Selectin-targeting glycosaminoglycan-peptide conjugate limits neutrophil-mediated cardiac reperfusion injury

Abstract

Aims

One of the hallmarks of myocardial infarction (MI) is excessive inflammation. During an inflammatory insult, damaged endothelial cells shed their glycocalyx, a carbohydrate-rich layer on the cell surface which provides a regulatory interface to immune cell adhesion. Selectin-mediated neutrophilia occurs as a result of endothelial injury and inflammation. We recently designed a novel selectin-targeting glycocalyx mimetic (termed DS-IkL) capable of binding inflamed endothelial cells. This study examines the capacity of DS-IkL to limit neutrophil binding and platelet activation on inflamed endothelial cells, as well as the cardioprotective effects of DS-IkL after acute myocardial infarction.

Methods and results

In vitro, DS-IkL diminished neutrophil interactions with both recombinant selectin and inflamed endothelial cells, and limited platelet activation on inflamed endothelial cells. Our data demonstrated that DS-IkL localized to regions of vascular inflammation in vivo after 45 min of left anterior descending coronary artery ligation-induced MI. Further, findings from this study show DS-IkL treatment had short- and long-term cardioprotective effects after ischaemia/reperfusion of the left anterior descending coronary artery. Mice treated with DS-IkL immediately after ischaemia/reperfusion and 24 h later exhibited reduced neutrophil extravasation, macrophage accumulation, fibroblast and endothelial cell proliferation, and fibrosis compared to saline controls.

Conclusions

Our findings suggest that DS-IkL has great therapeutic potential after MI by limiting reperfusion injury induced by the immune response.

Graphical Abstract

1. Introduction

Cardiovascular disease remains the leading cause of mortality in the USA. Approximately, 805,000 Americans will be diagnosed with coronary artery disease (CAD), the most common heart disease, this year.¹ Approximately half of those diagnosed with CAD will be susceptible to acute myocardial infarction (MI), as well as complications that can lead to heart failure and sudden cardiac death.² Although initial damage occurs during acute MI due to oxygen deprivation during the ischaemic phase, it is well documented that further injury can result from reperfusion.³ The effects of reperfusion injury (RI) are complex. In the absence of oxygen, ischaemic tissues experience severe disruptions to their homeostasis. A necessary shift from aerobic to anaerobic metabolism results in the accumulation of lactic acid, which contributes to a decline in intracellular pH, an impairment of ion exchange, and an increase in reactive oxygen species (ROS) production,^3,4 subsequently damaging cells within the vicinity. Paradoxically, the restoration of blood flow further contributes to cellular damage, partly due to inefficient oxidative phosphorylation that results in further production of ROS, reductions in nitric oxide (NO), and the recruitment of immune cells.^5,6

Ischaemia/reperfusion-induced damage to vascular endothelial cells (EC) initiates an inflammatory cascade that recruits neutrophils and platelets to the sites of damage.⁵ Increased permeability of the vasculature potentiates RI by allowing the infiltration of immune cells into the tissue. Reperfused ECs experience disruptions to their Ca²⁺ homeostasis and begin to contract,⁷ giving rise to endothelial gaps that facilitate the extravasation of leucocytes being recruited by ROS and cytokines. Though these cells have vital cardioprotective functions, overaccumulation of leucocytes and activated platelets exacerbates myocardial damage, contributing to the overall infarct size.

During RI, ECs shed their glycocalyx, a carbohydrate-rich protective layer that resides on the surface of healthy endothelium.^8,9 The glycocalyx is known to play a role in many cellular processes, including mechanotransduction,¹⁰ microvascular permeability,¹¹ and modulation of inflammatory mediators.¹² Diminution of the endothelial glycocalyx has been shown to contribute to vascular oedema as well as neutrophil and platelet adhesion.^10,13 While the endothelial glycocalyx has been shown to recover from enzyme-induced damage, this recovery was inversely proportional to white blood cell count.¹⁴ The inflammatory response hence plays a critical role in RI with respect to the glycocalyx, with neutrophil influx^15,16 and increased enzyme production^8,17 likely delaying glycocalyx regeneration. Therefore, a molecule that provides a stable, stealthy interface with the capacity to down-regulate platelet activation and neutrophil capture could prevent inflammation-mediated RI.

To address this critical problem, we have designed a novel multivalent selectin-targeting glycosaminoglycan conjugate (termed DS-IkL) that effectively abates neutrophil interactions within damaged regions of the vasculature. Given the elevated levels of matrix metalloproteinases and other proteolytic enzymes within the inflammatory zone, the glycosaminoglycan-derived backbone is conjugated to several selectin-binding peptides consisting of D-amino acids in order to increase the enzymatic stability of the molecule in the harsh inflammatory environment. We directly test the hypothesis that DS-IkL will significantly interfere with neutrophil capture and platelet activation on ECs in vitro and provide beneficial effects in vivo in a preclinical model of MI by reducing neutrophil extravasation, macrophage accumulation, fibroblast and EC proliferation, and fibrosis.

Our study demonstrates that DS-IkL interfered with neutrophil capture and adhesion on E-selectin substrates and cytokine-stimulated cardiac-derived ECs. DS-IkL provided protectioned against platelet activation on ECs in vitro, maintaining a similar activation state as on unstimulated ECs. In corroboration with these findings, we observed cardioprotection after a left anterior descending coronary artery (LAD) ligation-induced MI in mice in vivo. Treatment with DS-IkL resulted in improved cardiac function, reduced fibrosis, and a significant decrease in neutrophils, macrophages, proliferative fibroblasts, and proliferative ECs in the infarcted region. Taken together, these findings establish a therapeutic role of DS-IkL after MI by reducing reperfusion injury mediated by the immune response.

2. Materials and methods

For more detailed methods, please see Supplementary material online.

2.1 Animal model of IR surgery

We used 10- to 16-week-old male and female C57Bl/6J mice for this study. All animal handling and laboratory procedures were performed in accordance with the approved protocols of the Institutional Animal Care and Use Committee of the University of California, Davis, which conforms to the Guide for the Care and Use of Laboratory Animals published by the US National Institutes of Health (8th Edition, 2011). Mice were randomly selected to undergo either sham-operation, or ischaemia followed by 24 h [triphenyltetrazolium chloride (TTC) staining and cardiac troponin measurments], 36 h [in vivo imaging (IVIS) and fluorescence-activated cell sorting (FACS) analysis], or 2 weeks [electrophysiology, histology, and immunohistochemistry (IHC)] of reperfusion (I/R). Mice were anaesthetised with 80 mg/kg ketamine and 5 mg/kg xylazine, intraperitoneal, prior to surgery. After toe pinch and corneal reflexes were lost, ischaemia was induced by ligating the proximal LAD coronary artery for 45 min. Anaesthesia was maintained by supplementing 1% isoflurane throughout the I/R procedure. Mice were injected with 100 µL of either 0.9% saline or 30 µM DS-IkL in saline via the tail vein immediately after reperfusion and at t = 24 h. Mice were randomly selected to receive either DS-IkL or saline. Mice were given buprenorphine twice daily for 48–72 h post-operation (0.1 mg/kg subcutaneously). For euthanasia, mice were injected with 80 mg/kg ketamine and 5 mg/kg xylazine to achieve a surgical plane of anaesthesia followed by exsanguination upon removal of the heart. Cardiac structure and function, and neutrophil and macrophage aggregation, were assessed 2 weeks after surgery. For the in vivo imaging system and flow cytometry experiments, mice were allowed to recover for 36 h after surgery. Animals were imaged on an IVIS Spectrum in vivo imaging system (Perkin Elmer) under 1% isoflurane 1, 12, 24, and 36 h after reperfusion prior to euthanasia. For TTC staining and troponin analysis, mice were euthanized 24 h following reperfusion.

2.2 Peptide library

A combinatorial peptide library biased toward a known 7-mer selectin-binding peptide sequence, Ile-Glu-Leu-Leu-Gln-Ala-Arg,^18,19 was created using the one-bead-one-peptide split synthesis method²⁰ on Tentagel S resin. The library was split a total of 7 times, yielding 31⁷ unique sequences (Figure 1). Protecting groups were cleaved (82.5% trifluoroacetic acid, 5% phenol, 5% water, 5% thioanisole, and 2.5% triisopropylsilane) and resin screened for binding to recombinant human E-selectin Fc chimera. One sequence, Ile-(D)Lys-Leu-Leu-(D)Pro-Hydroxyproline-Arg (IkL) was selected for use in the experiments described herein.

Figure 1.

One-bead-one-peptide library synthesis. (A) A combinatorial peptide library was designed using the split synthesis method of Lam et al. biased toward a known E-selectin-binding sequence. (1) Resin was split in half; (2) half was coupled to the first amino acid of the selectin-binding sequence and the other half divided evenly into 30 tubes, each containing a different amino acid. (3) After coupling was complete, resin was combined and deprotected. (4) The cycle was repeated for each amino acid in the selectin-binding sequence to create a library with 31⁷ unique peptide sequences. (B) In each step, resin is divided in two and coupled to a different amino acid, b_x or a_x. In this example, the cycle is repeated three times to create a library with 2³ unique sequences. (C) An example of the screening result is shown. Resin with peptide that bound E-selectin can be seen in brown (red arrows). This figure was created with BioRender.

2.3 DS-IkL synthesis

Dermatan sulphate (DS, 41816 Da, Celsus Laboratories) was dissolved in phosphate buffer (pH 4.54) and reacted with 45 equivalents of 4-(4,6-dimethoxy-1,3,5-triazin-2-yl)-4-methyl-morpholinium chloride (DMTMM) for 5 min. HyNic-GRGsIkLLpHypR (IkL) (InnoPep) was dissolved in phosphate buffer and added to the reaction. The reaction was left to complete for 48–60 h at room temperature with constant shaking, quenched with water, then filtered using tangential flow filtration. Purified molecules were frozen and lyophilized for future use. The number of peptides bound was quantified by reading absorbance at 280 nm on a NanoDrop Microvolume Spectrophotometer (ThermoFisher Scientific) and comparing it to a standard curve of free peptide. Unless otherwise stated, a molecule with an average of 14–16 peptides bound per DS was used for the present studies. For fluorescent molecules, CF594 Dye Hydrazide (Biotum) was conjugated to DS using two equivalents of DMTMM and the complex purified before conjugation to IkL.

2.4 Cell culture

Human cardiac microvascular endothelial cells (HCMEC, PromoCell) p4-6 were used. Stimulation media was prepared by diluting to 0.4 ng/mL tumour necrosis factor alpha (TNF-α) and 0.3 ng/mL interleukin 1 beta (IL-1β) in complete endothelial growth medium MV (PromoCell).

2.5 Neutrophil and platelet isolation

Human whole blood was collected into EDTA (neutrophil experiments) or sodium citrate (platelet experiments) tubes in accordance with approved protocols of the Institutional Review Board Administration at UC Davis, which conform to the principles outlined in the Declaration of Helsinki. All participants gave informed consent prior to participation in the study. Neutrophils were isolated using an EasyStep Direct Human Neutrophil Isolation Kit (Stem Cell Technologies) and used at 0.3–2 × 10⁶ cells/ml in Hank’s Balanced Salt Solution with calcium and magnesium (HBSS^+/+) supplemented with 0.1% human serum albumin (HSA). For HCMEC binding experiments, neutrophils were stained with 1.5 nM Calcein-AM (BioLegend) for 30 min at 37˚C before use. For microsphere experiments, neutrophils were stained with AF488 anti-human CD11a/CD18 (clone m24) and PE anti-human CD15 (clone HI98) antibodies (BioLegend) for 20 min on ice. For platelet experiments, blood was separated by centrifugation for 20 min at 200 × g; the resultant platelet-rich plasma (PRP) layer was collected and used.

2.6 Platelet activation

HCMECs were grown to confluence on 96-well CellBind plates then treated with 30 µM DS-IkL, 450 µM IkL peptide, or 30 µM DS in stimulation media for 4 h. Wells were rinsed 2× with HBSS^+/+ then treated with 100 µL of PRP at 37˚C. After 1 h, 45 µl of PRP was removed from each well and added to tubes containing 5 µL of ETP [107 mM Ethylenediaminetetraacetic acid, disodium salt (EDTA), 12 mM Theophylline, and 2.8 µM Prostaglandin E1 in water]. Samples were briefly centrifuged and flash frozen.

2.7 NAP-2 and PF-4 ELISA

Unless otherwise stated, all incubations occurred at room temperature with shaking (700 RPM) and wells were rinsed 3× with PBS + 0.05% Tween 20 (PBST) or 1% BSA in PBS between steps. Mouse monoclonal anti-hNAP-2 IgG and anti-hPF4 IgG_2B capture antibodies (R&D Systems) were coated on 96-well EIA/RIA high binding plates (Corning) at 2 µg/mL in 1× PBS and incubated at 4˚C overnight. Wells were then blocked for 1 h with 1% BSA. Samples were thawed and centrifuged at 2000 × g for 20 min and supernatant diluted 1:5000 in 1% BSA. Blocking buffer was removed (no rinse) and samples added for 2 h. Biotinylated polyclonal goat anti-hNAP-2 IgG and anti-hPF4 IgG detection antibodies (R&D systems) were added at 0.2 µg/mL in 1% BSA for 2 h. Streptavidin-HRP (diluted 1:200 in PBS) was then added for 20 min. The colorimetric change was induced with 1:1 hydrogen peroxide: tetramethylbenzidine solution for 20 min. The reaction was stopped by addition of 2N sulfuric acid and absorbance measured at 450 nm and 540 nm. Signals were subtracted to obtain a final absorbance.

2.8 Neutrophil binding to E-selectin-coated microspheres

Fluoresbrite 641 1.75 µm carboxylate microspheres were coated with 10 µg/ml E-selectin using a PolyLink coupling kit (PolySciences, Inc.). Functionalized microspheres were treated with HBSS^+/+ (vehicle), EC-SEAL (30 µM), DS-IkL₁₀ (30 µM), or DS-IkL₁₅ (30 µM) for 1 h. Microspheres were pelleted at 1000 × g and resuspended in stained isolated human neutrophils, then incubated for 30 min at room temperature with rotation. Samples were fixed in 4% paraformaldehyde (PFA) and read on an Attune NxT Flow Cytometer (Invitrogen). Data were analysed using FlowJo software.

2.9 Neutrophil binding to HCMECs

HCMECs were grown to confluence in a tissue culture treated 96-well plate, stimulated for 4 h, then treated with DS-IkL (30 µM), DS (30 µM), IkL alone (450 µM), or HBSS^+/+ containing 0.2% HSA (vehicle) for 1 h. Cells were washed twice with HBSS^+/+ then treated with 50,000 Calcein-AM labelled neutrophils per well with rotation (150 RPM, 37˚C). After 30 min, cells were washed to remove non-adherent neutrophils and fluorescence read on a SpectraMax M5 plate reader (Molecular Devices).

2.10 CF594-DS-IkL binding to selectin surfaces

Piranha etched glass coverslips were coated with 0.2 mg/mL protein A/G and 25 mM (bis(sulfosuccinimidyl)suberate) (BS3) overnight before being adhered to sticky-Slides VI 0.4-untreated (Ibidi). Channels were coated with recombinant human E-selectin-FC or P-selectin-FC at 20 µg/mL (R&D Systems) for 1.5 h with rocking, blocked with 1% BSA for 1 h, then treated with serially diluted CF594-DS-IkL in HBSS^+/+ for 1 h with rocking. Channels were rinsed and stored in HBSS^+/+ at 4˚C overnight before imaging to remove non-specifically bound molecule, then imaged on a Nikon TE2000 inverted microscope (Nikon, Minato, Tokyo, Japan). For each channel, five images were acquired in random locations and average fluorescence (mean grey value) within a two-dimensional region of interest (ROI) quantified using ImageJ.

2.11 Echocardiography

Cardiac structure and function were monitored using 2D echocardiography (VisualSonic Vevo 2100 with a MS 550 D probe). All mice underwent echocardiography 2 weeks after either sham-operation or I/R surgery. Systolic function was recorded under conscious conditions, while the diastolic function was recorded with isoflurane (∼0.5–1%). Systolic function was acquired from 2-D M-mode images and B-mode videos at the short axis, and diastolic function was obtained from pulse-wave Doppler images.

2.12 Fibrosis measurements

After 2 weeks, hearts from all four groups were fixed in 4% PFA solution in PBS, followed by paraffin embedding. Five micrometre sections were cut at the transverse plane of the heart, at the point of ligation. Sections were then deparaffinized in a series of xylene/alcohol solutions (xylene, xylene, 1:1 xylene: ethanol, 100% ethanol, 100% ethanol, 95% ethanol, 70% ethanol, 50% ethanol; 3 min in each solution). These sections were stained for both Picrosirius Red and Masson’s Trichrome, which provide information on the collagen level. Images were analysed in a blinded fashion.

2.13 Cell isolation

Cells were isolated using a Langendorff perfusion system as previously described.²¹ After isolation, cells were fixed in 0.4% PFA, and then subjected to flow cytometry.

2.14 Flow cytometric analysis of non-myocyte cells

The single-cell suspension was obtained from 10- to 12-week-old C57BL/6 mice as described above. Cells were re-suspended in Ca²⁺ and Mg²⁺ free PBS, treated with phytoerythrin-conjugated anti-Thy1.2 (BD Bioscience, San Diego, CA, USA), anti-CD11b, anti-Ly-6C, anti-Ly-6G (BD Bioscience), anti-CD31 (BD Bioscience), anti-MyHC (Developmental Studies Hybridoma Bank, created by NICHD of the NIH and maintained at The University of Iowa, Department of Biology, Iowa City, IA, USA), and proliferation-specific Ki67 antibodies (BD Bioscience). Cells were also stained with 40 µg/mL 7‐amino‐actinomycin D (7AAD, BD Bioscience, San Jose, CA, USA) to measure the DNA content. Data was collected using a standard FACScan cytometer (BD Biosciences, San Jose, CA, USA) upgraded to a dual laser system with the addition of a blue laser (15 mW at 488 nm) and a red laser (25 mW at 637 nm Cytek Development, Inc., Fremont, CA, USA) or Becton Dickinson LSR-II Flow Cytometer. Data were acquired using CellQuest and DIVA 6.2 software (BD Bioscience). Cells stained with isotype-matched IgG antibodies were used as controls to determine the positive cell population. Data were analysed using FlowJo software (ver9.4 Treestar Inc., San Carlos).

2.15 IVIS imaging

Mice were injected with 30 µM CF594-DS-IkL or free fluorophore in 0.9% saline after surgery and after 24 h (100 µL via the tail vein). Prior to imaging, mice were anaesthetized by isoflurane inhalation. Mice (n = 6 per group) were imaged in an IVIS Spectrum Imaging System (PerkinElmer). Fluorescent images were taken 1, 12, and 24 h after the initial injection, and 12 h after the second injection (36 h after the initial injection). Images were analysed using Living Image system software (PerkinElmer, version 4.3.1). Two-dimensional ROI circles were drawn around the heart to quantify fluorescence and a control ROI was drawn on the flank to quantify background. Fluorescence values were normalized to the background by dividing the average radiant efficiency of the heart by that of background.

2.16 Immunohistochemistry

Hearts were fixed in 4% PFA overnight then washed with PBS the following day. Hearts were paraffin embedded and sliced into 5 µm sections. The sections were deparaffinized with xylene and rehydrated in a descending grade of ethanol. Antigen retrieval was performed in Citrate Buffer (pH 6.0, 15 min at 95°C). Sections were permeabilized with 0.25% Triton X-100 in PBS for 30 min at room temperature, and then blocked with 5% donkey serum diluted by 0.25% Triton X-100. For macrophage staining, sections were incubated with rat anti-F4/80 (BioLegend, monoclonal, 1:200) and mouse anti-α-Actinin (Sigma, monoclonal, 1:200) primary antibodies. For neutrophil staining, primary antibodies used were rat anti-Ly6G (BioLegend, monoclonal, 1:200) and mouse anti-α-Actinin (Sigma, monoclonal, 1:200). Primary antibodies were diluted in blocking solution and sections incubated overnight at 4°C. Secondary antibodies used were donkey anti-mouse Alexa Fluor 488 (Invitrogen, 1:500) or donkey anti-mouse Alexa Fluor 555 (Abcam, 1:500). Secondary antibodies were diluted in blocking solution and incubated for 1 h at room temperature. Slides were mounted and then imaged using a Zeiss confocal LSM 700 microscope. Images were analysed by ImageJ.

2.17 TTC staining

Hearts were embedded in 2% agarose gel in a heart matrix, then subsequently sliced into 1–1.5 mm sections. Sections were immersed in 1% TTC solution for 30 min at 37°C. They were then fixed in 4% PFA for 20 min before images were taken under a dissection microscope connected to a Dino-Lite USB microscope.

2.18 Troponin I measurements

Troponin I (CNTI) was measured using a commercial kit (Life Diagnostics) according to the manufacturer instructions. Sample concentrations were calculated from a standard curve generated in GraphPad Prism (San Diego, CA, USA) using four-parameter logistic regression. Sample concentrations were multiplied by their respective dilution factors to obtain final concentration values.

2.19 Statistics

All in vitro experiments were repeated at least three times for an n ≥ 3. Each experiment was performed with 3–4 technical replicates. Data were analysed using one-way or two-way ANOVA with post hoc Tukey multi-comparisons analysis. Data are represented as means ± SEM.

3. Results

3.1 Combinatorial peptide library screening generated the selectin-binding sequence, IkLLpHypR

Selectins mediate the initial capture of leucocytes to the vascular surface¹⁵ and are therefore a promising target for modulating leucocyte recruitment. We first set out to create a combinatorial peptide library containing unnatural amino acids, biased toward the selectin-binding sequence of EC-SEAL, a molecule we previously described.^18,19 We used the one-bead-one-compound combinatorial library method, which allows for the unrestricted incorporation of natural and unnatural amino acids (Figure 1),²⁰ to discover the sequence Ile-(D)Lys-Leu-Leu-(D)Pro-Hydroxyproline-Arg.

3.2 DS-IkL exhibited binding to E- and P-selectin-coated surfaces and reduced neutrophil binding to E-selectin in solution

To confirm that DS-IkL binds selectins, P- and E-selectin-coated surfaces were treated with increasing concentrations of DS-IkL or DS alone. Dermatan sulphate (DS) is a negatively charged proteoglycan that is known to interact with P-selectin.²² We directly determined if the addition of IkL would improve binding to P- or E-selectin compared to DS alone. Figure 2 shows that DS-IkL exhibited significantly enhanced binding to both selectins with increasing concentrations of molecule, which is not observed in the DS group. The ability of DS-IkL to interfere with E-selectin-binding to neutrophils in suspension was further confirmed in Supplementary material online, Figure S1. Treatment with DS-IkL inhibited neutrophil binding to E-selectin, but not P-selectin, at 5–30 µM concentrations by up to 35%. These reductions were similar to those seen on endothelial cell surfaces (Figure 3D). Higher concentrations were needed to reduce P-selectin binding. Similarly, in the presence of DS-IkL, stimulated neutrophils bound less E-selectin, but not intercellular adhesion molecule-1 (ICAM-1) (Supplementary material online, Figure S1), suggesting DS-IkL preferentially binds E-selectin.

Figure 2.

Improved binding to immobilized selectin surfaces with DS-IkL. P- and E-selectin-coated surfaces were treated with CF594-DS-IkL (30 µM, red) or CF594-DS (30 µM, black). CF594-DS-IkL binding increased with increasing concentration, as assessed by mean fluorescence intensity normalized to vehicle (HBSS). Data are represented as means ± SEM of three biological replicates. * P < 0.05, ** P < 0.01, *** P < 0.001 by two-way ANOVA with post hoc Tukey test.

Figure 3.

Treatment with DS-IkL reduced neutrophil binding and platelet activation. Platelet activation on stimulated HCMECs as assessed by (A) NAP-2 and (B) PF-4 was reduced after treatment with DS-IkL (30 µM) and/or IkL peptide (450 µM), but not DS (30 µM). n = 3 biological replicates. (C) E-selectin-coated microspheres were treated with HBSS, DS-IkL, IkL peptide, or DS prior to incubation with isolated human neutrophils (PMN). Data represent PMN binding to microspheres as assessed by flow cytometry. n = 3–6 biological replicates. (D) DS-IkL reduced PMN binding to stimulated HCMECs toward the level of unstimulated controls (P = 0.09). n = 3–4 biological replicates. All data are represented as means ± SEM. * P < 0.05, ** P < 0.01, *** P < 0.001 by one-way ANOVA with post hoc Tukey test.

3.3 Treatment with DS-IkL significantly reduced platelet activation on stimulated endothelial cells

Enhanced platelet activation and binding have been implicated in a multitude of chronic inflammatory diseases, including acute coronary and pulmonary syndromes.²³ Activated platelets release cytokines such as platelet factor 4 (PF-4) and neutrophil-activating peptide 2 (NAP-2) that can in turn activate neutrophils.²⁴ Given their contributions to neutrophil recruitment to sites of inflammation, we aimed to test if treatment of inflamed ECs with DS-IkL could reduce PF-4 and NAP-2 activation of platelets. Platelets were allowed to interact with stimulated HCMECs under standard culture conditions for 1 h before the supernatant was collected. Levels of PF-4 and NAP-2 that accumulated in the supernatant during this time were significantly diminished in DS-IkL treated cells (P < 0.05), as shown in Figure 3A,B. Notably, DS-IkL reduced NAP-2 activation on stimulated HCMECs to levels similar to unstimulated controls, implying our molecule has the potential to act as a barrier to platelet-induced neutrophil activation.

3.4 DS-IkL reduced neutrophil arrest on E-selectin-coated microspheres and stimulated endothelial cells

The neutrophil adhesion cascade begins with initial capture and slow rolling along selectins that are upregulated on inflamed endothelium and activated platelets; therefore, interfering with this initial capture could prove therapeutically beneficial. To test our molecule’s ability to block neutrophil-selectin interactions, E-selectin-coated microspheres were treated with DS (30 µM), IkL peptide (450 µM), DS-IkL (30 µM), DS-IkL₁₀ (30 µM), or vehicle prior to incubation with isolated human neutrophils. Only IkL peptide and DS-IkL were able to reduce neutrophil adhesion to microspheres to approximately 75% of control (P < 0.05, Figure 3C), with DS and DS-IkL₁₀ exhibiting similar neutrophil-microsphere colocalization as control. Given the importance of tightly regulated neutrophil capture and activation in proper wound healing,²⁵ this slight reduction could be the tipping point back toward a restorative state.

We next sought to investigate if the effect of DS-IkL on neutrophil-selectin interactions was retained on ECs, which are known to upregulate their selectin expression upon stimulation with inflammatory cytokines such as TNF-α and IL-1β.²⁶ HCMECs were treated with DS-IkL under stimulatory conditions prior to incubation with neutrophils. Neutrophils exhibited reduced adhesion to treated ECs as compared with vehicle controls (P < 0.01, Figure 3D), to a similar extent as on selectin-coated microspheres. These results support our hypothesis that DS-IkL reduces selectin-mediated neutrophil capture at sites of inflammation.

3.5 DS-IkL targeted to the heart after I/R

Since DS-IkL suppressed neutrophil adhesion in vitro, we investigated whether this would occur in vivo following a myocardial infarction, induced by 45 min of ischaemia at the LAD with subsequent reperfusion. To verify proper occlusion of the LAD, we monitored ECG recordings during occlusion (Supplementary material online, Figure S1). Only I/R mice with ST-segment elevations were included in the I/R group. Immediately after sham or LAD operation, we injected either 30 µM DS-IkL conjugated to a fluorophore (CF594-DS-IkL) or saline with an equal concentration of the fluorophore, and monitored in vivo fluorescent intensity at 1, 12, 24, and 36 h after surgery. Representative images of all four groups are shown in Figure 4A. As evidenced by these images, fluorescence signals were primarily localized to the cardiac region only in the I/R group after the CF594-DS-IkL injection. We observed some targeting to throat regions that sustained damage during tracheotomy, as well as background fluorescence in the abdomen, likely from autofluorescent components of mouse chow.²⁷ Quantitatively, signal intensity was significantly higher an hour after surgery in mice treated with DS-IkL after I/R, relative to mice treated with the fluorophore dissolved in saline (P < 0.001, Figure 4B). There was no difference in fluorescence intensity in the cardiac region in saline sham mice vs. saline I/R mice, suggesting that the molecule just circulated in the blood, but did not localize to the heart. In addition, fluorescent signal persisted for at least 24 h in the cardiac region after I/R surgery in mice treated with DS-IkL. After a subsequent dosage at the 24 h mark, no noticeable difference was observed in this group relative to the other groups at the 36 h time point. Taken together, our in vivo imaging suggested that DS-IkL targeted to the heart, and the binding of DS-IkL lasted for at least 24 h.

Figure 4.

DS-IkL targeted to the heart after I/R. Animals were injected with either CF594 dissolved in saline or CF594-DS-IkL after surgery and 24 h later. (A) Representative images taken 1 h after each injection, and every 12 h after reperfusion until the 36 h time point. (B) Data summary of CF594 (black bars) and CF594-DS-IkL (red bars) targeting to heart at 1, 12, 24, and 36 h after surgery. Black arrows depict injections. Data are represented as means ± SEM of n = 6 mice per group. * P < 0.05, ** P < 0.01; *** P < 0.001 by two-way ANOVA with post hoc Tukey test.

3.6 Mice treated with DS-IkL exhibited reduced infarct size and improved cardiac function after I/R

To examine the in vivo effects of DS-IkL on cardiac structure and function, we performed either sham or 45 min I/R. To assess the short-term effects of DS-IkL on cardiac injury, we administered one injection of either saline or DS-IkL immediately after surgery and examined the degree of myocardial infarction and extent of the cardiac injury after 24 h of reperfusion. Representative sections stained with TTC are displayed in Figure 5A. As evident from the sections, I/R induced a more pronounced increase in damaged, non-viable cardiac regions, relative to DS-IkL treated mice. Sham-operated sections are displayed in Supplementary material online, Figure S4. Quantification, percentage of the infarcted region relative to the left ventricular area, demonstrated a significant reduction in infarct size in DS-IkL hearts (Figure 5B). The cardiac injury was assessed by measuring cardiac Troponin I. Although there was not a significant difference, there was a trend towards a reduction in injury at 24 h post-injury (Figure 5C).

Figure 5.

Mice treated with DS-IkL exhibited reduced infarct size and improved cardiac function after I/R. (A) After 24 h of reperfusion, we assessed size of myocardial infarction using TTC staining. Representative sections are shown. (B) Quantification revealed an increase in infarct size with saline-treated mice. (C) Cardiac injury was also assessed using cardiac troponin I. After 2 weeks of reperfusion, mice were subjected to conscious echocardiography to examine the cardiac structure and function. (D) Representative M-mode images at the parasternal short-axis for all four groups are depicted. (E) Heart rate was not significantly different in these conscious mice. (F) I/R induced an increase in left ventricular mass, and (G) reduced fractional shortening and (H) global radial strain. (I) Summary data of region-specific radial strain and (J) circumferential strain are displayed. (K) Diastolic function was assessed by blood flow velocity through the mitral valve, as shown in representative images in all four groups. (L) The E/A ratio was significantly decreased in both groups but was more obvious in the saline I/R group. (M) Although the isovolumetric relaxation time (IVRT) was not significantly different in all four groups, (N) there was a difference in the mitral valve (MV) deceleration time in the saline I/R group vs. the saline sham group. In each graph, data are represented as means ± SEM of n = 5–7 mice per group. * P < 0.05, ** P < 0.01 by one-way ANOVA with post hoc Tukey test.

To determine the long-term effects of DS-IkL, we performed the same procedure, but mice were subjected to 2 weeks of reperfusion. To ensure robust activity of DS-IkL, we administered an additional dose of DS-IkL 24 h after surgery on top of the dose given immediately after surgery. Cardiac structure and function were assessed 2 weeks post-operation. Representative M-mode tracings at the parasternal short-axis are shown for all four groups and display the beat-to-beat changes in wall thickness and diameter of the left ventricle (Figure 5D). Recordings were performed in conscious mice; heart rate in all groups was therefore similar (Figure 5E). Structurally, I/R resulted in an increase in the left ventricular mass in groups treated with either saline or DS-IkL, relative to their respective sham-operated groups (P < 0.05, Figure 5F). Similar to our echocardiography findings, I/R induced an increase in heart weight to tibial length ratio in mice treated with either saline or DS-IkL, relative to their sham-operated controls (Supplementary material online, Figure S2). Even though I/R resulted in a reduction in fractional shortening (Figure 5G) and ejection fraction (Supplementary material online, Table S1) in both groups, there was a significant improvement in fractional shortening in the DS-IkL treated mice compared to saline alone (P < 0.05, Figure 5G). Furthermore, strain analysis revealed a significant reduction in global radial strain (Figure 5H) in both I/R groups relative to their respective sham groups. However, the decrease was significantly less in DS-IkL mice treated groups. Although the local radial strain was not significant (Figure 5I), the small improvement in all the different regions of the heart contributed to an overall increase in global radial strain in DS-IkL mice. Additionally, the absolute, global circumferential strain was significantly altered in saline-treated I/R mice relative to their sham control (Figure 5J). However, this was not observed in DS-IkL treated mice.

Diastolic function was assessed using the blood flow velocity through the mitral valve during the cardiac cycle. Representative tracings using pulse-wave Doppler show two distinct waveforms, which correspond to left ventricular filling during early diastole (E wave) and left ventricular filling during late diastole (A wave, Figure 5K). The ratio of the two provides an indication of diastolic function.²⁸ Mice treated with DS-IkL did not exhibit a change in diastolic function, as compared to their sham control; whereas, mice treated with saline showed a significant reduction in the E/A ratio, indicating an impairment in diastolic function (Figure 5L). Indeed, the E/A ratio of mice treated with DS-IkL was significantly higher than mice treated with saline, after I/R (P < 0.05). Although the isovolumetric relaxation time (IVRT, Figure 5M) did not change significantly, the MV deceleration time was significantly elevated in the I/R group treated with saline, relative to the saline sham group (P < 0.05, Figure 5N). This was not seen when the mice were treated with DS-IkL. Together, these data suggest that both systolic and diastolic function was significantly improved with the treatment with DS-IkL.

3.7 DS-IkL limited fibrosis after I/R

MI induces loss of cardiomyocytes with a concomitant increase in fibrosis.²⁹ Histological analyses of cardiac sections were performed 2 weeks after I/R. Representative whole heart images of all four groups, and short-axis sections stained with Masson’s Trichrome and Picrosirius Red are shown (Figure 6A). Percentages of fibrotic area relative to the total or left ventricular area were quantified in a blinded fashion and were low and not significantly different in the sham-operated groups. In contrast, there was a significant elevation in collagen deposition in both I/R groups, relative to their respective sham-operated controls (P < 0.01, Figure 6B). However, I/R mice treated with DS-IkL showed significantly less fibrosis than mice treated with saline (Figure 6B-D, P < 0.05), consistent with TTC data observed 24 h after I/R (Figure 5A,B). Together, the data show that DS-IkL limited fibrosis, which in part contributed to the improved cardiac function as assessed by echocardiography.

Figure 6.

DS-IkL limited fibrosis after I/R. (A) Representative images of whole hearts (top), Masson’s Trichome (middle), and Picrosirius Red (bottom) stained sections. Collagen deposition by Masson’s Trichrome (B and C) and Picrosirius Red (D and E) was significantly reduced in DS-IkL treated mice after I/R. Data are represented as means ± SEM of n = 5–7 mice per group; datapoints display 2–4 sections analysed per mouse. * P < 0.05, ** P < 0.01, *** P < 0.001 by one-way ANOVA with post hoc Tukey test.

3.8 DS-IkL prevented neutrophil and macrophage aggregation, fibroblast proliferation, and endothelial cell proliferation after I/R

I/R causes cardiac endothelial cell dysfunction, leading to an amplified inflammatory state and an increase in vascular permeability.³⁰ If not tightly regulated, immune cells recruited to the damaged regions accentuate damage and contribute to tissue fibrosis. To determine the effect of DS-IkL on neutrophil and fibroblast accumulation, we isolated non-myocyte cell populations from the hearts from all four groups 36 h after sham or I/R surgeries. Flow cytometric analysis showed a significant increase in CD11b⁺/Ly6-C/G⁺ neutrophils in the saline-treated I/R group compared to sham controls (P < 0.05), but no significant increase in the DS-IkL treated I/R group (Figure 7A,B), suggesting that treatment with DS-IkL limited neutrophil accumulation after I/R. Furthermore, there was a significant increase in both total (CD31⁺) and proliferative ECs (Ki67⁺/CD31⁺) in the saline I/R group, but not in the DS-IkL treated I/R groups (Figure 7C–E), suggesting a significant decrease in adverse vascular remodelling by DS-IkL. Moreover, the percentage of proliferative fibroblasts (Thy1.2⁺) in the saline I/R group was significantly higher than in the DS-IkL treated I/R group (Figure 7F–H), consistent with the significant increase in collagen deposition observed 2 weeks after reperfusion. It must be noted that CD45 was not used to gate out hematopoietic cells that may also be positive for CD31 or Thy1.2. Thus, while the changes in endothelial and fibroblast cell numbers are suggestive, future work is needed to definitively show these changes.

Figure 7.

DS-IkL prevented neutrophil and macrophage aggregation, fibroblast proliferation, and endothelial cell proliferation after I/R. We investigated the short- and long-term effects of DS-IkL on the immune response. (A) Flow cytometric analysis of isolated cardiac cells. Nucleated cells were selected from debris based on the incorporation of 7-AAD and the separation of myocytes from non-myocyte cells (NMC) using a cardiac myosin heavy chain (MF20)-specific antibody. Neutrophils and monocytes were separated based on the presence of CD11b/Lys6-C/G from all four groups as depicted. Neutrophils were further distinguished by gating on the SSC-high and FSC-low cells from CD11b⁺/Ly6C/G⁺ population. (B) Summary data show that there was a significant reduction in neutrophils in the DS-IkL I/R group, compared to the saline I/R group. Three technical replicates per mouse are shown. (C) Flow cytometric analysis of CD31⁺/Thy1.2⁻ endothelial cells in all four groups. (D) Summary data showing a significant increase in endothelial cells in saline I/R group relative to saline sham. DS-IkL prevented the increase. Data show two technical replicates per mouse. (E) Assessment of proliferation using the Ki67 proliferative marker showed a similar trend. Data represent averages of two technical replicates per mouse. (F) Fibroblasts (Thy1.2⁺/CD11b⁻/Ly6C/G⁻) were detected using flow cytometric analysis. (G) Summary data of Thy1.2 positive cells (two technical replicates per mouse) and (H) proliferative Thy1.2 positive cells using Ki67 (averages of two technical replicates per mouse) are shown. X and Y axes represent arbitrary units. (I) We examined the accumulation of neutrophils (Ly6G) and macrophages (F4/80) after 2 weeks of I/R using IHC. Representative images are shown for all four groups. DS-IkL prevented accumulation of both (J) macrophages and (K) neutrophils. n = 3–5 mice per group for flow cytometric analysis. n = 3 mice per group, 2–3 sections per mouse for IHC. Data are represented as means ± SEM. * P < 0.05, ** P < 0.01, *** P < 0.001 by one-way ANOVA with post hoc Tukey test.

Although the immune response is heightened shortly after MI, the effects can manifest long-term as the body tries to re-establish homeostasis. To determine the long-term effects of DS-IkL on neutrophil and macrophage accumulation, we performed immunohistochemistry on all four groups after 2 weeks of I/R. Representative images of IHC are shown in Figure 7I. Tissue sections were stained with α-actinin to stain for cardiomyocytes, F4/80 to stain for macrophages, and Ly6G to stain for neutrophils. DS-IkL treatment significantly reduced the number of macrophages per area (Figure 7I,J) and neutrophils per area (Figure 7I,K) at 2 weeks (Figure 7J,K). Together, our data suggest that DS-IkL limited neutrophil and macrophage aggregation, as well as fibroblast and EC proliferation. Mechanistically, DS-IkL treatment prevented multiple aspects of the immune response by acting as a molecular bandage on the damaged endothelial surface, which ultimately resulted in less fibrosis and improved cardiac function.

4. Discussion

Inflammatory responses significantly contribute to further injury after the initial insult of an MI. Ischaemic coronary vessels undergo drastic alterations to their microstructure and homeostasis upon reperfusion, conducive to immune cell infiltration. Ion imbalance disrupts cell–cell junction integrity along the endothelium,³¹ creating migration points for neutrophils and macrophages that are recruited to these cytokine-rich environments.³⁰ Furthermore, ECs exhibit signs of endothelial cell dysfunction, namely enhanced cell contractility, upregulated cell adhesion molecules such as E- and P-selectin, and the loss of the glycocalyx.^10,30 The dysfunctional EC phenotype facilitates neutrophil recruitment and extravasation. Neutrophils that have been recruited to these sites activate and secrete a milieu of pro-inflammatory cytokines that in turn recruit additional leucocytes,³² shifting the inflammatory process from restorative to destructive. Therapeutics that limit the initial binding and activation of circulating immune cells could tip the balance back to a restorative state.

4.1 Design of the novel DS-IkL

We previously described a molecule termed ‘EC-SEAL’ composed of oxidized dermatan sulphate (DS), to which known selectin-binding peptides were conjugated via a heterobifunctional crosslinker.¹⁸ Although EC-SEAL mitigated the immune response in vitro, it is not an acceptable drug candidate for the clinic. The chemistry used to functionalize the DS uses glycol splitting by periodate oxidation, a mechanism which cleaves carbohydrate rings at vicinal diols to create a reactive intermediate.³³ This chemistry renders the molecule unstable; thus, it cannot be stored long-term. Furthermore, the chain conformation changes dramatically, resulting in a coiled polymer chain rather than an extended structure seen with natural glycosaminoglycans.

In this study, we designed an improved molecule termed DS-IkL. We demonstrated that DS could be modified with peptides without cleaving the ring structure and still maintain the ability to bind to the selectin receptors to interfere with platelet and neutrophil binding. Importantly, this new chemistry, which better maintains the DS structure, eliminates the reactive intermediates that reduced the stability of EC-SEAL. Additionally, the selectin-binding sequence in EC-SEAL consisted solely of L-amino acids, which are readily degraded by proteolytic enzymes present at sites of inflammation. Though not directly addressed here, we believe that the incorporation of D-amino acids within the selectin-binding sequence delays DS-IkL’s degradation and clearance, supporting its presence on ECs for at least 24 h.

To design DS-IkL, we first created a combinatorial peptide library containing D-amino acids, biased toward the E-selectin-binding sequence of EC-SEAL.^18,19 We chose to use the one-bead-one-compound combinatorial library method, which allows for the unrestricted incorporation of natural and unnatural amino acids,²⁰ to discover the sequence I-k-L-L-p-Hyp-R. Our preliminary efforts focused on the ability of IkL peptide and our DS-IkL glycoconjugate to bind selectins (Figure 2) and interfere with neutrophil binding to E-selectin-coated microspheres (Figure 3A). Selectins mediate the initial capture of leucocytes to the vascular surface,¹⁵ and are therefore a promising target for modulating leucocyte recruitment. We hypothesized that selectins blanketed with our glycoconjugate would be masked from circulating leucocytes, therefore limiting their capture. In line with this hypothesis, we observed significant reductions in neutrophil binding to both selectin-coated microspheres and stimulated ECs upon treatment with DS-IkL, suggesting DS-IkL interacted with upregulated selectins on the cell surface.

4.2 DS-IkL reduced platelet activation on endothelial cells

Given the destructive role platelets play in myocardial I/R injury,^16,34 we were interested in DS-IkL’s effect on the activation state of platelets incubated on stimulated ECs. Interestingly, we saw significant reductions in NAP-2 levels and a trend toward reduction in PF-4 in cultures that had been pre-treated with DS-IkL as compared to HBSS treated controls, suggesting ECs treated with DS-IkL presented fewer available platelet-activating ligands such as endothelial P-selectin. Though we did not observe significant reductions in PF-4, a chemokine which induces firm adhesion of neutrophils in the presence of TNF-α,^24,35 we believe our modest results may be an artefact of the experiment. PF-4 has an affinity for GAGs³⁵ and therefore may have been sequestered by DS-IkL or heparin on the cell surface or within the media.

In addition to direct damage caused by activated platelets, these bioactive blood components have been shown to play a pivotal role in neutrophil capture along inflamed endothelium, offering an anchor point via platelet P-selectin and facilitating endothelial transmigration.^23,34 Although E- and P-selectin are very similar molecules, both in structure and function, we observed slight differences in binding to E- and P-selectin-coated surfaces (Figure 2), with DS-IkL exhibiting more consistent binding to E-selectin at higher concentrations. Our results suggest that IkL peptide plays a more influential role in E- and P-selectin binding when DS-IkL is administered at higher concentrations, whereas at lower concentrations, selectin binding may be influenced more by DS. By interacting with both P- and E-selectin, DS-IkL may simultaneously protect from neutrophil interactions, platelet aggregation, and the formation of platelet-neutrophil complexes. Therefore, DS-IkL presents a two-fold cardioprotective potential by limiting damage from both activated platelets and neutrophils.

4.3 DS-IkL improved cardiac function by disrupting components of the immune response

Since one of the hallmarks of reperfusion injury is overactivation of the immune system due to leucocyte accumulation and platelet activation, we assessed the potential therapeutic benefits of DS-IkL in a clinically relevant murine model of I/R. Our data show that DS-IkL localized to the damaged cardiac region (Figure 4) and limited neutrophil and fibroblast accumulation after I/R, which resulted in less tissue fibrosis (Figure 6), reduced myocardial infarction, and improved cardiac function (Figure 5). Interestingly, DS-IkL treated I/R mice did not exhibit the increase in total and proliferative ECs that was observed in the saline I/R group. In some cases, a heightened presence of proliferative ECs may indicate tissue regeneration;³⁶ however, the saline-treated I/R group exhibited poor cardiac function and enhanced fibrosis. Our results therefore suggest that rather than regeneration, early remodelling and myocardial hypertrophy was beginning to occur, necessitating an increase in cardiac angiogenesis.³⁷

4.4 Proposed mechanism of DS-IkL

Other attempts have been made to limit I/R induced injury after MI,³⁸ including ischaemic pre- and post-conditioning, pharmacologic interventions targeting metabolic pathways that contribute to RI,^39,40 and the use of monoclonal antibodies to limit platelet and neutrophil adhesion.³⁸ In particular, antibody therapy to P-selectin^41,42 and ICAM-1 have been investigated as cardioprotective therapies against neutrophil and platelet mediated RI.⁴³ However, as evidenced by the SELECT-ACS trial evaluating the effect of Inclacumab,⁴⁴ a recombinant monoclonal antibody against P-selectin, a major drawback of antibody therapy is the need for high doses for therapeutic effect,⁴⁵ lending to high-production costs.⁴⁶ It is worth noting that monoclonal P-selectin antibody crizanluzumab⁴⁷ (awarded Breakthrough Therapy designation) and the pan-selectin inhibitor Rivipansel⁴⁸ (failed at Phase 3) have shown promise in vaso-occlusive crises in sickle cell disease. However, neither of these have been tested in RI.

Our strategy is unique from antibody therapeutics. DS-IkL is generally smaller than antibody therapeutics (∼62 kDa) and has been designed to interrupt multiple parts of the immune pathway simultaneously by masking adhesion molecules on the endothelial surface. We designed a molecular bandage that targets DS—a native component of the glycocalyx—to inflamed endothelium. By blanketing DS over the dysfunctional endothelial surface, we aimed to directly reduce selectin availability while indirectly interfering with immune cell interactions with other adhesion molecules on the endothelial surface. We demonstrated that DS-IkL bound E- and P-selectin; limited neutrophil and platelet adhesion to inflamed ECs; reduced neutrophil, macrophage, and fibroblast accumulation after MI (Figure 7) and improved cardiac function after MI. We conducted our current study at a treatment concentration of 30 µM. Figure 2 suggests DS-IkL may have a greater therapeutic benefit at higher concentrations; therefore, future studies will be required to determine the concentration of DS-IkL that produces the optimum therapeutic effect, as well as to further elucidate the mechanism by which DS-IkL works.

5. Conclusions

We created DS-IkL using the one-bead-one-peptide synthesis approach. In vitro, DS-IkL bound to both P- and E-selectin-coated surfaces and prevented neutrophil binding and platelet activation. In vivo, DS-IkL localized to the cardiac region in mice after MI and limited neutrophil and fibroblast accumulation. Mice treated with DS-IkL exhibited improved cardiac function and less fibrosis. Taken together, our data suggest that our recently developed, novel DS-IkL molecule reduced the inflammatory response induced by MI, which resulted in cardioprotection.

Data availability statement

The data underlying this article are available in the Dryad Digital Repository, https://doi.org/10.25338/B8M32V.

Supplementary material

Supplementary material is available at Cardiovascular Research online.

Authors’ contributions

T.D. and P.N.T. designed the research study, conducted experiments, acquired data, analysed data, and wrote the manuscript. H.S., L.R., and P.S. conducted experiments, acquired data, analysed data. C.N. and V.T. conducted experiments. J.L.O. analysed data. X.L. and K.L. helped conceptualize and design the peptide library. N.C. and A.P. designed the research study, provided monetary support, and wrote the manuscript.

Funding

This work was supported by a Predoctoral Fellowship from Tobacco-Related Disease Research Program (TRDRP) T29DT0237 (TD), Postdoctoral Fellowships from NIH T32 Training Grant in Basic & Translational Cardiovascular Science NIH T32 HL086350 and NIH F32 HL149288 (PNT), NIH R01 HL085727, HL085844, HL137228, and S10 RR033106, Research Award from the Rosenfeld Foundation, VA Merit Review Grant I01 BX000576 and I01 CX001490 (NC). NC is the holder of the Roger Tatarian Endowed Professorship in Cardiovascular Medicine and a part-time staff physician at VA Northern California Health Care System, Mather, CA. The authors would like to thank the Combinatorial Chemistry and Chemical Biology Shared Resource at University of California Davis for assistance of synthesis and sequence decoding of OBOC peptide library. Utilization of this Shared Resource was supported by the UC Davis Comprehensive Cancer Center Support Grant awarded by the National Cancer Institute (NCI P30CA093373).

Conflict of interest: none declared.

 

Time for primary review: 40 days

 

Translational perspective

Cardiovascular disease remains the leading cause of mortality worldwide. Acute inflammation from myocardial infarction (MI) results in damaged endothelial cells, leading to the loss of glycocalyx. Here, we designed a novel selectin-targeting glycocalyx mimetic (termed DS-IkL) capable of binding inflamed endothelial cells. DS-IkL limits neutrophil binding and platelet activation on inflamed endothelial cells. Treatment with DS-IkL in a preclinical model of MI reduces infarct size by preventing neutrophil extravasation, macrophage accumulation, fibroblast and endothelial cell proliferation, and fibrosis. Our findings suggest that DS-IkL has great therapeutic potential after MI by limiting reperfusion injury induced by the immune response.