If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Address reprint requests to Song Hong, Ph.D., LSUHSC-Center of Neuroscience Excellence, Lions Building, 2020 Gravier St., Suite D, New Orleans, LA 70112
Impaired macrophage functions imposed by diabetic complications and the suppressed formation of 14S,21R-dihydroxydocosa-4Z,7Z,10Z,12E,16Z,19Z-hexaenoic acid (14S,21R-diHDHA) in wounds contribute significantly to deficient wound healing in diabetics, but how are macrophage functions and 14S,21R-diHDHA formation associated? We studied 14S,21R-diHDHA generation from macrophages using liquid chromatography/mass spectrometry. The role in macrophage-mediated wound healing functions was determined using a murine splinted excisional wound healing model and in vitro assays. 14S,21R-diHDHA acts as a macrophage-generated autacoid, and its attenuated formation in macrophages of diabetic db/db mice was accompanied by impairment of macrophage prohealing functions. 14S,21R-diHDHA restored db/db macrophage-impaired prohealing functions by promoting wound re-epithelialization, formulation of granulation tissue, and vascularization. Additionally, 12/15-lipoxygenase-deficient macrophages, which are unable to produce 14S,21R-diHDHA, exhibited impaired prohealing functions, which also were restored by 14S,21R-diHDHA treatment. The molecular mechanism for 14S,21R-diHDHA-induced recovery of impaired prohealing functions of db/db macrophages involves enhancing their secretion of vascular endothelial growth factor and platelet-derived growth factor BB, decreasing hyperglycemia-induced generation of reactive oxygen species, and increasing IL-10 expression under inflammatory stimulation. Taken together, these results indicate that deficiency of 14S,21R-diHDHA formation by diabetic macrophages contributes to their impaired prohealing functions. Our findings provide mechanistic insights into wound healing in diabetics and suggest the possibility of using autologous macrophages/monocytes, treated with 14S,21R-diHDHA, or related compounds, to promote diabetes-impaired wound healing.
Impaired wound healing associated with diabetes mellitus presents a major unresolved medical challenge. It is predicted that dysregulated cellular and molecular processes are responsible for impairments in wound healing in diabetes,
but the mechanisms underlying these processes require further elucidation. Macrophages play critical roles in wound healing: they engulf apoptotic polymorphonuclear leukocytes (PMNs), tissue debris, and infectious microorganisms; they promote wound healing, including accelerating re-epithelialization by promoting keratinocyte migration from the wound edge to decrease the epithelial gap and increasing the formation of granulation tissue by promoting fibroblasts and endothelial cells moving into wounds to cover the wound bed
In addition to amino acid-based cytokines and growth factors, which are known to mediate wound healing and associated angiogenesis, lipid-derived molecules also may be important mediators. Docosahexaenoic acid (DHA), an essential ω-3 fatty acid, is a relatively abundant, endogenous lipid component in blood and in wounded full-thickness skin.
Previous studies showed that DHA can ameliorate diabetes complications, such as cardiovascular disease, by improving endothelial function through anti-inflammatory mechanisms and reduced platelet aggregation, as well as by reducing blood triglyceride levels and increasing high-density lipoprotein cholesterol.
Resolvins, neuroprotectins/protectins, and maresins are lipid mediators generated naturally from DHA during acute inflammation that have potent anti-inflammatory and protective functions.
Novel docosatrienes and 17S-resolvins generated from docosahexaenoic acid in murine brain, human blood, and glial cells Autacoids in anti-inflammation.
Their discovery has opened a new avenue for unraveling the mechanisms behind the beneficial effects of DHA. The DHA-derived lipid mediator neuroprotectin D1 (NPD1; 10R,17S-dihydroxydocosa-4Z,7Z,11E,13E,15Z,19Z-hexaenoic acid) enhances corneal wound healing,
and 14S,21R-dihydroxydocosa-4Z,7Z,10Z,12E,16Z,19Z-hexaenoic acid (14S,21R-diHDHA), a novel endogenous DHA-derived lipid mediator, is generated by macrophages and in wounded skin.
14S,21R-diHDHA is produced by sequential enzymatic actions of 14S-hydroperoxylation-specific leukocyte-12-lipoxygenase (L-12-LOX) and ω-1-hydroxylation-specific P450.
Diabetes and the accompanying hyperglycemia and oxidative stress damage DNA, proteins, and lipids in various tissues. These conditions cause cellular and enzymatic dysfunction during the formation of cytokines and eicosanoids associated with cell growth, angiogenesis, and/or inflammation,
We have observed that 14S,21R-diHDHA formation was significantly halted in diabetic wounds, compared with nondiabetic wounds in mice, which suggests that diabetic complications dysregulate the generation of DHA-derived lipid mediators.
indicating that 14S,21R-diHDHA is an important prohealing lipid mediator. Also, decreased 14S,21R-diHDHA formation in diabetes may affect prohealing functions of cells that participate in wound healing.
The findings presented here demonstrate that diabetes-impaired prohealing functions of db/db macrophages are associated with decreased formation of 14S,21R-diHDHA, and could be ameliorated by treating diabetic macrophages with 14S,21R-diHDHA.
Materials and Methods
Mice
Animal protocols were approved by the Institutional Animal Care and Use Committee and the Institutional Review Board of the Louisiana State University Health Sciences Center, New Orleans. Studies were blinded. Diabetic db/db (BKS.Cg-m+/+Leprdb) and nondiabetic db/+ mice (10-week-old females; Jackson Laboratory, Bar Harbor, ME) were used when blood glucose was 22 to 35 mmol/L for db/db mice and 5 to 10 mmol/L for db/+ mice, as measured with a glucometer. Both B6.129S2-Alox15tm1Fun/J [12/15 (or leukocyte-12)-lipoxygenase deficient (12/15-LOX-/-)] and wild-type control (C57BL/6J) mice were 10-week-old females obtained from the Jackson Laboratory.
Cell Cultures
Macrophages were isolated and identified as previously described.
Briefly, mice were treated with thioglycolate for 3 days. Macrophages were obtained by peritoneal lavage and were cultured in RPMI 1640 medium (Invitrogen, Carlsbad, CA) containing glucose at different concentrations; nonadherent cells were deleted by changing the culture medium. For identification of macrophages, cells were seeded into eight-well culture slides (BD Biosciences, San Jose, CA). After being fixed in 4% paraformaldehyde (Sigma-Aldrich, St. Louis, MO) for 15 minutes and permeabilized in 0.1% Triton-100 (Sigma-Aldrich) on ice for 5 minutes, cells were incubated with primary antibody (monoclonal rat anti-mouse F4/80; Santa Cruz Biotechnology, Santa Cruz, CA) at 4°C for 12 hours. Cells were then incubated with Cy3-labeled goat anti-rat IgG (Invitrogen) at 4°C for 12 hours. Nuclei were stained with Hoechst 33342 dye (blue). Images were captured using a Zeiss Axioplan 2 microscope in an Everest digital microscopy workstation with associated software (Slidebook version 4.2; Zeiss). More than 95% of cells were F4/80+ (see Supplemental Figure S1A at http://ajp.amjpathol.org).
Murine fibroblast cell line CF-1 was obtained from the American Type Culture Collection (ATCC, Manassas, VA), and cultured in Dulbecco's modified Eagle's medium containing 10% fetal bovine serum. Dermal microvascular endothelial cells (DMVECs) were isolated from skin of db/db mice and confirmed by immunocytochemical analysis, which demonstrated their purity (>95%) based on the staining of endothelial markers CD31 (red) and VE-cadherin (green); nuclei were stained with Hoechst 33342 blue dye (see Supplemental Figure S1B at http://ajp.amjpathol.org).
Human peripheral blood provided by the New Orleans Blood Bank was collected from a code-deidentified healthy donor who had not taken medication for 2 weeks before donation. Blood was also collected by cardiac puncture from C57BL/6J mice under deep anesthesia.
Polymorphonuclear leukocytes and platelets were separated from human or murine blood as described previously.
Briefly, peripheral blood samples were mixed with the same volume of PBS containing 4% dextran and were centrifuged at 150 × g for 15 minutes. Platelet-rich supernatant was further centrifuged at 1600 × g for 15 minutes to collect platelets. Cell deposit was used to isolate PMNs by density gradient centrifugation using Histopaque 1.077 cell separation medium (Sigma-Aldrich). PMNs had at least 95% purity by Hoechst 33342 dye staining and 95% viability by Trypan Blue dye (Sigma-Aldrich) exclusion. Lymphocytes were isolated from the spleens of db/+ mice by magnet-based microbead CD45R (B cells) and CD90.2 (T cells) according to the manufacturer's protocol (Miltenyi Biotec, Bergisch Gladbach, Germany). Lymphocytes showed more than 95% viability by Trypan Blue dye exclusion.
Formation of 14S,21R-diHDHA and Its Biosynthetic Intermediates and Analysis by Liquid Chromatography–Ultra Violet Detector–Tandem Mass Spectrometry
The formation of 14S,21R-diHDHA and its biosynthetic intermediates and analysis by liquid chromatography-tandem mass spectrometry were performed as described previously,
with some modifications. DHA, DHA-d5, and porcine L-12-LOX were supplied by Cayman Chemical (Ann Arbor, MI). Neuroprotectin/protectin D1 (10R,17S-dihydroxy-docosa-4Z,7Z,11E,13E,15Z,19Z-hexaenoic acid) prepared by total organic synthesis was kindly provided by Dr. Charles N. Serhan (Harvard Medical School)
Novel docosatrienes and 17S-resolvins generated from docosahexaenoic acid in murine brain, human blood, and glial cells Autacoids in anti-inflammation.
Anti-inflammatory actions of neuroprotectin D1/protectin D1 and its natural stereoisomers: assignments of dihydroxy-containing docosatrienes [Erratum appeared in J Immunol 2006, 176:3843].
Samples and standards of lipid mediators were handled under dimmed light, stored at −80°C in vials purged with nitrogen gas and sealed tightly. Aqueous solutions of DHA and DHA derivatives were freshly prepared (∼4°C) immediately before use.
The separation and analysis of DHA-derived compounds were performed using the liquid chromatography-photodiode array ultra violet detector-linear ion trap mass spectrometer (LC-UV-MS/MS). The LC system was equipped with a chiral column. The mobile phase had a flow-rate of 0.2 mL/min, eluted as B (methanol:H2O:acetic acid = 27:73:0.01) from 0 to 1 minute, ramped from B:methanol 40:60 to B:methanol 20:80 by 50 minutes, ramped to 100% methanol by 55 minutes, and then maintained in 100% methanol. An isotope dilution LC-MS/MS method was used to quantify 14S,21R-diHDHA or mono-HDHAs with 14S,21R-diHDHA-d4 or 14S-hydroxy-DHA-d5 (14S-HDHA-d5) as the internal standard, respectively, which was added to each macrophage sample at the time of collection (10 ng of each standard).
Electrospray mass spectrometric analysis of 5-hydroperoxy and 5-hydroxyeicosatetraenoic acids generated by lipid peroxidation of red blood cell ghost phospholipids.
The UV spectrum was acquired without an internal standard. 14S-HDHA, 14S-HDHA-d5, 14S,21R-diHDHA, or 14S,21R-diHDHA-d4 from the effluent of the chiral LC coupled to the mass spectrometer was analyzed on the basis of selected MS/MS ion chromatograms at a mass-to-charge ratio (m/z) 205 of MS/MS 343, m/z 205 of MS/MS 348, m/z 253 of MS/MS 359, or m/z 253 of MS/MS 363, respectively. The amount of 14S-HDHA or 14S,21R-diHDHA in each macrophage sample was then quantified based on its LC peak area ratio relative to that of its deuterated internal standard 14S-HDHA-d5 or 14S,21R-diHDHA-d4, respectively, using a calibration curve obtained from a series of diluted standards using stable isotope dilution LC-MS/MS method.
Electrospray mass spectrometric analysis of 5-hydroperoxy and 5-hydroxyeicosatetraenoic acids generated by lipid peroxidation of red blood cell ghost phospholipids.
14S,21R-diHDHA and 14S,21S-diHDHA or their deuterated isotopomers were generated by incubating 14S-HDHA (50 μg) or 14S-HDHA-d5 (50 μg), respectively, with 1 nmol human cytochrome P450 kit (h-P450, containing recombinant human CYP2E1, NADPH cofactor, and buffer salts in the company's proprietary ratios; Codexis-BioCatalytics, Pasadena, CA) (30°C, 12 hours). Under the same conditions, 14R,21R-diHDHA and 14R,21S-diHDHA were generated from 14R-HDHA by h-P450.
14S-HDHA or 14S-HDHA-d5 were prepared by incubating DHA or DHA-d5 (50 μg each), respectively, with porcine L-12-LOX (1000 U) in buffer containing 0.1 mol/L Tris at pH 7.4 for 20 minutes at 37°C. At the end of the incubation, 3 mg of NaBH4 was added.
Additionally 14S-HDHA and 14R-HDHA were prepared from racemic 14-S/R-HDHAs (Cayman Chemical) by the above chiral LC-UV-MS/MS-based separation. The DHA and 14S-HDHA used to make 14S-HDHA and 14S,21R-diHDHA, respectively, were not less than 98% pure (without detectable oxidation artifacts), as analyzed by chiral LC-UV-MS/MS.
The biosynthesis of 14S,21R-diHDHA and intermediates was assayed by incubating (20 minutes) purified macrophages, PMNs, lymphocytes (T cells/B cells 1:1), fibroblast cells, or DMVECs (1 × 107 cells/each type), as well as platelets (1 × 108 cells), in PBS containing DHA (3 μmol/L) or 14S-HDHA-d5 (1 μmol/L); this was followed by stimulating the cells (37°C, 1 hour) with TNFα (10 ng/mL), IL-1β (10 ng/mL), and lipopolysaccharide (100 ng/mL). LC-MS/MS analysis indicated that DHA (free acid) in full-thickness murine skin wounds can reach levels as high as 1 to 9 μmol/kg. Thus, 3 μmol/L DHA is within the concentration range found in the pathophysiological condition of skin wounds. TNFα and IL-1β are generated in tissue injury. LPS treatment simulates the potential for infection in wounds.
TNFα, IL-1β, and LPS increase the expression of phospholipase A2, which releases DHA free acid from membranes and/or other parts of the cells. They also likely enhance 12/15-lipoxygenase-like activity,
Each incubation was extracted for analysis and preparative separation of 14S,21R-diHDHA or its biosynthetic intermediates. The pH of each incubation was adjusted to pH 4.5 with HCl (1 mol/L, on ice) and then the incubation was mixed with three volumes of ice-cold methanol (sample:methanol = 1:3 v/v). Each mixture was sonicated and centrifuged (4500 × g for 15 minutes at ∼4°C). The pellet was extracted with methanol twice. The supernatants for each sample were pooled together and adjusted to 10% methanol with water, and partially purified via C18 solid-phase extraction (500 mg Bond Elut LRC-C18 INT sorbent; Agilent Technologies, Santa Clara, CA). The methanol contained 0.1% butylated hydroxytoluene to suppress auto-oxidation.
Electrospray mass spectrometric analysis of 5-hydroperoxy and 5-hydroxyeicosatetraenoic acids generated by lipid peroxidation of red blood cell ghost phospholipids.
The final extracts were reconstituted into methanol for chiral LC-UV-MS/MS separation or analysis. For chiral LC preparative separation and fraction collection of compounds generated from incubation, a fraction (10%) of LC flow was split to MS/MS for detection. The fraction of 14S,21R-diHDHA was dried completely with N2 gas and reconstituted in ethanol. The purified 14S,21R-diHDHA and 14S,21R-diHDHA-d4 were found to be not less than 95% pure by LC-UV-MS/MS using neuroprotectin/protectin-D1
Briefly, paired 5-mm circular, full-thickness wounds were made symmetrically along the midline on the dorsal skin of mice. Two days later, 14S,21R-diHDHA (50 ng/wound) or macrophages and 14S,21R-diHDHA-treated macrophages (106 cells/wound) were applied to the wound bed (10 μL/bed) and were injected intradermally at four points (10 μL/site) distributed evenly near the wound edge (50 μL total/wound). A donut-shaped silicone splint was adhered around the wound to minimize wound contraction; the splinting did not influence re-epithelialization.
This model closely parallels human wound healing. The adherent macrophages were incubated with 100 nmol/L 14S,21R-diHDHA in a dish (37°C, 5% CO2, 24 hours), resuspended in RPMI 1640 medium, and used as 14S,21R-diHDHA-treated macrophages.
Wound Healing Analysis
Complete re-epithelialization takes ∼10 days in wounds of db/db mice and ∼7 days in C57BL/6J mice.
To enable quantification of the healing parameters appropriate for our research, re-epithelialization, granulation tissue formation, and angiogenesis were analyzed for wounds at different times for different mice.
Briefly, wounds rimmed with 2 mm of unwounded skin were excised from sacrificed mice. Cryosections (10 μm thick) of wounds from db/db mice (day 8 after wounding) and 12/15-LOX−/− or C57BL/6J mice (day 5 after wounding) were analyzed to determine the epithelial gap and granulation tissue formation after H&E staining. The splint has a constant area, and it was used to normalize the wound sizes
The relative epithelial gap was calculated as a percentage: (advancing epithelial gap/original epithelial gap) × 100. Granulation tissue locates between epidermal layer and subcutaneous layer, and exhibits tight granular morphology. Relative granulation tissue area was calculated as a percentage: (granulation tissue area in treatment/granulation tissue area in control) × 100. Wound vascularity in wounds from db/db mice (day 8 after wounding) and 12/15-LOX−/− or C57BL/6J mice (day 7 after wounding) were assessed by staining vessel endothelial cells with monoclonal rat anti-mouse CD31 antibody followed by goat anti-rat IgG horseradish peroxidase detection kit (BD Biosciences, San Jose, CA), and then by hematoxylin for nuclei. Three to four images per section were captured. Percentage vascularity was calculated as (CD31+ area per field/total wound bed area per field) × 100. Images were captured on a microscope (Zeiss Axio Imager.M1) with an AxioCam MRC camera and Axiovision release 4.6 software. Distance of the epithelial gap, granulation tissue area, and CD31+ area were measured using ImageJ version 1.40 software (NIH, Bethesda, MD).
Macrophage-Conditioned Medium Preparation
Quiescent db/db and db/+ macrophages (3 × 105 cells) were prepared by incubating cells in RPMI 1640 medium containing 25 mmol/L glucose (for 24 hours), and then in fresh RPMI 1640 medium containing 25 mmol/L glucose with or without 100 nmol/L 14S,21R-diHDHA (for 24 hours). The end product, macrophage-conditioned medium (MCM), was collected after centrifugation (3000 × g for 15 minutes at 4°C).
DMVEC Migration and Vasculature Formation
Analysis of DMVEC migration and vasculature formation was performed as described previously.
Briefly, quiescent db/db DMVECs were resuspended in RPMI 1640 medium with 25 mmol/L glucose and added to the upper chamber of 24-well Transwell supports (8-μm pore; BD Biosciences) at 1 × 105 cells/well; RPMI 1640 medium with 25 mmol/L glucose or MCM was added to the lower chamber for migration (37°C, 5% CO2, 4 hours). For vasculature formation, quiescent db/db DMVECs were cultured on Matrigel substrate in a 96-well culture plate (BD Biosciences) at 2 × 104 cells/well in RPMI 1640 medium with 25 mmol/L glucose or MCM (37°C, 5% CO2, 24 hours). Images were captured on a microscope (Nikon TS100) with a CCD camera and software (NIS-Elements F version 2.30).The vasculature length was measured using NIH ImageJ version 1.40 software.
Bio-Plex Protein Array and Enzyme-Linked Immunosorbent Assay
Vascular endothelial growth factor (VEGF) and platelet-derived growth factor BB (PDGF-BB) secreted by macrophages treated with or without 14S,21R-diHDHA for 24 hours were quantified using a Bio-Plex kit (Bio-Rad, Hercules, CA). In another experiment, quiescent db/db macrophages were treated with 14S,21R-diHDHA with or without LPS for 24 hours. IL-10 in the supernatants was quantified using an enzyme-linked immunosorbent assay kit (Fisher Scientific, Rockford, IL).
Western Blot Analysis
The Western blot experiments were conducted as previously described.
p38 and ERK1/2 coordinate cellular migration and proliferation in epithelial wound healing: evidence of cross-talk activation between MAP kinase cascades.
Briefly, The db/db, db/+,12/15-LOX−/− and C57BL/6J peritoneal macrophages were analyzed for 12/15-LOX expression. In another experiment, quiescent macrophages were cultured in RPMI 1640 medium containing 5.5 mmol/L glucose (for db/+, 12/15-LOX−/−, and C57BL/6J) or 25 mmol/L glucose (for db/db) with or without 14S,21R-diHDHA and/or IL-4 for 24 hours. Macrophages were analyzed for arginase-1 expression.
Western blot analysis was performed by loading 30 μg total protein/lysed sample to a well in a 4% to 15% Tris-HCl SDS-PAGE gel (Bio-Rad). The electrotransferred protein bands were stained by primary antibodies and then by fluorescent-labeled secondary antibodies (LI-COR, Lincoln, NE), and finally were quantified using a LI-COR Odyssey imaging system.
Reactive Oxygen Species Generation Analysis in Macrophages
Analysis of reactive oxygen species (ROS) generation was performed as described previously.
Briefly, The db/db and db/+ macrophages were cultured in RPMI 1640 medium containing 0.1% fetal bovine serum and 5.5 or 25 mmol/L glucose, with or without 100 nmol/L 14S,21R-diHDHA, for 3 days. Medium was changed every day. ROS generation in cells was assessed using the probe 2,7-dichlorofluorescein diacetate (DCFH-DA, 10 μmol/L; Sigma-Aldrich). Cells were cultured with DCFH-DA in RPMI 1640 medium for 30 minutes and DCFH-DA fluorescence was measured using an excitation wavelength of 480 nm and an emission wavelength of 535 nm.
Statistical Analysis
Statistical analysis was performed using one-way analysis of variance followed by Fisher's least significant difference post hoc comparison. P < 0.05 was considered significant.
Results
14S,21R-diHDHA Formation Is Decreased in Macrophages of Diabetic db/db Mice
To determine effects of diabetes on macrophage production of the prohealing, DHA-derived lipid mediator 14S,21R-diHDHA
and the biosynthetic mechanism, we analyzed 14S,21R-diHDHA from murine macrophages using a LC-UV-MS/MS (Thermo Fisher Scientific, Waltham, MA) equipped with a chiral column (Chiralpak-IA, 150 mm × 2.1 mm × 5 μm; Chiral Technologies, West Chester, PA). Murine macrophages were isolated from the peritoneal cavity, and 14S,21R-diHDHA was generated by macrophages in both db/db and db/+ mice (Figure 1). The tandem MS/MS spectrum of peak III (Figure 1A) possessed ions at m/z 359 [M-H]−, 341 [M-H-H2O]−, 323 [M-H-2H2O]−, 297 [M-H-CO2-H2O]−, and 279 [M-H-CO2-2H2O]−, demonstrating two hydroxyl groups and one carboxyl group in the molecular ion m/z 359. The 14-hydroxyl was demonstrated by ions m/z 205, 233, 189, and 161. The 21-hydroxyl was demonstrated by ions m/z 271 and 253, which involved the cleavage of C20-C21 bond in the molecular ion at m/z 359 [M-H]− and loss of CH(O)CH3 (44 atomic units) derived from the terminal group of carbons at the tail of the molecular ion (Figure 1, A and B). Therefore, the peak III compound was 14,21-diHDHA.
Figure 1Differences between macrophages from diabetic db/db and nondiabetic db/+ mice in producing 14S,21R-diHDHA and its biosynthetic intermediates 14S-HDHA and 21R-diHDHA. Samples were analyzed by Chiralpak-IA-based chiral LC-UV-MS/MS. A: 14S,21R-diHDHA identification (peak III): MS/MS spectrum, UV spectrum (top left), MS/MS-based structure elucidation (top right), and chiral LC chromatogram (top middle) with diastereomers at peak I, II, and IV as annotated. B: MS/MS fragmentation pathways for 14S,21R-diHDHA. C: Amount of 14S,21R-diHDHA, 14S-HDHA, or 21R-HDHA generated by macrophages (ng/107 cells). Data are reported as means ± SEM (n = 3). *P < 0.05.
There are four 14,21-diHDHA diastereomers resulting from four different combinations of R/S configurations of C14 and C21 (14R,S and 21R,S). The chiral LC retention time of 14,21-diHDHA in peak III (Figure 1A, top middle) matched that of 14S,21R-diHDHA, which had different retention time than the other three 14,21-diHDHA diastereomers.
Therefore, peak III is 14S,21R-diHDHA. The formation of 14S,21R-diHDHA was decreased in diabetic db/db macrophages, compared with nondiabetic db/+ macrophages (Figure 1C), based on the chiral LC-MS/MS quantification.
To obtain biochemical insights into the impairment of 14S,21R-diHDHA biosynthesis in diabetes, we studied the biosynthetic intermediates 14-HDHA and 21-HDHA in db/db and db/+ macrophages.
A significantly higher level of 14S-HDHA was generated by db/db macrophages, compared with db/+ macrophages (Figure 1C). This coincided with up-regulated expression of L-12-LOX (see Supplemental Figure S2A at http://ajp.amjpathol.org), which is involved in DHA 14S-hydroxylation. However, the 21R-HDHA level was significantly lower in db/db macrophages than in db/+ macrophages (Figure 1C), which suggests that diabetes may impede 14S,21R-diHDHA biosynthesis by interfering with 21-hydroxygenation.
The results from our chiral LC-MS/MS analysis show that PMNs, platelets, and skin cells (including fibroblasts and DMVECs) did not produce 14S,21R-diHDHA; whereas lymphocytes (T cells and B cells) produced a much lesser amount of 14S,21R-diHDHA, compared with macrophages, and platelets produced the intermediate 14S-HDHA for the biosynthesis of 14S,21R-diHDHA (Figure 1A; see also Supplemental Figure S3 at http://ajp.amjpathol.org). These data suggest that macrophages produce 14S,21R-diHDHA in wounds, especially at days 2 to 4 after wounding, when macrophages are the major leukocytes in wounds.
These results motivated us to further study the relation of decreased generation of 14S,21R-diHDHA to impaired prohealing functions of diabetic macrophages.
14S,21R-diHDHA Recovers Impaired Prohealing Functions of Macrophages of Diabetic db/db Mice
To determine whether the impaired prohealing functions of db/db macrophages is associated with decreased 14S,21R-diHDHA generation, db/+, db/db macrophages and 14S,21R-diHDHA-treated db/db macrophages were injected into wounds of db/db mice at day 2 after wounding, when considerable numbers of endogenous macrophages are recruited in the natural course of wound healing.
Compared with vehicle controls, the administration of db/+ macrophages or db/db macrophages significantly promoted re-epithelialization by decreasing the epithelial gap and increasing formation of granulation tissue (Figure 2, A–C). In addition, db/db macrophages showed impaired functions in accelerating wound healing, compared with db/+ macrophages (Figure 2, A–C), which is consistent with a previous report.
14S,21R-diHDHA-treated db/db macrophages considerably decreased the epithelial gap and increased granulation tissue area, compared with db/db macrophage treatment (Figure 2, A–C). These results provide evidence that reduced generation of 14S,21R-diHDHA in diabetic db/db macrophages contributed to impaired prohealing functions, and that these functions were restored by 14S,21R-diHDHA treatment.
Figure 214S,21R-diHDHA recovers impaired prohealing functions of db/db macrophages (Mfs) in cutaneous wounds. Splinted excisional wounds were created in the skin of db/db mice. At day 2 after wounding, db/db macrophages, 14S,21R-diHDHA-treated db/db macrophages, db/+ macrophages (106 cells/wound), or RPMI 1640 vehicle control medium was administered to wounds. Wounded skin was analyzed at day 8 after wounding. A: Representative micrographs of H&E-stained cryosections of wounds. A long black line marks the epithelial gap; dashed lines circumscribe areas of granulation tissue (G). B: Relative epithelial gap (n = 8). C: Relative granulation tissue area (n = 8). D: Representative microphotographs of CD31+ blood vessels in wounds. E: Quantification of wound vascularity (n = 18) at day 8 after wounding. Data are reported as means ± SEM. *P < 0.05 versus control; †P < 0.05 versus db/+ macrophage treatment; ‡P < 0.05 versus db/db macrophage treatment. Scale bars, 500 μm (A); 100 μm (D).
Our data demonstrate that wounds of db/db mice injected with db/+ or db/db macrophages exhibited increased vascularity, compared with wounds of control animals (Figure 2, D and E). In addition, significantly more CD31+ vessels were found in the wound beds injected with 14S,21R-diHDHA-treated db/db macrophages than with untreated db/db macrophages (Figure 2, D and E). Indeed, untreated db/db macrophages were less efficient than db/+ macrophages in promoting angiogenesis (Figure 2, D and E). Thus, 14S,21R-diHDHA counteracted diabetes impairment and recovered macrophage prohealing functions, including enhanced angiogenesis.
To further understand the cellular mechanism by which 14S,21R-diHDHA rescues impaired macrophage prohealing functions in diabetes, we studied diabetic db/db DMVEC migration and vasculature formation in MCM. Media conditioned by db/+ and db/db macrophages considerably increased db/db DMVEC migration and enhanced vasculature formation, compared with controls (non-MCM) (Figure 3, A–D). The db/db MCM was significantly less effective in promoting db/db DMVEC migration and vasculature formation, compared with db/+ MCM, demonstrating that macrophage functions in angiogenesis were impaired in diabetic mice. When conditioned medium from 14S,21R-diHDHA-treated db/db macrophages was used, db/db DMVEC migration and vasculature formation was significantly enhanced, relative to db/db MCM (Figure 3, A–D). These data indicate that 14S,21R-diHDHA enhanced functions of db/db macrophages in angiogenesis through promoting DMVEC migration and vasculature formation.
Figure 314S,21R-diHDHA treatment restores db/db macrophage function in cellular processes of angiogenesis in db/db DMVECs. Representative micrographs of DMVEC migration (A) and vasculature formation (B) in RPMI 1640 vehicle control medium or in MCM. Data are quantified as migrated DMVECs per field (n = 15) (C) and as vasculature length per field (n = 4) (D). VEGF (E) and PGDF-BB (F) secreted from db/db macrophages, 14S,21R-diHDHA-treated db/db macrophages, or db/+ macrophages were measured using a Bio-Plex protein array (n = 3). Scale bars = 100 μm. Data are reported as means ± SEM. *P < 0.05 versus control; †P < 0.01 versus db/+ MCM; ‡P < 0.05 versus db/db MCM.
We postulated, therefore, that 14S,21R-diHDHA enhances the functions of db/db macrophages in angiogenesis, in part, by increasing their angiogenic cytokine production. The results demonstrated that lesser amounts of VEGF and PDGF-BB were released by db/db macrophages than by db/+ macrophages (Figure 3, E and F), consistent with impaired functions of db/db macrophages. This impairment was diminished by 14S,21R-diHDHA, which activated db/db macrophages to secrete significantly more VEGF and PDGF-BB under simulated hyperglycemia (Figure 3, E and F). Taken together, these results suggest that 14S,21R-diHDHA treatment restored the impaired prohealing functions of db/db macrophages in angiogenesis by enhancing db/db macrophage paracrine capacity.
To further confirm that endogenous 14S,21R-diHDHA positively regulated macrophage prohealing functions, 12/15-LOX−/− macrophages were compared with wild-type C57BL/6J macrophages. We found that 12/15-LOX−/− macrophages did not produce 14S,21R-diHDHA, whereas macrophages from C57BL/6J, BALB/c, db/db, or db/+ mice did
in 14S,21R-diHDHA biosynthetic pathways and 12/15-LOX could not be detected in 12/15-LOX−/− macrophages. After injection into wounds, C57BL/6J macrophages significantly promoted re-epithelialization with a decreased epithelial gap and promoted granulation tissue formation, as shown by an augmented granulation tissue area, compared with vehicle control (Figure 4, A–C). 12/15-LOX−/− macrophages had no effect on re-epithelialization, but they did cause a slight increase in granulation tissue formation (Figure 4, A–C). Thus, 12/15-LOX−/− macrophages exhibited impaired functions in accelerating wound healing, compared with C57BL/6J macrophages (see Supplemental Figure S2 at http://ajp.amjpathol.org). After treatment with 14S,21R-diHDHA, 12/15-LOX−/− macrophages significantly accelerated wound healing. 14S,21R-diHDHA-treated 12/15-LOX−/− macrophages decreased the epithelial gap and increased granulation tissue area, compared with untreated 12/15-LOX−/− macrophages (Figure 4, A–C).
Figure 414S,21R-diHDHA treatment restores macrophage prohealing functions impaired by 12/15-LOX deficiency. Splinted excisional wounds were created in the skin of C57BL/6J mice. Wounds were then treated with vehicle control, 12/15-LOX−/− macrophages, 14S,21R-diHDHA-treated 12/15-LOX−/− macrophages, or C57BL/6J macrophages (106 cells/wound) at day 2 after wounding. A: Representative micrographs of H&E-stained cryosections of wounded skin collected at day 5 after wounding. A long black line marks the epithelial gap; dashed lines circumscribe areas of granulation tissue (G). B: Relative epithelial gap (n = 8). C: Relative granulation tissue area (n = 8). D: Representative microphotographs of CD31+ blood vessels in wounds. E: Quantification of wound vascularity (n = 18) at day 7 after wounding. Data are reported as means ± SEM. *P < 0.05 versus control; †P < 0.05 versus C57BL/6J macrophage treatment; ‡P < 0.05 versus 12/15-LOX−/− macrophage treatment. Scale bars, 500 μm (A); 100 μm (D).
The effect of decreased 14S,21R-diHDHA production on macrophage functions for angiogenesis was further addressed. We found that wounds injected with 12/15-LOX−/− macrophages did not exhibit increased vascularity, relative to the control, and also had much less vascularity relative to C57BL/6J macrophage treatment (Figure 4, D and E). However, significantly higher wound vascularity was found in wounds injected with 14S,21R-diHDHA-treated 12/15-LOX−/− macrophages than in wounds injected with untreated 12/15-LOX−/− macrophages (Figure 4, D and E). Thus, impaired prohealing functions of 12/15-LOX−/− macrophages are associated with decreased biosynthesis of 14S,21R-diHDHA.
Given that i) macrophages are the main leukocytes, ii) there are relatively fewer PMNs in wounds at days 2 to 3 after wounding, and iii) macrophages produce 14S,21R-diHDHA, we wanted to clarify whether application of 14S,21R-diHDHA can ameliorate wound healing after macrophages become the dominant leukocytes. Wound healing in 12/15-LOX−/− mice was compared with that in wild-type C57BL/6J mice. Previous data showed decreased 14S,21R-diHDHA formation in wounds of 12/15-LOX−/− mice, compared with wounds of C57BL/6J mice.
In addition, correlative with diminished formation of 14S,21R-diHDHA, histological analysis showed that 12/15-LOX−/− mice exhibited impaired cutaneous wound healing, compared with C57BL/6J mice, as well as delayed re-epithelialization (Figure 5, A and B), decreased granulation tissue formation (Figure 5, A and C), and decreased vascularity (Figure 5, D and E). Application of 14S,21R-diHDHA into wounds of 12/15-LOX−/− mice at day 2 after wounding significantly promoted re-epithelialization and granulation tissue formation, as well as vascularity (Figure 5). These data suggest that 14S,21R-diHDHA may ameliorate wound healing through modulating macrophage functions.
Figure 514S,21R-diHDHA ameliorates impaired cutaneous wound healing in 12/15-LOX−/− mice. Splinted excisional wounds were made in the skin of 12/15-LOX−/− mice and C57BL/6J mice. Wounds were then treated with 14S,21R-diHDHA (50 ng/wound) at day 2 after wounding. Wounds of C57BL/6J mice received only vehicle. A: Representative micrographs of H&E-stained cryosections of wounds at day 5 after wounding. A long black line marks the epithelial gap; dashed lines circumscribe areas of granulation tissue (G). B: Relative epithelial gap (n = 8). C: Relative granulation tissue area (n = 8). D: Representative microphotographs of CD31+ blood vessels in wounds. E: Quantification of wound vascularity (n = 18) at day 7 after wounding. Data are reported as means ± SEM. *P < 0.05 versus C57BL/6J mice; †P < 0.05 compared 12/15-LOX−/− mice treated with vehicle control. Scale bars, 500 μm (A); 100 μm (D).
We therefore investigated whether 14S,21R-diHDHA treatment reduces ROS generation. A high concentration of glucose (25 mmol/L) induced ROS generation in db/+ macrophages. A low glucose concentration (5.5 mmol/L) resulted in slightly decreased ROS generation in db/db macrophages (Figure 6A). The db/db macrophages produced more ROS, compared with db/+ macrophages (Figure 6A). 14S,21R-diHDHA treatment decreased ROS generation by db/+ and db/db macrophages in high-glucose conditions (Figure 6A). Thus, 14S,21R-diHDHA may affect macrophage function by reducing ROS generation under high-glucose conditions.
Figure 614S,21R-diHDHA reduces ROS generation induced by high concentrations of glucose and increases IL-10 expression in db/db macrophages under LPS stimulation. A: ROS generation. db/db and db/+ macrophages were cultured in 5.5 mmol/L or 25 mmol/L glucose and treated with 14S,21-diHDHA. ROS generation in cells was assessed using the probe DCFH-DA (n = 4). B: IL-10 expression. IL-10 secreted from db/db macrophages treated with 14S,21R-diHDHA, LPS, or 14S,21R-diHDHA plus LPS was measured using an IL-10 enzyme-linked immunosorbent assay kit (n = 3). C: Arginase-1 expression. Quiescent db/db and db/+ macrophages were cultured for 24 hours in RPMI 1640 medium with 14S,21R-diHDHA, IL-4, or both IL-4 and 14S,21R-diHDHA. Macrophage lysates were analyzed by Western blot. Arginase-1 quantities relative to β-actin were measured based on optical densities of the gel bands and were normalized to that of nontreated db/db macrophages (n = 3). 14S,21R-diHDHA, 100 nmol/L; IL-4, 10 ng/mL; LPS, 100 ng/mL. Data are reported as means ± SEM. *P < 0.05.
LPS stimulation resembles the potential of infection to wounds. To determine the effect of 14S,21R-diHDHA on IL-10 expression, we treated db/db macrophages with 14S,21R-diHDHA alone, 14S,21R-diHDHA plus LPS, LPS alone, or vehicle control. We found that 14S,21R-diHDHA significantly promoted LPS-induced IL-10 generation by macrophages, although 14S,21R-diHDHA alone did not affect IL-10 generation by db/db macrophages (Figure 6B). These results indicate that 14S,21R-diHDHA enhanced the anti-inflammatory functions of db/db macrophages. We also assessed the effect of 14S,21R-diHDHA on IL-4-induced alternative activation,
which results in a prohealing phenotype in macrophages and characteristically involves arginase-1 expression. 14S,21R-diHDHA did not affect this type of activation of macrophages under the in vitro conditions used in the present study (Figure 6C).
Discussion
Wound healing is a dynamic process that is characterized by inflammation, tissue regeneration, and remodeling.
Our data show that macrophages possess a greater capability to produce 14S,21R-diHDHA, compared with several other typical cell types usually present in wounds. 14S,21R-diHDHA formation was reduced in diabetic db/db macrophages, compared with nondiabetic db/+ macrophages (Figure 1), which suggests that diabetes dysregulation of macrophage biosynthesis of 14S,21R-diHDHA could contribute to diabetes impairment of 14S,21R-diHDHA production in diabetic wounds.
was elevated in diabetic wounds. Nevertheless, the decreased 21R-HDHA formation (Figure 1) suggests that diabetes attenuates 21R-hydroxylation, which could contribute to reduced 14S,21R-diHDHA generation. The 21R-hydroxylation is likely to be catalyzed by a specific P450
Cytochrome P-450 4F18 is the leukotriene B4 omega-1/omega-2 hydroxylase in mouse polymorphonuclear leukocytes: identification as the functional orthologue of human polymorphonuclear leukocyte CYP4F3A in the down-regulation of responses to LTB4.
Diabetic hyperglycemia and accompanying oxidative stress might affect biosynthesis of 14S,21R-diHDHA by altering P450 or disrupting the transfer of biosynthetic intermediates
between 12-LOX and P450 within or between cells. 12/15-LOX−/− macrophages are deficient in 14S,21R-diHDHA generation; we therefore used 12/15-LOX−/− macrophages as negative control to study the role of this autacoid-type mediator in macrophage functions. Although platelets are unable to produce 14S,21R-diHDHA, and lymphocytes barely produce this mediator, both cell types produce the intermediate 14S-HDHA of 14S,21R-diHDHA biosynthesis (see Supplemental Figure S3 at http://ajp.amjpathol.org). The 14S-HDHA produced by platelets or lymphocytes was likely transported to macrophages and there converted to 14S,21R-diHDHA through the transcellular biosynthesis mechanism for lipid mediators.
They are activated to phagocytose apoptotic PMNs, debris, and bacteria, and then coordinate later events of healing, including re-epithelialization, granulation tissue formation, and angiogenesis.
Impaired macrophage functions are among the crucial defects responsible for delayed wound healing in diabetics. Compared with normal macrophages, diabetic macrophages display detrimental pathophysiological responses, such as expressing low levels of angiogenic cytokines,
including reduced expression of VEGF and PDGF-BB (Figure 3). Moreover, the association of decreased formation of 14S,21R-diHDHA with impaired prohealing functions of db/db macrophages indicates that impaired functions of diabetic macrophages also includes impaired formation of this autacoid. This association was further confirmed by the fact that 12/15-LOX−/− macrophages could not produce 14S,21R-diHDHA and exhibited attenuated promotion of healing that could be restored by 14S,21R-diHDHA treatment (Figure 4). Impaired functions of 12/15-LOX−/− macrophages may not only result from halted 14S,21R-diHDHA formation, but may also stem from other detrimental effects of 12/15-LOX deletion; however, 14S,21R-diHDHA treatment can restore macrophage functions, indicating that 14S,21R-diHDHA can redeem those effects. Additionally, 12/15-LOX−/− mice exhibiting delayed wound healing. The application of 14S,21R-diHDHA at day 2 after wounding, when macrophages significantly infiltrate into wounds, counteract such impairment (Figure 5). This further suggests the reparative roles of 14S,21R-diHDHA via action on macrophages in wounds. However, the paracrinic action of macrophage 14S,21R-diHDHA on other kinds of cells in wounds cannot be excluded and should be investigated in future studies.
During the inflammatory phase of wound repair, ROS generated by PMNs and macrophages destroy harmful microorganisms and stimulate expression of some growth factors.
However, overproduction of ROS leads to excessive oxidative stress and results in cellular dysfunction and functional alterations of DNA, lipids, and proteins.
Wound inflammation in diabetic ob/ob mice: functional coupling of prostaglandin biosynthesis to cyclooxygenase-1 activity in diabetes-impaired wound healing.
High glucose concentrations induce significant increases in ROS production in macrophages (Figure 6). Thus, diabetic hyperglycemia might stimulate macrophages to produce excessive ROS, which cause macrophage dysfunction. Moreover, the suppression of oxidative stress has been shown to reduce the abnormally high levels of proinflammatory cytokines in diabetic macrophages.
14S,21R-diHDHA was found to decrease ROS generation (Figure 6), providing another possible means by which 14S,21R-diHDHA treatment mediates recovery of macrophage prohealing functions in diabetes.
Heterogeneity of macrophages has recently been proposed. Alternatively activated macrophages, as distinct from classically activated macrophages, promote wound healing, and macrophages are capable of switching phenotypes under different stimuli.
However, 14S,21R-diHDHA did not affect arginase-1 expression under IL-4 stimulus (Figure 6), a property of alternatively activated macrophages, suggesting that 14S,21R-diHDHA does not act on the macrophage switching to an alternative phenotype. 14S,21R-diHDHA increased IL-10 expression by db/db macrophages under LPS stimulation (Figure 6), suggesting that the recovery of diabetic prohealing functions by 14S,21R-diHDHA may involve augmenting IL-10 generation. Of note, db/db macrophages expressed more arginase-1, compared with db/+ macrophages (Figure 6). Arginase-1 effectively competed with nitric oxide synthase for arginine and converted arginine to urea and ornithine, thus preventing the overproduction of nitric oxide.
Increased expression of arginase-1 and IL-10 in db/db macrophages, compared with db/+ macrophages, may be an adaptive response under the stress of diabetes.
The DHA-derived lipid mediator 14S,21R-diHDHA acts as an macrophage-produced autacoid that promotes the prohealing functions of macrophages (Figure 7). Reduced formation of 14S,21R-diHDHA in db/db macrophages is associated with their impaired prohealing functions. 14S,21R-diHDHA treatment of macrophages reduced ROS generation, promoted IL-10 generation, and enhanced diabetic macrophage prohealing functions. These observations provide mechanistic insights into wound healing and point toward possible treatment of wounds in diabetic patients using autologous macrophages treated with 14S,21R-diHDHA or related compounds.
Figure 7Impairment of prohealing functions of diabetic db/db macrophages is associated with reduced 14S,21R-diHDHA formation. Diabetes decreases the generation of 14S,21R-diHDHA by macrophages. Treatment with 14S,21R-diHDHA rescues diabetes-impaired macrophage promotion of wound re-epithelialization, granulation tissue formation, and vascularization. The mechanism of 14S,21R-diHDHA action on macrophages includes reducing ROS generation under hyperglycemia, increasing IL-10 expression, increasing the secretion of VEGF and PDGF-BB, and enhancing the macrophage promotion of endothelial cell migration and vasculature formation.
We thank Drs. Nicolas G. Bazan and Haydee E.P. Bazan (Neuroscience Center of Excellence at the Louisiana State University Health Sciences Center, New Orleans, LA) for 12/15-LOX knockout mice and also for helpful advice. We thank Dr. Eric C.B. Milner for excellent editing and Dr. Qiang Wu and Jose A. Galindo for technical assistance.
Identification of db/db, db/+, 12/15-LOX-/-, and C57BL/6J macrophages as well as db/db DMVECs. Cultured macrophages and DMVECs were confirmed by immunocytochemical analysis, which demonstrated the expression of the F4/80 in macrophages (A) as well as CD31 and VE-cadherin in DMVECs (B). Scale bar = 100 µm.
Expression of 12/15-LOX in db/db, db/+, 12/15-LOX-/-, and C57BL/6J macrophages (Mf). Cell lysates from db/db, db/+, 12/15-LOX-/-, and C57BL/6J macrophages were analyzed by Western blot using 12/15-LOX and β-actin antibodies. Representative Western blot images of 12/15-LOX and β-actin in macrophages of (A) db/db and db/+ as well as (B) 12/15-LOX-/- and C57BL/6J mice. 12/15-LOX quantities relative to β-actin were measured based on optical densities of the gel bands, and normalized to that db/+ or C57BL/6J macrophages. Data are expressed as mean ± SEM (n = 3). *P < 0.05.
Reprehensive LC-MS/MS chromatograms show that platelets, PMNs, fibroblasts, or DMVECs did not produce 14S,21R-diHDHA and lymphocytes barely produced this mediator whereas platelets or T/B cells can generate the biosynthetic intermediate 14S-HDHA and/or 21R-HDHA. Samples were analyzed by Chiralpak-IA based chiral LC-UV-MS/MS.
References
Falanga V.
Wound healing and its impairment in the diabetic foot.
Anti-inflammatory actions of neuroprotectin D1/protectin D1 and its natural stereoisomers: assignments of dihydroxy-containing docosatrienes [Erratum appeared in J Immunol 2006, 176:3843].
Electrospray mass spectrometric analysis of 5-hydroperoxy and 5-hydroxyeicosatetraenoic acids generated by lipid peroxidation of red blood cell ghost phospholipids.
p38 and ERK1/2 coordinate cellular migration and proliferation in epithelial wound healing: evidence of cross-talk activation between MAP kinase cascades.
Cytochrome P-450 4F18 is the leukotriene B4 omega-1/omega-2 hydroxylase in mouse polymorphonuclear leukocytes: identification as the functional orthologue of human polymorphonuclear leukocyte CYP4F3A in the down-regulation of responses to LTB4.
Wound inflammation in diabetic ob/ob mice: functional coupling of prostaglandin biosynthesis to cyclooxygenase-1 activity in diabetes-impaired wound healing.
00644-4&id=gr1.jpg){kind=link}
00644-4&id=gr2.jpg){kind=link}
00644-4&id=gr3.jpg){kind=link}
00644-4&id=gr4.jpg){kind=link}
00644-4&id=gr5.jpg){kind=link}
00644-4&id=gr6.jpg){kind=link}
00644-4&id=gr7.jpg){kind=link}