Reagents were purchased from Sigma-Aldrich (St. Louis, MO) and Thermo Fisher Scientific (Waltham, MA). Recombinant mouse IFNγ, CCL2, and CCL5 were purchased from Invitrogen Corporation (Carlsbad, CA). Human fibrinogen and thrombin were obtained from Enzyme Research Laboratories (South Bend, IN). Collagen IV, collagen I, and laminin were purchased from Corning (Corning, NY). CEP-BSA (21:1 molar ratio) and CEP-Fg (29:1 molar ratio) were prepared through a Paar-Knorr pyrrole synthesis with (9H-fluoren-9-ylmethyl ester 4,7-dioxo-heptanoic acid) DOHA-Fm as described previously (28). Anti-CD68 mAb was from eBioscience. Polyclonal antibody against the α_D I-domain was described previously (11). The antibody recognizes both human and mouse α_D I-domains and has no cross-reactivity with recombinant human and mouse α_M, α_X, and α_L I-domains. The antibody was isolated from rabbit serum by affinity chromatography using α_D I-domain-Sepharose. Anti-macrophage antibody F4/80 was from eBioscience (San Diego, CA). A polyclonal antibody against CEP and monoclonal IgM anti-CEP antibody were described previously (30, 31). Blocking IgG anti-CEP antibody (Clone 3C9) was generated in Dr. Tatiana Byzova’s laboratory (23).

Wild type (C57BL/6J, stock # 000664) and integrin α_D-deficient (B6.129S7-Itgad^tm1Bll /J, stock # 005258) mice were bought from Jackson Laboratory (Bar Harbor, ME). α_D-deficient have been backcrossed to C57BL/6 for at least ten generations. All procedures were performed according to animal protocols approved by East Tennessee State University IACUC.

Notably, the terms M1 and M2 macrophages are artificial and do not represent all variabilities of macrophage subsets in vivo. We used the term M1 in our manuscript to refer to in vitro polarized macrophages activated by IFN-γ which serve as a model for pro-inflammatory macrophages in vivo. Therefore, peritoneal macrophages from 8-12-week-old mice (WT and α D − / − ) were harvested by lavage of the peritoneal cavity with 5 ml of sterile PBS 3 days after intraperitoneal (IP) injection of 4% thioglycollate (TG; 0.5ml). The cells were washed twice with PBS and resuspended in RPMI media supplemented with 10% FBS, 0.1 mg/ml streptomycin, and 0.1 unit/ml penicillin. The cell suspension was transferred into 100mm Petri dishes and incubated for 2h at 37°C in humidified air containing a 5% CO₂ atmosphere. Nonadherent cells were washed out with PBS, and the adherent macrophages were replenished with complete RPMI media. The macrophages were differentiated to M1 phenotypes by treatment with recombinant mouse interferon-γ (IFN-γ) (100 U/ml, Thermo Fisher) for 4 days. Medium with IFN-γ was changed every 2 days or as required. Macrophages were dissociated from the plates using 5mM EDTA in PBS and used for the experiments thereafter.

The HEK293 cells stably expressing human integrin α_Dβ₂ were generated as described previously (11) and maintained in DMEM/F-12 (Invitrogen) supplemented with 10% FBS, 2 mM glutamine, 15 mM HEPES, 0.1 mg/ml streptomycin, and 0.1 unit/ml penicillin.

The construct for the α_D I-domain was generated and recombinant protein was isolated as described in our previous paper (23). Briefly, α_D in the active conformation (Pro¹²⁸-Lys³¹⁴) was inserted in the pET15b vector, expressed in E. Coli as a His-tag fusion protein, and purified using affinity chromatography on Ni-chelating agarose (Qiagen Inc., Valencia, CA).

Flow cytometry analysis was performed to assess the expression of integrin α_D on mouse peritoneal macrophages and α_Dβ₂-transfected HEK293 cells. Cells were harvested and pre-incubated with 4% normal goat serum for 30 min at 4°C, then 2x10⁶ cells were incubated with a specific antibody for 30 min at 4°C. Non-conjugated antibodies required additional incubation with Alexa 488 or Alexa 647-donkey anti-mouse IgG (at a 1:1000 dilution) for 30 min at 4°C. Finally, the cells were washed and analyzed using a Fortessa X-20 (Becton Dickinson).

The adhesion assay was performed as described previously (23) with modifications. Briefly, 96-well plates (Immulon 2HB, Cambridge, MA) were coated with different concentrations of fibrinogen, collagen IV, collagen I, or laminin for 3 h at 37°C. The wells were post-coated with 0.5% polyvinyl alcohol for 1 h at room temperature. Mouse peritoneal macrophages or HEK293 cells transfected with integrin α_Dβ₂ were labeled with 10 µM Calcein AM (Molecular Probes, Eugene, OR) for 30 min at 37°C and washed with HBSS and resuspended in the same medium at a concentration of 1 × 10⁶ cells/mL. α_Dβ₂-transfected HEK 293 cells were also supplemented with 2mM MnCl₂. Aliquots (50 µL) of the labeled cells were added to each well. After 30 minutes of incubation at 37°C in a 5% CO₂ humidified atmosphere, the nonadherent cells were removed by washing with PBS. The fluorescence was measured in a Synergy H1 fluorescence plate reader (BioTek, Winooski, VT), and the number of adherent cells was determined from a labeled control.

Glass coverslips were coated with fibrinogen (20 µg/ml), Collagen IV (5 μg/ml) or CEP-BSA (750 μM) for 3 hours at 37°C and post-coated with 0.5% polyvinylpyrrolidone for 1 hour at 37°C. To determine cell spreading, inflammatory macrophages isolated from the WT and α D − / − mice were allowed to adhere to coverslips for 1 hour at 37°C. Macrophages were fixed with 4% paraformaldehyde, permeabilized with 0.1% Triton X-100, and incubated with Phalloidin-iFluor 555 and DAPI. Fluorescent images were obtained with EVOS FL auto cell fluorescent imaging system using 20x objective with multi-frame automatic setup (9-fields). The cell areas were calculated using Imaris 8.0 software.

Migration assay was performed as described previously (26). Polarized M1 macrophages were labeled with PKH26 red fluorescent dye. Cell migration assay was performed for 48 hours at 37°C in 5% CO₂ in a sterile condition. An equal number of WT macrophages was evaluated by cytospin of mixed cells before the experiment and at the starting point before migration. Labeled WT (1.5x10⁵) activated macrophages were plated on the membranes of transwell inserts with a pore size of 8 μm and 6.5 mm in diameter (Costar, Corning, NY) precoated with fibrinogen (Fg). Fibrin gel (100 µl/sample) was generated by mixing 0.75mg/ml Fg containing 1% FBS and 1% P/S and 0.5 U/ml thrombin. In some samples 9 µM CEP was incorporated into the gel during polymerization. 30 nM of MCP-1 was added on top of the gel to initiate the migration. Migrating cells were detected by Leica Confocal microscope (Leica-TCS SP8), and the results were analyzed and reconstructed using IMARIS 8.0 software.

The binding parameters of the interaction between the α_D I-domain with CEP or fibrinogen were determined using an Octet K2 instrument (FortéBio, Sartorius Group). CEP-BSA or fibrinogen was immobilized on the Amine Reactive Second-generation (AR2G) biosensor using the amine coupling kit according to the manufacturer’s protocol. Different concentrations of the active forms of α_D I-domain (50-1000 nM) were applied in the mobile phase, and the association between the immobilized and flowing proteins was detected. Experiments were performed in 20 mm HEPES, 150 mm NaCl, 1 mM CaCl₂, 1 mM MgCl₂, 0.02% (v/v) Tween-20, 0.02% BSA, pH 7.5. The protein-coated biosensors were regenerated with 25 mm NaOH and 2M NaCl. Analyses of the binding kinetics were performed using DataAnalysis HT11.0 software (ForteBio). The value of the equilibrium dissociation constant (K_D) was obtained by fitting a plot of response at equilibrium (Req) against the concentration.

Cryosections (10 μm) of aorta roots (from WT and ApoE^-/- mice after five weeks on a western diet) were warmed to room temperature for 30 minutes prior to immunofluorescence staining. Tissue sections were fixed in ice-cold acetone for 10 minutes followed by permeabilizing with 0.2% Tween-20 for 10 min to increase the signal of intracellular binding sites. Tissue sections were washed in PBS and incubated with SuperBlock (PBS) Blocking buffer (Thermo Scientific, Rockford, IL, USA) for 45 min to block nonspecific binding. Tissues were then incubated at 4°C overnight with the primary antibody (rabbit polyclonal anti-CEP and rat anti-mouse CD68 (macrophage marker). After washing several times with PBS, the sections were incubated with Alexa Fluor 488-conjugated donkey anti-rabbit IgG and Alexa Fluor 568-conjugated donkey anti-rat for 1 hour at room temperature. The sections were subsequently washed and sealed. The tissue sections were examined by Leica Confocal microscope (Leica-TCS SP8). Control sections without the primary antibody were also performed at the same time.

Male ApoE-deficient mice were kept on a Western diet for 16 weeks. After that, ApoE^-/- and the same age WT C57BL/6 mice (as the control) were euthanized, their vascular system was perfused with 10 ml PBS, and the aortas were cleaned from the outer layers of connective tissue and removed. The isolated aortas were digested with Collagenase type I, type XI, Hyaluronidase, and DHAase I according to the published protocol (12). The cells were separated by centrifugation. The non-cellular supernatant was high speed centrifugated to exclude aggregates and cell traces, and used for Western Blot analysis.

The same concentration of proteins isolated from ApoE^-/- and WT aortas was loaded on SDS-electrophoresis analyzed by 7% SDS-PAGE electrophoresis and Western blotting. The Immobilon-P membranes (Millipore) were incubated with rabbit polyclonal anti-CEP antibody, followed by incubation with goat anti-rabbit secondary antibody conjugated to horseradish peroxidase and developed using enhanced SuperSignal Chemiluminescent Substrate (Pierce).

The evaluation of CEP-modified proteins was performed as described previously (23). Collagen I, collagen IV, and laminin were coated on a 96-well plate at a concentration of 20 μg/ml overnight at 4°C. Wells were post-coated with 0.5% polyvinyl alcohol for 1 h at 37°C. 20 μM DHA in 20mM Hepes, 150 mM NaCl, 0.01% H₂O₂, and 1mM CaCl₂ were added to the wells and incubated in an oxygen-free environment (under argon) for 18 hours at 37°C. After incubation, the plate was washed out with PBS supplemented with 0.05% Tween-20 3 times and incubated with anti-CEP polyclonal antibody (0.9 μg/ml) for 2 hours at 37°C. After washing, wells were incubated with goat-anti-rabbit HRP conjugated antibody for 1 h at 37°C and the binding was developed using TMB-ELISA substrate solution (Pierce). The result was detected by a plate reader using a wavelength of 450 nm.

Statistical analyses were performed using a Student’s t-test or Student’s paired t-tests where indicated in the text using SigmaPlot 13. A value of p<0.05 was considered significant. Error bars represent the SEM of experiments.

We and others have demonstrated the presence of CEP in different inflammatory tissues using immunohistochemistry (23, 30–34). Particularly, anti-CEP antibody detected strong immunostaining in the sections of atherosclerotic lesions of ApoE^-/- mice on a Western diet ( Figure 2A ). However, the direct formation of CEP-protein adducts was shown previously only in the peritoneal cavity, but not in the tissue (23). To demonstrate the link between CEP immunostaining and the formation of CEP-modified proteins in the inflamed tissue during chronic inflammation, we evaluated CEP-protein adduct formation in atherosclerotic lesions ( Figure 2B ). Male ApoE-deficient mice were kept on a Western diet for 16 weeks (n=3). After that, ApoE^-/- and the same age WT C57BL/6 mice (as the control) were euthanized, their vascular system was perfused and aortas were removed. The isolated aortas were digested according to the published protocol (12) and cells were separated by centrifugation. The non-cellular supernatant was high speed centrifugated to exclude aggregates and cell traces, and used for Western Blot analysis.

The same concentration of proteins isolated from ApoE^-/- and WT aortas was loaded on SDS-electrophoresis ( Figure 2B , left panel) and analyzed using an anti-CEP antibody ( Figure 2B , right panel). We found a dramatic difference in anti-CEP staining of proteins isolated from the atherosclerotic lesions of ApoE-deficient and WT mice. While only one weak band was spotted in the sample isolated from WT mice, several intensive bands were detected in the ApoE^-/- mice samples that represent several CEP-modified proteins. Notably, the faded CEP signal from the WT mice corresponds to the concept that the natural oxidation of PUFA in a healthy organism can lead to a low level of adduct formation. We detected such signal previously in commercial fibrinogen that we used in our experiments as a control (23).

To confirm the ability of CEP to form the covalent adduct with ECM proteins, we tested the formation of CEP modification in collagen I, laminin, and collagen IV using in vitro DHA oxidation assay that we developed previously (23). ECM proteins were incubated with DHA and 0.01% H₂O₂ for 18 h at 37°C. After incubation, CEP formation was detected by ELISA with an anti-CEP antibody ( Figure 2C ). We found that DHA oxidation led to the formation of CEP-collagen IV adduct and CEP-laminin adduct, while collagen I was not modified.

These experiments clearly demonstrated that ECM proteins are modified by CEP, and inflamed tissue (e.g. atherosclerotic lesions) can be a source of specific adhesive substrate for macrophages.

Pro-inflammatory, M1-like macrophages, are the major source of inflammatory stimuli during chronic inflammation. We tested how protein modification with CEP affects M1 macrophage adhesion to ECM proteins such as collagen IV, laminin, and fibrinogen (which leaks from blood to the ECM through the damaged endothelial monolayer during inflammation). Different concentrations of collagen IV, laminin, CEP-modified BSA, and BSA alone as a control were immobilized on the 96-well plate and tested as substrates for M1 macrophage adhesion ( Figure 3A ). We found that the adhesion of pro-inflammatory macrophages to CEP-modified BSA dramatically overreached the adhesion of macrophages to natural ECM ligands. Notably, BSA alone does not support the adhesion of macrophages. In a separate experiment, we compared the adhesion of fibrinogen-modified by CEP and native fibrinogen. Clearly, CEP modification dramatically increased macrophage adhesion to fibrinogen ( Figure 3B ).

To verify these results macrophage spreading on CEP, collagen IV, and fibrinogen was evaluated. Cells were labeled with phalloidin-PE, examined by fluorescent microscopy, and analyzed using the ImarisCell module (Imaris 8.0). Corresponding to the adhesion data, the results demonstrate stronger spreading to CEP with larger cell areas and focal adhesion complexes ( Figures 3C, D ).

To evaluate the effect of CEP modification in ECM on cell motility we performed a 3D migration assay by comparing M1 macrophage migration in a polymerized fibrin matrix and polymerized fibrin matrix supplemented with 9 μM CEP. Based on our previous result the migration of M1 macrophages in a fibrin matrix is lower compared to non-activated or M2 macrophages (26). The goal of this experiment was to evaluate the additional effect of CEP on this migration. Cells were plated on one side of the gel and migration was initiated by CCL2 and CCL5 added to the other side of the gel ( Figure 4A ). We found that migration of M1 macrophages was dramatically reduced in fibrin gel supplemented with CEP ( Figures 4B–D ), which corresponds to strong adhesion events and promotion of macrophage retention at the site of chronic inflammation.

To more critically evaluate these results, we tested the morphology of macrophages in a 3D matrix supplemented with CEP. 3D gels were generated using polymerized fibrin and polymerized fibrin/9 uM CEP. The same macrophages were incorporated into both matrixes during polymerization and migration was stimulated with a gradient of CCL2. After 18 hours, the cells inside the gel (at least 80 µM from the gel surface) were analyzed using the confocal microscope. We found that cells in the fibrin matrix present a more elongated shape when compared with the more rounded cells in the matrix with supplemented CEP ( Figure 4E ) The difference between the length-to-diameter ratio was extremely significant (P<0.0001, n=30-35/group) ( Figure 4F ). According to published data, elongated cell morphology in a 3D matrix corresponds to migrating cells.

Therefore, our results demonstrate that modification with CEP affects M1 macrophage adhesion, spreading, and migration-dependent morphology.

The major adhesive receptors for CEP on macrophages are integrins α_Dβ₂ and α_Mβ₂. Integrin α_D is upregulated on M1 macrophages (12), ( Figure 5A ), while α_M maintains the same level of expression. This difference determines the distinct roles of these β₂ integrins in inflammation. While α_Mβ₂ supports macrophage migration and demonstrates some protective function during chronic inflammation (27, 35), integrin α_Dβ₂ inhibits M1 macrophage migration and contributes to the development of chronic inflammation (12, 26). We hypothesized that integrin α_Dβ₂ is critical for M1 macrophage adhesion to CEP-modified proteins. To test this hypothesis, we compared the adhesion of M1 macrophages isolated from the WT and α_D-deficient mice to CEP. The adhesion assay demonstrated a dramatic reduction in macrophage adhesion in α_D-deficient mice ( Figure 5B ). Moreover, macrophage spreading to CEP was significantly affected in α D − / − M1 macrophages, the total cell areas were reduced as well as cell shapes were irregular ( Figures 5C, D ).

Therefore, our data demonstrate that α_D-deficiency significantly reduced adhesion and spreading of M1 macrophages on CEP, which confirmed the critical role of α_Dβ₂ in macrophage adhesion and potential retention in CEP-modified substrate.

Macrophages are a complex system with a variety of adhesive receptors that have overlapping ligand binding properties. To verify the role of CEP as a strong α_Dβ₂ ligand, we tested α_Dβ₂-transfected HEK293 cells ( Figure 6A ). HEK293 cells do not express any receptors that interact with CEP such as integrin α_Mβ₂, CD36, or TLRs (23, 34). Adhesion of α_Dβ₂-transfected HEK293 cells was tested to CEP-BSA versus BSA and CEP-fibrinogen versus fibrinogen. As we expected α_Dβ₂-cells do not adhere to BSA, but demonstrate the strong concentration-dependent adhesion to CEP-BSA. ( Figure 6B , upper).

Fibrinogen is a natural ligand for integrin α_Dβ₂ and a common element of the ECM during inflammation. Markedly, CEP modification significantly improves the adhesive properties of fibrinogen to α_Dβ₂-HEK293 transfected cells similar to the result, which was obtained for M1 macrophages ( Figure 3B ). Notably, it has been shown previously that cell adhesion to fibrinogen achieved maximum at fibrinogen concentration 2-3 µg/ml and is significantly reduced at higher concentrations (10, 36). The mechanism is not known but may depend on the surface-induced lateral aggregation of fibrinogen (37). This result demonstrated that CEP modification in different proteins provides a similar augmentation of cell adhesive properties.

To evaluate a difference in the binding of α_Dβ₂ to CEP and fibrinogen, we isolated α_Dβ₂ ligand-binding motif – I-domain ( Figure 6C ) and examined its binding properties using BioLayer Interferometry. CEP and fibrinogen were immobilized on amino coupling biosensors and tested with different concentrations of α_D I-domain using two approaches such as kinetic titration (single cycle kinetics) ( Figure 6D ) and equilibrium analysis (steady-state analysis) ( Figures 6E, F ). The binding affinity was analyzed using DataAnalysis HT11.0 software (ForteBio). We found that the affinity of α_D binding to CEP (K_D=1.0x10^-7M) is 2-fold stronger than the α_D interaction with fibrinogen (K_D=2.2x10^-7).

Therefore, α_Dβ₂ generates a more powerful adhesion to CEP-modified proteins due to a higher affinity mediated by a negative charge of a carboxyl group in CEP’s structure. This result suggests the unique role of CEP-modified ECM proteins in macrophage adhesion that generates strong α_Dβ₂-dependent adhesion via CEP that can be critical for macrophage retention.