Citrullinated Histone 3 Causes Endothelial Barrier Dysfunction

Animals

Mice used in these studies were male C57BL/6J purchased from Jackson Laboratory. Genotyping was performed by Transnetyx using real-time PCR. Mice were maintained under a 12/12-hour light/dark cycle with food and water ad libitum. All animal studies were approved by the University of South Florida Institutional Animal Care and Use Committee and was performed in accordance with the NIH Guide for the Care and Use of Laboratory Animals.

Intravital microscopic analysis of protein transvascular flux

To examine plasma protein flux across mesenteric microvessels [21], mice (10–20g) were anesthetized with an intramuscular injection of urethane (1.75 g/kg) and shaved at the abdomen. The jugular vein was cannulated for IV infusion of solutions. A midline laparotomy was performed, and the mesentery was exteriorized over an optical stage for microscopic observation. The microcirculation of the mesentery was imaged using a Nikon Eclipse E600FN microscope with Evolve 512 digital camera (Photometrics, AZ, USA). Mice were given an IV bolus of fluorescein isothiocyanate conjugated bovine albumin (FITC-albumin) at 100 mg/kg followed by continuous infusion of 0.15 mg/kg/min to maintain a constant plasma concentration. Postcapillary venules were selected for analysis of FITC-albumin flux and stimulated with 10 μg/mL recombinant H3Cit or vehicle control. FITC-albumin leaking into extra-vascular space was accumulated over time. Fluorescent images were acquired every five minutes for one hour. Protein flux was quantified using the formula IOI_Rel=(I_i-I_o)/I_i, where I_i=intensity inside the vessel and Io=intensity outside the vessel.

HUVEC cell culture

Human umbilical vein endothelial cells (HUVECs) were purchased from Lonza and grown in Endothelial Cell Basal Medium supplemented with EGM-2 MV Bulletkits (Lonza, MD, USA), or from PromoCell and grown in Endothelial Cell Growth Medium 2 supplemented with SupplementMix C-39216 (PromoCell GmbH, Heidelberg, Germany). Cells were seeded onto 0.1% gelatin-coated plates and incubated in 5% CO₂ humidified incubator at 37°C for 2–3 days past confluence for use in experimental assays.

Transendothelial electrical resistance

HUVECs were seeded onto 8W10E+ PET electrode arrays to be used with an Electric Cell-Substrate Impedance Sensing (ECIS) system (Applied Biophysics, Troy, NY). Transendothelial electrical resistance (TER) was continuously recorded over time as an indicator of cell-cell adhesive barrier function [22]. Cells were stimulated with human recombinant H3Cit (Item No. 17926; Cayman Chemical, Ann Arbor, MI) or vehicle control (0.1% BSA in PBS) with or without inhibitors. Representative tracings are presented normalized to baseline. Change in resistance was quantified by subtracting the lowest resistance value from the baseline resistance value of each ECIS well and averaging maximum change from baseline at each concentration. Inhibitors used in this study include Rho kinase inhibitor Y27632 (Cayman), Rho inhibitor Rhosin (Calbiochem), and MLCK inhibitor peptide 18 (Cayman). Barrier enhancing agent forskolin (MP Biomedicals) was used as a pharmacological therapeutic.

Cell Viability Assay

LIVE/DEAD™ Viability/Cytotoxicity Kit, for mammalian cells (Cat. No L3224; Invitrogen) was used per manufacturer’s instructions. Absorbance was read at 530 nm (calcein AM) to determine live cells and at 645 nm (ethidium homodimer-1) to determine dead cells.

Immunofluorescence confocal microscopy

HUVECs were seeded onto coverslips and treated with human recombinant H3Cit or vehicle control for one minute for immunofluorescence staining as we’ve previously described. For adherens junction staining, cells were fixed with 4% paraformaldehyde for 10 minutes at room temperature, washed twice with PBS, blocked with 10% donkey serum in PBS for one hour, and labeled with VE-cadherin (D87F2) XP® Rabbit mAb #2500 (CST; 1:500) overnight at 4°C in a humidified chamber. The next day, cells were washed with PBS, incubated with donkey anti-rabbit IgG Alexa Fluor 488 (Invitrogen; Cat #R37118) for one hour at room temperature, washed again with PBS, and mounted to slides with ProLong Diamond Antifade Mountant with DAPI (Life Technologies). Fluorescent confocal images were captured with an Olympus FV1200 Laser Scanning Confocal Microscope. VE-cadherin intensity was analyzed with FIJI (version 2.0.0-rc-65/1.51w) software by taking the sum of Z-stack (6 slices), subtracting the background (radius=10), applying Gaussian blur (1.00), converting to mask, and analyzing the particles (size 50-Infinity, circularity 0–0.25) per field of view (~212 μm²). Actin staining was performed in the same manner, except cells were permeabilized with 0.1% PBS-T for 5 minutes before blocking and were stained with Alexa Fluor 568 Phalloidin (F-actin) and Alexa Fluor 488 DNAse I (G-actin) for 20 minutes at room temperature. Intensity of each type of actin was analyzed using FIJI (version 2.0.0-rc-65/1.51w) software by taking the sum of Z-stack (10 slices) and measuring total intensity per field of view (~318 μm²), represented as F:G-actin ratio.

Statistical analysis

Statistics were performed using GraphPad Prism (version 6.0f/7.0d). Pairwise-comparisons were made using unpaired two-tailed t-test and group-wise comparisons were made using ordinary one-way ANOVA with Tukey’s post hoc multiple comparisons test. Statistical significance was defined as p ≤ 0.05.

H3Cit causes microvascular leakage and endothelial barrier dysfunction without cell death

To determine the direct effects of H3Cit in the microvascular endothelial barrier, we used intravital microscopy to examine the protein transvascular flux and found that H3Cit (10 μg/mL) caused extravasation of fluorescent-labeled albumin across microvessels (Fig. 1). Furthermore, we used the ECIS system to measure TER, an indicator of cell-cell adhesive barrier strength and found that stimulation of HUVECs with H3Cit decreased TER in a concentration-dependent manner, indicating endothelial barrier dysfunction (Fig. 2A and B); this response was not due to cell death (Fig. 2C). Taken together, we concluded that H3Cit is capable of disrupting the microvascular endothelial barrier without causing cell toxicity.

H3Cit-mediated endothelial barrier dysfunction is not dependent on Rho or MLCK signaling pathways

Limited information is available regarding the receptor and signaling mechanisms underlying histone-induced endothelial injury. Similar to other components of NETs, such as extracellular DNA, histones can be considered to act as a pattern recognition molecule. So far, cellular pathways seemingly activated by extracellular histones include toll-like receptor [23, 24] and inflammasome [25, 26] signaling, as well as membrane integration and calcium influx [2, 27]. However, modifications of histones can affect their size, charge, and structure [28], which could change the way they interact with the endothelium. For example, we were unable to prevent H3Cit-induced endothelial barrier dysfunction with TLR4 antagonism (data not shown). Thus, it is important to distinguish the specific effects of different histones like H3Cit that are modified during pathological conditions so that more specific targeted therapies can be developed.

In attempt to determine the signaling events responsible for H3Cit-mediated endothelial barrier dysfunction, we investigated two well-known pathways involved in barrier regulation, namely the Rho and myosin light chain kinase (MLCK) pathways. Endothelial paracellular permeability is dynamically regulated by intercellular junctions that are anchored to the cytoskeleton where the activation status of myosin light chain (MLC) determines the contractile status of cells. MLC phosphorylation, which can be increased by both the Rho (MLC phosphatase inhibition) and MLCK (MLC kinase activation) signaling pathways, triggers cell contraction and subsequent junction opening [22]. Surprisingly, inhibition of Rho pathway with rho-associated protein kinase (ROCK) inhibitor Y27632 (Fig. 2D) or Rho GEF binding domain inhibitor Rhosin (Fig. 2E) was not sufficient to prevent the TER drop induced by H3Cit. Similarly, inhibition of MLCK pathway with MLCK inhibitor peptide 18 was unable to reduce H3Cit-mediated endothelial barrier dysfunction (Fig. 2F). Therefore, we suggest that H3Cit-induced barrier opening is not dependent on MLCK or Rho activity; rather, it may affect through actin polymerization.

H3Cit causes EC barrier dysfunction by opening adherens junctions and reorganizing the actin cytoskeleton

To further investigate potential molecular mechanisms of H3Cit-induced barrier dysfunction, we sought to determine its effects on endothelial cell-cell adherens junctions and the actin cytoskeleton. Polymerization and redistribution of the actin cytoskeleton to form stress fibers in endothelial cells promotes the opening of cell-cell junctions [29]. Stimulation of HUVECs with H3Cit caused rapid thinning of adherens junction protein VE-cadherin at cell-cell borders coupled with intercellular gap formation (Fig. 3A and B). Furthermore, H3Cit invoked centralized F-actin bundles and an increase in the ratio of filamentous (F)-actin to globular (G)-actin, indicating actin stress fiber formation (Fig. 3C and D). In consistence with our finding, histones have recently been shown to polymerize G-actin and bundle with F-actin [30], supporting the possibility of histone-actin interactions and subsequent endothelial barrier opening. We conclude that H3Cit contributes to endothelial barrier dysfunction by opening adherens junctions and reorganizing the actin cytoskeleton.

Barrier-enhancing adenylyl cyclase activator forskolin prevents H3Cit-mediated barrier disruption

Although the precise endothelial cell signaling mechanisms activated by H3Cit responsible for barrier dysfunction warrants further investigation, we sought to determine whether a known barrier enhancer was capable of preventing the response therapeutically. The pharmacological agent forskolin, typically recognized as a barrier enhancer due to its activation of adenylyl cyclase, contributes to actin cytoskeleton stabilization [29]. Indeed, enhancing the endothelial barrier with forskolin (10 μM, 30 min pre-treatment) substantially attenuated the TER response to H3Cit (Fig. 4A and B). Therefore, enhancing the endothelial barrier therapeutically by stabilizing the actin cytoskeleton may be an option to protect against H3Cit-mediated tissue injury.

In summary, this study demonstrates novel evidence for H3Cit-induced endothelial barrier dysfunction leading to microvascular leakage characterized by opening of adherens junctions and reorganization of the actin cytoskeleton; the response can be prevented by enhancing the barrier with the adenylyl cyclase activator forskolin. Targeting H3Cit therapeutically may hold potential to limit the exacerbation of the inflammatory response during a number of conditions.