response observed in IBD (55, 56) and in experimental murine colitis (57, 58). GPCR signaling can promote remodeling and turnover of integrin-containing cell-matrix contacts to regulate epithelial migration and mucosal wound repair. In macrophages, MaR1 binds to the GPCR LGR6, leading to enhanced phagocytosis, efferocytosis, and phosphorylation of ERK and CREB; pathways involved in facilitating wound repair. In epidermis, LGR6 signaling is important for stem cell homeostasis, suggesting that MaR1 might also have a role in epithelial stem cell regulation. Unfortunately, a MaR2 receptor in IEC has not been identified. Many of the other SPM receptors are GPCRs, and it is likely that the MaR2 receptor is also in this family of proteins. We have observed that MaR2 signaling is enhanced after incubation of epithelial cells with TNFα. This suggests that the MaR2 receptor, as with other epithelial GPCRs, is up-regulated during an inflammatory response. These findings further support the current paradigm that the pro-inflammatory environment sets the stage for resolution of inflammation and repair. This delicate balance of inflammation, resolution, and repair is perturbed in chronic inflammatory disorders such as IBD, thereby contributing to the disease pathogenesis.

We observed that MaR2 activates Src-FAK signaling in IECs to promote collective epithelial cell migration required for wound repair. This is in keeping with previous studies showing that Src-FAK signaling is critical for regulating epithelial migration (29, 59, 60). In addition to MaR2 mediated activation of Src-FAK signaling, we also observed increased activation of paxillin (pTyr118) at the leading edge of migrating IECs. This finding is supportive of a previous report showing increased paxillin activation leading to enhanced epithelial migration in corneal epithelial cells (61). Furthermore, MaR2 signaling promotes recruitment of paxillin binding FA structural proteins talin and vinculin that have been shown to control migration of corneal and breast epithelial cells (62–64). We observed increased phosphorylation and activation of FA proteins at the leading edge the migrating cells which provide an important directional driving force to orchestrate collective epithelial migration and wound repair (Fig. 6).

Fig. 6.

MaR2 enhances pro-repair signaling in intestinal epithelial cells during intestinal mucosal wound repair. Disruption of the intestinal epithelial barrier results in inflammation and recruitment of immune cells that release specialized pro-repair molecules including MaR2 that facilitate transition into the repair phase (Upper Right). MaR2 signaling drives epithelial wound healing through regulation of IEC migration mediated by Src-paxillin and talin-vinculin.

Like other SPMs, MaR2 is a thermolabile lipid that requires storage at −80 °C in order to retain biological activity. An important consideration when developing therapeutics is transportation and storage conditions, where RT and long-term storage are the goal. Therapies that require ultracold storage will be limited in their reach, regardless of how well they function, because specialized freezers that reach these temperatures are expensive and not widely available. These clinical limitations are highlighted by the recent global vaccination campaign for SARS-CoV-2 (Severe Acute Respiratory Syndrome Coronavirus 2), where vaccines that required ultracold storage were of limited utility in rural areas and developing nations. Previous studies have shown that encapsulation of SPMs within NPs increases both their biostability and thermostability. Here we report that NP encapsulation of MaR2 results in retention of pro-repair effects in vivo even following extended storage at RT. These data highlight the increased thermostability and wide-reaching potential clinical utility of MaR2 NPs. This expanded thermostability for MaR2 NPs is supported by a study showing IFN-α-loaded liposomes remained functionally active following storage at 8 °C for 3 mo (65). To our knowledge, there are no instances of the use of encapsulated SPM nanoparticles in the clinic. Nevertheless, similar NPs are currently in human clinical trials by Cour Pharmaceuticals and Selecta Biosciences. In both cases, particles are administered by intravenous infusion. In fact, Cour Pharmaceuticals has two therapies in human trials pertaining to controlling inflammation in the gut for peanut allergy and/or celiac disease (66). Although beyond the scope of the current investigation, future work will consider more translational routes of encapsulated SPM NPs delivery including IV infusion and oral use.

Taken together, this study demonstrates for the first time that MaR2 promotes IEC migration to drive colonic mucosal wound repair. Developing lyophilized NPs that enhance the thermostability of MaR2 would facilitate long-term storage at readily achievable temperatures thus expanding the potential clinical utility of MaR2 for use in promoting mucosal repair in healthcare facilities lacking specialized cold storage equipment.

Materials and Methods

Mice.

Ten to twelve weeks old male and female C57BL/6 mice were purchased from the Jackson Laboratory and maintained within the animal facilities at the University of Michigan. Studies performed in mice were evaluated and ratified by the University of Michigan, an AAALAC accredited institution.

In Vivo Wounding of Colonic Mucosa.

Distal colonic mucosa was injured at five sites per mouse along the dorsal artery using a high-resolution, miniaturized colonoscope system equipped with biopsy forceps (Coloview Veterinary Endoscope, Karl Storz; Germany). Images of wounded mucosa were obtained at 24 and 72 h post-injury to determine the percentage of wound closure as published previously (23).

Lipidomic Analysis of MaR2 Levels.

25 to 30 punch biopsies (1 mm²) of colonic mucosa from intact or wounded tissue were obtained at 48, 72, and 96 h after wounding. Tissue levels of MaR2 were analyzed by the Queen Mary University London Lipid Mediator Unit, William Harvey Research Institute, Barts and The London School of Medicine and Dentistry, as previously reported (23).

Dextran Sulfate Sodium-mediated Murine Colitis and Recovery Model.

Mice were supplied ad libitum with drinking water containing 3% and 5% weight/volume of DSS salt (40 kDa, J63606, Thermo Fisher Scientific; Ward Hill, MA) for 5 d to induce colitis, followed by 4 d with normal drinking water for recovery. Assessment of disease activity was monitored daily. DAI was calculated using the sum of scores for the percentage loss of body weight (score 0-4), stool consistency (score 0-4), and blood in stool (scores 0-4. Fisher Healthcare™ Sure-Vue™; Houston, TX) / 3, as previously described (30). Higher DAI values indicate a more severe colitis. In addition, colons were isolated and rolled into Swiss rolls for H&E staining and evaluation of the histological colitis score (HCS) using the formula: [inflammation/injury percentage + (2 * erosion/ulceration percentage)] / 10 was performed, as described previously (67).

Cell Lines and Colonoids for Epithelial Wounding In Vitro.

The model human IEC line HT29/B6 along with primary human colonoids was used to perform wound closure assays as previously published (68). Colonoids isolated from healthy human donors were grown as 3D cysts and maintained in a proliferative state to further seed 2D monolayers as has been previously published (69). Confluent monolayers of IECs and 2D colonoids were wounded, and wound closure was analyzed by time-lapse imaging over a 24 h time course.

Reagents.

The following reagents were used: Maresin-2 (cat. 1639809-46-3; Cayman Chemicals), hIFNγ (cat. 285-IF, R&D Systems, Inc.), and hTNFα (cat. 210-TA, R&D Systems, Inc.).

MSD Measurement.

Cell migration efficiency (cell migration speed, distance traveled, and migration directionality) was determined for scratch-wounded IECs and colonoid monolayers by tracking cells over time as described previously (27). Briefly, trajectories of cells at the leading edge of wounds were tracked with images obtained every hour over a 24 h time course using ImageJ software version 1.52a (NIH; https://imagej.nih.gov/ij). The values obtained were analyzed with the algorithm published by Gorelik and Gautreau (27) to generate MSD values in μm².

Immunoblotting and Immunofluorescence Labeling.

Grid-wounded monolayers of HT29/B6 IECs and 2D colonoids were harvested in radioimmunoprecipitation assay lysis buffer (10 mM Tris-HCl, pH 8.0, 140 mM NaCl, 1 mM EDTA, 1% Triton X-100, 0.1% sodium deoxycholate, 0.1% SDS, pH 7.4) with protease and phosphatase inhibitors (Sigma-Aldrich) for immunoblotting analysis as published previously (10). Protein concentrations were quantified using the BioRad DC Protein Assay (Hercules, CA) and an Epoch microplate spectrophotometer (BioTek). Equal quantities of protein samples were loaded into 10% polyacrylamide gels for SDS–PAGE and immunoblotting by standard methods. Levels of protein were quantified using the digital densitometry of protein bands and normalized to GAPDH as a standard protein loading control. IF was performed following standard protocols. Briefly, IEC monolayers were fixed in 4% paraformaldehyde (43368, Thermo Fisher Scientific) and permeabilized using 0.2% Triton X-100 (1122980101, Sigma-Aldrich) before blocking with 1% donkey serum. Monolayers were incubated with primary antibody (>10 μg/mL) at 4 °C overnight. Following, monolayers were incubated with a secondary antibody (2.7 μg/mL for 45 min and 2 min of Hoechst for nuclei staining (1:2,000, H3570, Invitrogen, Thermo Fisher Scientific) to further mount with ProLong Gold (P36934, Invitrogen, Thermo Fisher Scientific) antifade reagent. Fluorescence images were obtained using a Nikon A1 confocal microscope (Tokyo, Japan) and analyzed using ImageJ.

Antibodies against the following proteins were used for immunoblotting: FAK (cat. 610088; BD Biosciences); pFAK^Tyr861 (cat. PS 1008; Calbiochem), pFAK^Tyr397 (cat. 3283; Cell Signaling Technology), Src (cat. 2108), pSrc^Tyr416 (cat. 2101), pSrc^Tyr527 (cat. 2105), vinculin (cat. 4650), talin (cat. 4021), ppaxillin^Tyr118 (cat.2541), GAPDH (cat. G9545, Sigma-Aldrich), HRP-anti-mouse (cat. 115-005-146), and HRP-anti-rabbit (cat. 111-035-144, Jackson Inmmunoresearch Laboratories, Inc.).

Nanoparticles.

PLA particles were prepared using the oil-in-water (o/w) emulsion solvent evaporation technique as previously described (67). Briefly, polymer (PLA, Lactel Absorbable Polymers, 0.21 dL/g, approximately 0.5% of polymer chains) was dissolved in dichloromethane (DCM, Sigma-Aldrich) at 74 mg/mL. 100 µg of MaR2 in 1 mL ethanol was added to the solution of PLA, and MaR2 dissolved in ethanol was added to the solution of PLA. Following this 1% w/v poly(ethylene-alt-maleic acid) (PEMA, MW 400,000, Polysciences, Inc.) in water was added and sonicated at 100% amplitude for 10 s using a Cole-Parmer Ultrasonic processor (model XPS130). The resulting o/w emulsion was added to magnetically stirred 0.5% PEMA overnight to allow evaporation of the DCM. Particles were collected by centrifugation and washed three times with deionized water. Cryoprotectants (4% w/v sucrose, 3% w/v mannitol) were added to the particle suspension, frozen at −80 °C, and lyophilized for storage at RT or 4 °C.

Statistical Analysis.

Statistical significance was performed by unpaired two-tailed Student’s t test or using one- or two-way ANOVA with Bonferroni’s multiple comparison tests. Results are expressed as means ± SEM. P values of 0.05 or less were considered significant.

Supplementary Material

Appendix 01 (PDF)

Data, Materials, and Software Availability

All study data are included in the article and/or SI Appendix.