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
Corresponding author. Cardiovascular Research Centre (CVRC), School of Medical Sciences, Örebro University, Södra Grev Rosengatan 32, 703 62, Örebro, Sweden.
School of Medical Sciences, Örebro University, Örebro, SwedenCardiovascular Research Centre (CVRC), School of Medical Sciences, Örebro University, Örebro, Sweden
School of Medical Sciences, Örebro University, Örebro, SwedenCardiovascular Research Centre (CVRC), School of Medical Sciences, Örebro University, Örebro, Sweden
School of Medical Sciences, Örebro University, Örebro, SwedenCardiovascular Research Centre (CVRC), School of Medical Sciences, Örebro University, Örebro, Sweden
School of Medical Sciences, Örebro University, Örebro, SwedenCardiovascular Research Centre (CVRC), School of Medical Sciences, Örebro University, Örebro, Sweden
1 These authors share last authorship of this work.
Liza U. Ljungberg
Footnotes
1 These authors share last authorship of this work.
Affiliations
School of Medical Sciences, Örebro University, Örebro, SwedenCardiovascular Research Centre (CVRC), School of Medical Sciences, Örebro University, Örebro, Sweden
1 These authors share last authorship of this work.
Ashok K. Kumawat
Footnotes
1 These authors share last authorship of this work.
Affiliations
School of Medical Sciences, Örebro University, Örebro, SwedenCardiovascular Research Centre (CVRC), School of Medical Sciences, Örebro University, Örebro, Sweden
IL-6 trans-signaling induces LAMA4-to-LAMA5 switch in endothelial cells, which is accompanied by secretion of several inflammatory mediators.
•
These alterations contribute to favored trans-endothelial migration of mononuclear cells over granulocytic cells.
•
In human atherosclerotic plaques, the expression of LAMA4 and LAMA5 is altered and correlates with markers of granulocytic and lymphocytic cells.
Abstract
Background and aims
Laminins are essential components of the endothelial basement membrane, which predominantly contains LN421 and LN521 isoforms. Regulation of laminin expression under pathophysiological conditions is largely unknown. In this study, we aimed to investigate the role of IL-6 in regulating endothelial laminin profile and characterize the impact of altered laminin composition on the phenotype, inflammatory response, and function of endothelial cells (ECs).
Methods
HUVECs and HAECs were used for in vitro experiments. Trans-well migration experiments were performed using leukocytes isolated from peripheral blood of healthy donors. The BiKE cohort was used to assess expression of laminins in atherosclerotic plaques and healthy vessels. Gene and protein expression was analyzed using Microarray/qPCR and proximity extension assay, ELISA, immunostaining or immunoblotting techniques, respectively.
Results
Stimulation of ECs with IL-6+sIL-6R, but not IL-6 alone, reduces expression of laminin α4 (LAMA4) and increases laminin α5 (LAMA5) expression at the mRNA and protein levels. In addition, IL-6+sIL-6R stimulation of ECs differentially regulates the release of several proteins including CXCL8 and CXCL10, which collectively were predicted to inhibit granulocyte transmigration. Experimentally, we demonstrated that granulocyte migration is inhibited across ECs pre-treated with IL-6+sIL-6R. In addition, granulocyte migration across ECs cultured on LN521 was significantly lower compared to LN421. In human atherosclerotic plaques, expression of endothelial LAMA4 and LAMA5 is significantly lower compared to control vessels. Moreover, LAMA5-to-LAMA4 expression ratio was negatively correlated with granulocytic cell markers (CD177 and myeloperoxidase (MPO)) and positively correlated with T-lymphocyte marker CD3.
Conclusions
We showed that expression of endothelial laminin alpha chains is regulated by IL-6 trans-signaling and contributes to inhibition of trans-endothelial migration of granulocytic cells. Further, expression of laminin alpha chains is altered in human atherosclerotic plaques and is related to intra-plaque abundance of leukocyte subpopulations.
The vascular endothelium, a monolayer of endothelial cells (ECs) that line the inside of blood vessels, is separated from the rest of the vessel wall by a thin layer of specialized extracellular matrix known as the endothelial basement membrane [
]. The assembly of endothelial basement membrane is primarily determined by laminins, large glycoproteins (400–900 kDa) that form networks with other proteins/glycoproteins such as collagen type IV, nidogen (entactin) and perlecan [
]. There are five α, four β, and three γ chain variants discovered so far in humans, which in combination give rise to 16–18 different laminin isoforms [
]. Loss-of-function studies using mice models have demonstrated that genetic deletion of almost any of the laminin chains leads to embryonic lethality signifying its vital role during development [
]. The systematic and widely accepted nomenclature of laminin isoforms is based on the specific αβγ chain composition. For example, an isoform made up of α1, β1, and γ1 chains is referred as Laminin111 or LN111.
The endothelial basement membrane is predominantly made up of α4 (LAMA4) and α5 (LAMA5) containing isoforms (i.e. LN521/LN511, LN421/LN411) with some degree of variations depending on the developmental stage, as well as size and type of vasculature [
]. The expression of LAMA4 starts during early embryogenesis and is expressed in all types of vessels throughout postnatal life. However, LAMA5 chain expression appears postnatally, with patchy expression in venules and postcapillary venules and higher expression in capillaries and arterioles, while some larger arteries lack LAMA5 expression [
]. Laminins provide structural support and anchoring to ECs in addition to their contribution in transduction of mechano-sensing signals, maintenance of barrier function and overall stability of blood vessels [
]. Murine ECs cultured on LAMA4 containing isoforms favor transmigration of neutrophils while ECs grown on LAMA5 containing isoforms preferentially allow transmigration of T-lymphocytes [
Mechanisms that govern the temporal and spatial expression of laminins across the vasculature under pathophysiological conditions are largely unknown. Inflammation and aging have been suggested to modulate expression of laminins in various extracellular matrices including in endothelial basement membrane [
]. In addition, a strong upregulation of α4 laminin expression by ECs has been shown in vitro in response to pro-inflammatory agents such as TNF-α, IL-1β and LPS [
Endothelial cell laminin isoforms, laminins 8 and 10, play decisive roles in T cell recruitment across the blood-brain barrier in experimental autoimmune encephalomyelitis.
IL-6 is a pleotropic cytokine that has a role in both pro-inflammatory and regenerative or homeostatic functions. The so-called classic-signaling encompasses binding of IL-6 to a heterodimer receptor consisting of membrane-bound IL-6Rα and gp130. Whereas IL-6 trans-signaling is initiated through binding of IL-6 to a soluble form of IL-6R (sIL-6R) which then binds to a membrane bound gp130 [
]. Despite the general assumption that expression of membrane bound IL-6R is restricted to few types of cells like hepatocytes and immune cells, we have recently shown that ECs also express the membrane bound IL-6R and gp130 and hence, respond to both IL-6 classic-as well as trans-signaling [
Activation of the JAK/STAT3 and PI3K/AKT pathways are crucial for IL-6 trans-signaling-mediated pro-inflammatory response in human vascular endothelial cells.
]. While IL-6 classic-signaling appears to regulate basic endothelial homeostatic functions, IL-6 trans-signaling induces expression of a pro-inflammatory chemokine MCP-1 and impaired angiogenic response of ECs [
Activation of the JAK/STAT3 and PI3K/AKT pathways are crucial for IL-6 trans-signaling-mediated pro-inflammatory response in human vascular endothelial cells.
]. Moreover, several mice studies demonstrated that interfering with IL-6 pathway, specifically IL-6 trans-signaling pathway, is protective of atherosclerotic plaque development [
]. These findings strongly indicate the involvement of IL-6 in the development of atherosclerosis. Activation of endothelial cells is included among the proposed mechanisms that IL-6 mediates its atherogenic effect. In the current study, we aimed to investigate the role of IL-6 in regulating endothelial laminin production and characterize the impact of altered laminin composition on the phenotype, inflammatory response, and function of ECs. We also aimed to investigate laminin composition in atherosclerotic plaques in association with the immune cell composition at site.
2. Materials and methods
2.1 Cell culturing
We cultured Human Umbilical Vein Endothelial Cells (HUVECs) and Human Aortic Endothelial Cells (HAECs), both from Life technologies (USA) in 75 cm2 flasks (Sarstedt, Germany) containing complete endothelial medium [VascuLife basal medium supplemented with VEGF or ENGS LifeFactors kit (LifeLine Cell Technologies, USA)] and antibiotics [Penicillin (0.1U/ml) + Streptomycin (100 ng/ml)-PEST, Gibco, Life Technologies, USA]. In an environment kept at 37 °C and 5% CO2, the cultures were maintained until passage 10. Medium was replaced every 48–72 h.
2.2 Coating plates with laminin
Cell culture plates (Sarstedt, Germany) and glass-slide chambers (ibidi, Germany) were coated with 2 μg/cm2 of laminin isoforms (BioLamina, Sweden) for 2 h at 37 °C or overnight at 4 °C. The laminin isoforms were diluted in PBS containing Mg2+ and Ca2+.
2.3 Treatment of HUVECs
HUVECs were seeded at cell densities of 3 × 105 cells/well in 6-well plates and 6 × 104 cells/well in 24-well plates containing complete endothelial medium containing antibiotics. After overnight incubation, the medium was replaced with fresh antibiotics free medium and cells were stimulated with recombinant human proteins IL-6 and sIL-6R (both from R&D systems, USA) for 24 h–72 h. Culture supernatants and cells were collected at the end of incubation and kept at −80 °C until further analysis.
2.4 RNA isolation and cDNA synthesis
Using E.Z.N.A® Total RNA Kit (OMEGA bio-tek inc, USA), RNA was extracted from frozen cells. Following manufacturer's instruction, cells were lysed with 1:1 mixture of TRK lysis buffer containing 2% β-Mercaptoethanol and 70% ethanol. The cell lysates were transferred into HiBind RNA columns, centrifuged for 1min at 10,000g, and washed three times. RNA was eluted using RNase free water and concentrations were determined using NanoDrop™ 2000 (Thermo Fisher Scientific, USA) spectrophotometer.
Synthesis of cDNA was achieved using high-capacity cDNA reverse transcription kit (Thermo Fisher Scientific, USA) following manufacturer's instructions. In total reaction volume of 20 μl, 1 μg total RNA was mixed with master mix composed of buffer, random primers, dNTPs and reverse transcriptase enzyme. The preparation was allowed to run in thermocycler for: 10 min at 25 °C, 120 min at 37 °C, 5 min at 85 °C and kept at 4 °C before storage at −20 °C.
2.5 Real-Time PCR
TaqMan qPCR primers/probes (Applied Biosystems, Life technologies, USA), were used to analyze gene expression. A total reaction volume of 10 μl consisting of LuminoCt qPCR ready mix (Sigma-Aldrich, USA), TaqMan Primer/Probe (Applied Biosystems, Life technologies, USA), water and cDNA was allowed to run in QuantStudio 7 Flex Realtime PCR system (Applied Biosystems, Foster City, USA). The cycling condition used was as follows: at 95 °C for 1 s and at 60 °C for 20 s for 40 cycles in addition to one step initialization at 95 °C for 20 s. GAPDH was used as housekeeping gene to normalize relative quantities recorded for each well.
2.6 Protein extraction and quantification
HUVECs were lysed using ice-cold RIPA lysis buffer (Millipore, USA) and quantity of protein in the lysates were analyzed using Micro BCA™ Protein Assay kit (Thermo Scientific, USA). Absorbance was measured at 540 nm using Cytation 3 Imaging reader (BioTek, Winooski, USA).
2.7 Immuno-(Western) blotting
A mixture of cell lysates and SDS sample buffer was denatured for 5 min at 95 °C. Electrophoretic separation was achieved by loading the mixture (10–80 μg of protein/well) into 4–12% NuPAGE® Novex Bis-Tris gels and 3–8% NuPAGE® Novex Tris-Acetate gels combined with respective running buffers i.e. MOPS and TA SDS running buffers (both Invitrogen, USA). MagicMark™ XP Western Protein Standard (Invitrogen, USA) was used to determine molecular masses of the proteins. Proteins were blotted onto Immobilon-FL PVDF membranes (Millipore, USA). TBS-T (10 mM Tris-HCl pH 8.0, 150 mM NaCl, 0.1% (w/v) Tween-20) was used in further steps. For detecting proteins, membranes were incubated with the following primary antibodies: anti-Laminin-alpha4 antibody (Abcam, United Kingdom, ab242198; 1:750 dilution), anti-Laminin-alpha5 antibody (Abcam, United Kingdom, ab210957; 1:750 dilution), anti-VE-cadherin antibody (Abcam, United Kingdom, ab33168; 1 μg/ml), anti-PECAM-1 antibody (Abcam, United Kingdom, ab24590; 1 μg/ml) and anti-β-Tubulin antibody (Millipore, USA, #05–661; 1:2000 dilution). Following this, the membranes were incubated with horseradish peroxidase (HRP)-conjugated goat anti-rabbit IgGs (Cell Signaling Technology, USA, #7074; 1:2000) or horse anti-mouse IgGs (Cell Signaling Technology, USA, #7076; 1:2000). Protein bands were visualized using Immobilon™ Western Chemiluminescent HRP Substrate solution from Millipore (Millipore, USA), and chemiluminescence was recorded by a Li-Cor Odyssey Fc imager and analyzed with Image Studio Software (both Li-Cor Biotechnology UK Ltd., United Kingdom).
2.8 Immunofluorescent staining
Immunofluorescence staining of cells cultured on laminins or treated with IL-6+sIL-6R was performed using anti-Laminin-alpha4 antibody (Abcam, United Kingdom, ab242198; 1:200 dilution), anti-Laminin-alpha5 antibody (Abcam, United Kingdom, ab210957; 1:200 dilution), anti-VE-cadherin antibody (Abcam, United Kingdom, ab33168; 1:150 dilution) and anti-PECAM-1 antibody (Abcam, United Kingdom, ab24590; 1:300 dilution). Antibodies were diluted in TBS containing 2% BSA. Briefly, cells were fixed with ice cold 4% paraformaldehyde for 40min at room temperature. Following rinsing with PBS, cells were permeabilized with 1% BSA prepared in 0.1% Triton X-100 for 30min at room temperature. PBS was used in subsequent washing steps. Cells were incubated with primary antibodies for 1 h followed by 1 h incubation with secondary antibodies [anti-mouse IgG (Alexa Flour®488) Abcam, United Kingdom, ab150113; 1:1000 dilution; anti-rabbit IgG (Alexa Flour®647) Abcam, United Kingdom, ab15005; 1:1000 dilution]. The slides were incubated with DAPI for 5min in the dark, air dried and mounted in PERTEX (Histolab, Sweden). Stained slides were analyzed using Leica SP8 UV/Visible Laser Confocal Microscope (Leica Microsystems, Germany).
2.9 Proteomics analyses
Using Cardiovascular III and Inflammation panels by Olink proteomics (Uppsala, Sweden), the release of up to 184 proteins were analyzed in cell culture supernatants. The analysis employs proximity extension assay (PEA) where two antibodies tagged with complementary nucleotide chains are targeted against two epitopes of a protein. Up on binding to the protein, the nucleotide chains hybridize and can then be amplified with PCR. The PCR product is proportional to the amount of the target protein which is reported as normalized protein expression (NPX) on a log2 scale. Statistical analysis of the NPX data was performed using t-test followed by multiplicity correction with Benjamini–Hochberg false discovery rate (FDR). This proteomic data was further analyzed in Ingenuity Pathway Analysis (IPA®, QIAGEN Inc., The Netherlands, https://www.qiagenbioinformatics.com/products/ingenuity-pathway-analysis) online tool [
]. Enriched functions by the IL-6 trans-signaling regulated proteins were identified by performing a core analysis in IPA using a cut-off value of 1.5 for fold change (FC) in protein level and FDR value of 20%. The IPA analyses was performed on 12 May, 2020.
2.10 Sandwich ELISA
Quantification of secreted proteins from culture supernatants was performed using sandwich ELISA kits (R&D systems, USA) following the manufacturer's instructions. Optical densities at the end of the assay were recorded at 450 nm using Cytation 3 Imaging reader (BioTek, Winooski, USA).
2.11 Matrix adhesion assay
HUVECs were labeled with BCECF-AM (2′,7′-bis-(2-carboxyethyl)-5-(and-6)-carboxyfluorescein, acetoxymethyl ester; Thermo Fisher Scientific, USA). On to laminin coated 96-well plates, the labeled HUVECs were seeded at 5 × 104 cells/well density and incubated for 1 h in a 37 °C and 5% CO2 environment. The fluorescence intensity of the labeled cells in each well was measured (excitation and emission at 439 nm and 535 nm respectively) using Cytation 3 Imaging reader (BioTek, Winooski, USA). This measurement was repeated after removing non-adherent cells by washing the wells with endothelial medium (VascuLife basal medium, LifeLine Cell Technologies, USA) containing 0.05% BSA (Gibco, Life Technologies, USA). The results were calculated as percentage of laminin-specific adhesion after washing against initial cell input.
2.12 Quantifying cell number
Cell counting kit-8 (CCK-8, Sigma-Aldrich, USA) assay was used to determine number of HUVECs cultured on LN421 and LN521 or on uncoated plastic. Briefly, HUVECs were seeded (4 × 103 cells/well) in laminin-coated 96-well plates using complete endothelial medium containing antibiotics. After overnight incubation, the medium was replaced with antibiotic-free medium, and incubation continued for 48 h. Following manufacturer's instruction, 10 μl of the CCK-8 solution (Tetrazolium salt) was added to each well and incubated for up to 4 h. Optical density of the formazan dye formed was measured at 450 nm using Cytation 3 Imaging reader (BioTek, Winooski, USA).
2.13 Leukocyte migration assays
Granulocytes and peripheral blood mononuclear cells (PBMCs) were isolated form peripheral blood of healthy donors collected in EDTA tubes. First PBMCs were isolated using Lymphoprep™ density gradient medium (Stemcell Technologies, Canada) following manufacturer's instructions. From the sediment, granulocytes were isolated by lysing the RBCs using a lysis buffer (BD BioSciences, USA). Isolation of T-lymphocutes was achived using EasySep™ Human CD3 Positive Selection Kit II (STEMCELLTechnologies, Canada). The migration assay was performed using the CytoSelect™ kit (Cell Biolaba Inc, USA) according to manufacturer's instructions. Briefly, leukocytes were labeled with a fluorochrome (leukotracker, Cell Biolabs Inc, USA) and added on to the apical side of trans-well inserts (2 × 105/well) with or without cultured ECs, or on trans-well inserts coated with laminin isoforms (2 μg/cm2). Leukocytes were allowed to migrate for 3–24 h, and the migrated cells were resuspended and lysed with a buffer. A sample of 150 μl of the lysate was transferred to a 96-well plate and fluorescence was detected at 485nm/535 nm for excitation and emission using Cytation 3 Imaging reader (BioTek, Winooski, USA). Blood samples were collected from healthy donors with an informed consent and the protocol has been approved by the regional ethics committee.
2.14 Proteomic and transcriptomic profiling of atherosclerotic and normal vessels
The gene and protein expression of laminin chains as well as immune cell markers was studied using the Biobank of Karolinska Endarterectomies (BiKE), a cohort consisting of carotid plaques taken from patients undergoing endarterectomy. Detailed description of the cohort has been published previously [
]. Patients were grouped as ‘symptomatic’ and ‘asymptomatic’. Symptomatic patients (n = 87) were identified with the presence of transient ischemic attack, minor stroke, and retinal transient ischemic attack. Patients lacking these symptoms during 6 months before the surgery were grouped as asymptomatic patients (n = 40). Atherosclerosis-free iliac and radial arteries (n = 9) and aorta (n = 1) obtained from organ donors without a history of cardiovascular disease were used as control/normal vessels. The study was approved by the regional ethics committee and the investigation conformed to the principles outlined in the Declaration of Helsinki. Samples were collected with informed consent from patients/organ donor or their guardians.
RNA extracts were prepared from the vessels as described previously [
] and microarray analyses was performed using Affymetrix HG-U133 plus 2.0 Genechip arrays. The microarray dataset is available at the NCBI GEO data repository (accession number GSE125771). In addition, proteomic analysis on plaques and adjacent peripheral tissue (n = 18 patients) was achieved using liquid chromatography mass spectrometry/mass spectrometry (LC-MS/MS). Details of the sample preparation were previously described [
Further, 5 μm sections obtained from endarterectomy specimens were used for immunohistochemical staining using primary antibodies against LAMA4 (Abcam, United Kingdom, ab242198) and LAMA5 (Abcam, United Kingdom, ab210957). Isotype mouse IgG was used as negative control and all antibodies were diluted in Da Vinci Green solution. Briefly, sections were deparaffinized with tissue clear followed by rehydration in ethanol and water. Antigen retrieval was achieved by high pressure boiling in DIVA buffer. The sections were then blocked with background sniper and incubated with primary antibodies against LAMA4 and LAMA5 at room temperature for 1 h. The sections were further incubated in probe-polymer reagent tagged with alkaline phosphatase followed by Warp Red solution. The sections were finally Hematoxylin-counterstained, dehydrated and mounted using PERTEX (Histolab, Sweden). Microscopic analyses and image acquisition were performed using Nikon OPTIPHOT-2 microscope. The control sections used in the immunohistochemical staining were from a control internal carotid artery from a 61-year-old man undergoing surgical excision of a neck tumor.
2.15 Statistical analysis
Data were analyzed using GraphPad Prism® statistical software version 9.0 (GraphPad Software, Inc., USA). Data are presented as mean ± standard error of mean (SEM) of at least 3 sets of independent experiments. For comparison between groups, one-way ANOVA for repeated measures followed by Bonferroni post-hoc test, paired t-test, and One sample t-test were used. Pearson correlation was used to assess the relationship between the expression of laminin alpha chains and granulocyte markers in human atherosclerotic plaques. p value less than 0.05 was considered as statistically significant.
3. Results
3.1 IL-6 trans-signaling alters the production of endothelial laminin alpha chains
To investigate the effect of IL-6 trans-signaling on laminin production by ECs, we stimulated HUVECs with IL-6 in combination with sIL-6R and analyzed the expression of laminin chains at mRNA and protein levels. We found that activation of IL-6 trans-signaling in vascular ECs significantly downregulated gene expression of LAMA4 while upregulating gene expression of LAMA5 (Fig. 1A). Furthermore, we saw that the downregulated LAMA4 and upregulated LAMA5 expressions by IL-6 trans-signaling were evident on protein level (Fig. 1B–E). In contrast, HUVECs stimulation with IL-6 alone (classic-signaling) did not affect the expression of either LAMA4 or LAMA5 (Supplementary Fig. 1). To validate these findings in ECs from adult vasculature, we treated human aortic ECs (HAECs) with IL-6 in combination with sIL-6R and assessed expression of LAMA4 and LAMA5. We found that LAMA4 was downregulated while LAMA5 was upregulated (Supplementary Fig. 2). Moreover, the effects of IL-6 trans-signaling on endothelial LAMA4 and LAMA5 were reversed by soluble gp130 (sgp130), the natural antagonist of IL-6 trans-signaling (Fig. 1B and C). However, IL-6 trans-signaling in HUVECs did not affect gene expression of laminin beta chains (LAMB1 or LAMB2) and LAMC1 (Supplementary Fig. 3). Collectively, these findings demonstrate that IL-6 trans-signaling in ECs induces a switch in laminin alpha chain expression from LAMA4 to LAMA5.
Fig. 1IL-6 trans-signaling induces LAMA4 to LAMA5 switch in human vascular ECs.
(A) Gene expression of LAMA4 and LAMA5 in HUVECs exposed to IL-6 and sIL-6R (100 ng/ml each) for 24–72 h. Representative immunoblot and bar graphs showing (B) LAMA4 and (C) LAMA5 expression in ECs exposed to IL-6 and sIL-6R (100 ng/ml each) (48 h) in the presence or absence of sgp130Fc (1 μg/ml). Immunofluorescence staining showing expression of (D) LAMA4 and (E) LAMA5 in unpermeabilized HUVECs with or without treatment with IL-6 and sIL-6R (100 ng/ml each) for 48 h. *p < 0.05, **p < 0.01, ***p < 0.001 and ns (non-significant) compared to control. ††p < 0.01 comparing IL-6/sIL-6R/sgp130Fc to IL-6/sIL-6R.
3.2 Endothelial LAMA4 to LAMA5 switch induced by IL-6 trans-signaling is accompanied with altered release of inflammatory mediators
To further characterize the response of ECs to IL-6 trans-signaling that accompany the switch in laminin expression, we analyzed the release of 184 proteins (Olink® proteomics; Inflammation and Cardiovascular III panels) from supernatants of human vascular ECs stimulated with IL-6 in combination with sIL-6R for 48 h. We found that a total of 97 proteins were above detection limits, out of which IL-6 trans-signaling significantly altered the release of 20 proteins (3 downregulated, 17 upregulated) compared to untreated controls (p-value <0.05, FDR 5%) (Fig. 2A). Supplementary Table 1 shows the full list of detected proteins with respective fold changes, p-values, and FDR. In contrast, IL-6 classic-signaling moderately upregulated secretion of only 3 proteins (CXCL11, CCL23 and HGF) as shown in Supplementary Fig. 4. The proteomic data was further analyzed in Ingenuity Pathway Analysis online tool (IPA®, QIAGEN Inc) to identify functions and networks that are regulated by IL-6 trans-signaling. The top enriched functions were associated with adhesion/binding and migration of leukocytes. We found that the differentially regulated proteins by IL-6 trans-signaling resulted in enrichment of the functions ‘Binding of granulocytes’ (p-value = 2.26E-08) and ‘Migration of granulocytes’ (p-value = 2.14E-11) (Fig. 2B). Simultaneously, IL-6 trans-signaling regulated proteins enriched the functions ‘Binding of lymphocytes’ (p-value = 7.45E-08) and ‘Lymphocyte migration’ (p-value = 2.18E-17) (Fig. 2C). In addition, the IPA® analyses predicted that the differential regulation of proteins due to IL-6 trans-signaling leads to inhibition of binding and migration of granulocytes (Fig. 2B) while simultaneously leading to activation of binding and migration of lymphocytes (Fig. 2C). Interestingly, we noticed that 9 of the differentially regulated proteins by IL-6 trans-signaling enriched the functions associated with both granulocyte and lymphocyte recruitment suggesting for contrasting impacts of those proteins to favor trans-migration of a specific leukocyte sub-population over the other. Previously, we have reported that IL-6 trans-signaling downregulates CXCL8 and upregulates CXCL10 gene expression in a time-dependent fashion [
]. In this study, we analyzed the gene expression of additional key proteins and found that IL-6 trans-signaling resulted in upregulated gene expression of CXCL9, CXCL11 and CCL23 while downregulating gene expression of AXL (Fig. 2D). Moreover, we confirmed the IL-6 trans-signaling induced changes in secretion of endothelial CXCL8, CXCL9, CXCL10 as well as CCL-2 using HAECs (Supplementary Fig. 5). Overall, our data suggest that IL-6 trans-signaling alters the secretion of proteins/chemokines from ECs which inhibit binding and migration of granulocytes and promote binding and migration of lymphocytes.
Fig. 2IL-6 trans-signaling regulates secretion of proteins from human vascular ECs that control immune cell binding and migration.
(A) Volcano plot showing the fold change (FC) and False Discovery Rate (FDR) of proteins released from HUVECs stimulated with IL-6/sIL-6R (100 ng/ml each) for 48 h. FDR was calculated using Benjamini-Hochberg method and the dotted line marks FDR = 5%. Ingenuity pathway analyses showing that the differentially regulated proteins enrich pathways associated with binding and migration of (B) granulocytes and (C) lymphocytes. The red color indicates upregulation in release of proteins while green indicates downregulation. Blue lines indicate that a protein leads to predicted inhibition of function while orange lines indicate predicted activation of function. The yellow lines show disagreement between state of the differentially regulated protein expression and the predicted state of function (i.e., binding or migration). (D) qPCR analyses showing the time-course of relative expression of CXCL9, CXCL11, CCL23 and AXL in HUVECs treated with a combination of IL-6 and sIL-6R (100 ng/ml each). *p < 0.05, ***p < 0.001, compared to respective controls.
Chemokines released from ECs induce activation of integrin receptors on leukocytes that are needed for establishing firm adhesion with endothelial adhesion molecules such as ICAM-1 and VCAM-1 [
]. Although most of such adhesion molecules are common for both granulocytes and lymphocytes, some endothelial proteins that are involved in diapedesis have been implied to preferably allow migration of certain leukocyte subpopulations such as CD99L2 [
]. Hence, we investigated whether IL-6 trans-signaling regulates expression of proteins involved in diapedesis of leukocyte. We have previously reported that both ICAM-1 and VCAM-1 are significantly upregulated in HUVECs in response to IL-6 trans-signaling activation [
Activation of the JAK/STAT3 and PI3K/AKT pathways are crucial for IL-6 trans-signaling-mediated pro-inflammatory response in human vascular endothelial cells.
]. We found that the gene expression of JAMA, JAMAC, CD99 and CD99L2 in HUVECs were not regulated by IL-6 trans-signaling activation (Supplementary Fig. 6).
To assess whether the LAMA4 to LAMA5 switch contributes to the IL-6 trans-signaling induced changes in secretion of inflammatory mediators, we cultured HUVECs on laminin isoforms LN421 and LN52 and analyzed culture supernatants for basal as well as IL-6 trans-signaling induced release of inflammatory mediators. For this, we used the Olink® proteomics (Inflammation panel) platform and found that out of the 92 proteins in the panel, 44 proteins were above the detection limit. The basal levels of proteins released from ECs were comparable between those cultured on LN421 and LN521 (Supplementary Fig. 7). Similarly, there was no difference in IL-6 trans-signaling induced release of inflammatory mediators from ECs cultured on LN421 and LN521 in comparison to ECs cultured on plastic (Fig. 3A). The impact of cell proliferation could be ruled out as cell number after an overnight and 48 h of incubation was comparable between ECs cultured on LN421 versus LN521 (Supplementary Fig. 8). Nevertheless, we noted that adherence of unstimulated ECs to wells coated with LN521 was significantly higher than those coated with LN421 during the first hour after seeding (Fig. 3B). Moreover, unstimulated ECs cultured on LN521 showed upregulated VE-cadherin expression (48 h), but no difference in PECAM-1, compared to those cultured on LN421 (Fig. 3D and E). Under bright field microscope, the morphology of ECs appeared similar regardless of the laminin isoform the ECs were cultured on (Supplementary Fig. 9). Overall, these results suggest that the LAMA4 to LAMA5 transition induced by IL-6 trans-signaling has negligible impact in release of inflammatory mediators from ECs. Further, our findings imply that LAMA5 supports a more rapid adhesion of ECs which is accompanied by increased expression of VE-cadherin that might enhance the tightness of endothelial inter-cellular junctions.
Fig. 3Properties of human vascular ECs cultured on LAMA4 and LAMA5 containing isoforms.
(A) Secretion of inflammatory mediators from HUVECs cultured on cell culture plates pre-coated with LN421 or LN521 or uncoated plastic in response to stimulation with IL-6/sIL-6R (100 ng/ml each). The color codes show fold change (FC) on log2 scale with respective controls. (B) Adhesion of HUVECs to cell culture plates pre-coated with LN421 or LN521 during the first 1 h after seeding. (C) Immunofluorescence staining showing expression pattern of VE-cadherin and PECAM-1 in HUVECs cultured on LN421 and LN521. Immunoblot and quantified expression of (D) VE-cadherin and (E) PECAM-1. A.U. = Arbitrary Unit. **p < 0.01.
3.3 Endothelial LAMA4 to LAMA5 switch induced by IL-6 trans-signaling contributes to inhibition of granulocyte trans-migration
To test whether IL-6 trans-signaling in ECs modulate leukocyte trafficking, we performed in vitro trans-well migration experiments. HUVECs were cultured overnight on trans-well inserts and treated with IL-6+sL-6R (IL-6 trans-signaling) or left untreated (Fig. 4A). After 48 h of incubation, we analyzed the secretion of two key chemokines for granulocyte (CXCL8) and PBMC (CCL-2) migration in both apical and basolateral side of the trans-well inserts containing ECs activated with IL-6 trans-signaling (Fig. 4B). In agreement with the Olink® proteomics analyses, we found that the level of CXCL8 was significantly lower while CCL-2 was significantly higher in both apical and basolateral side of the trans-well inserts containing HUVECs activated with IL-6 trans-signaling. Next, freshly isolated granulocytes or PBMCs were added on the apical side and allowed to migrate for 24 h. As depicted in Fig. 4C, we found that migration of granulocytes through IL-6 trans-signaling activated ECs was significantly reduced compared to untreated cells or TNF-α treated ECs (positive control). Meanwhile, trans-migration of PBMCs (Fig. 4D) or purified CD3+ lymphocytes (Supplementary Fig. 10) was similar between IL-6 trans-signaling activated and untreated HUVECs while their trans-migration across TNF-α treated cells, used as a positive control, was significantly enhanced.
Fig. 4Trans-migration of granulocytes and PBMCs across human vascular ECs.
(A) In vitro trans-migration of granulocytes/PBMCs across HUVECs treated with combination of IL-6 and sIL-6R (100 ng/ml each) or TNF-α (50 ng/ml). (B) ELISA data showing the level of CXCL8 and CCL-2 on the apical and basolateral side of the cell culture inserts after 48 h of incubation before addition of the leukocytes. Bar graph showing the percentage of (C) granulocytes and (D) PBMCs that migrated across HUVECs treated with IL-6+sIL-6R or TNF-α relative to control (Control set to 1.0). (E) In vitro trans-migration setup and bar graph showing granulocytes and PBMCs that trans-migrated into wells containing conditioned media from ECs treated with or without IL-6+sIL-6R for 48 h (Control set to 1.0). (F) In vitro trans-migration setup and a bar graph showing granulocytes and PBMCs that migrated across inserts coated with LN521 and LN421 (LN421 set to 1.0) towards conditioned medium from HUVECs treated with IL-6+sIL-6R. *p < 0.05, **p < 0.01, ***p < 0.001, compared to respective controls.
To determine the role of secreted mediators and laminins separately, we first performed trans-well migration experiments using conditioned medium from HUVECs treated with or without IL-6+sIL-6R. As shown in Fig. 4E, the trans-well migration of granulocytic cells was significantly lower towards wells containing conditioned medium from IL-6 trans-signaling activated cells compared to the one from untreated cells. However, trans-migration of PBMCs was similar across conditioned medium from IL-6 trans-signaling treated cells compared to the one from untreated cells (Fig. 4E). Next, we repeated the migration experiments using trans-well inserts coated with laminins and conditioned medium from HUVECs treated with IL-6+sIL-6R. As shown in Fig. 4F, we found that human granulocytic cells showed limited trans-migration across inserts coated with L521 compared to those coated with LN421. This was also the case when conditioned medium from TNF-α treated ECs was used (Supplementary Fig. 11). Meanwhile, trans-migration of PBMCs across inserts coated with LN421 and LN521 was comparable (Fig. 4F). To test whether migration ability of leucocytes is altered due to activation by IL-6 trans-signaling, we compared migration of leucocytes (3 h and 24 h) across condition medium from unstimulated HUVECs with or without freshly added IL-6+sIL-6R. Here, we found that migration of both granulocytes and PBMCs was comparable across both conditioned media suggesting that presence of IL-6+sIL-6R per se has insignificant effect on migration of leucocytes (Supplementary Fig. 12). Altogether, these findings suggest that the IL-6 trans-signaling induced inhibition of granulocyte trans-migration is mediated by both changes in secretion of chemokines and alterations in expression of laminin alpha chains.
3.4 Expression of laminin alpha chains in human atherosclerotic plaques and their correlations with intra-plaque leukocyte composition
To investigate whether endothelial laminin alpha chains are altered and are associated with granulocytic and lymphocytic cell composition in a chronic inflammatory condition, we analyzed human atherosclerotic plaques in comparison with control vessels collected in the BiKE cohort. The clinical characteristics of participants in the cohort have been described previously [
]. The gene and protein expression of laminin alpha chains was first normalized to the expression of an endothelial cell marker (Von Willebrand factor/vWF) to estimate endothelial specific laminin expressions. As such, we found gene expression of endothelial LAMA4 and LAMA5 were significantly downregulated in plaques compared to control vessels (Fig. 5A). In addition, the gene expression of endothelial LAMA5, but not LAMA4, was significantly lower in plaques from symptomatic patients compared to asymptomatic patients (Fig. 5A). On protein level, we found that the expression of endothelial-specific LAMA5, but not LAMA4, was significantly downregulated in plaques compared to matched atherosclerosis-free adjacent carotid tissue (Fig. 5B). Furthermore, endothelial LAMA5 protein expression was significantly lower in plaques from symptomatic patients compared to those from asymptomatic patients (Fig. 5B). Similar results were obtained when the comparisons were repeated on gene and protein expression data before normalization to the expression of vWF (Supplementary Fig. 13). The LAMA5-to-LAMA4 ratio on gene expression level, but not on protein level, was significantly lower in plaques compared to control vessels (Supplementary Fig. 14A). Meanwhile, LAMA5-to-LAMA4 ratio, both on gene and protein level, was significantly lower in plaques from symptomatic patients compared to plaques from asymptomatic patients (Supplementary Fig. 14B).
Fig. 5Expression of laminin alpha chains normalized to endothelial marker Von Willebrand factor (vWF) in atherosclerotic plaques and macroscopically healthy vessels.
(A) Microarray data showing mRNA expression of LAMA4 and LAMA5 in macroscopically healthy vessels compared to atherosclerotic plaques; and comparison of the gene expression in plaques obtained from symptomatic and asymptomatic patients. (B) Proteomic data showing expression of LAMA4 and LAMA5 in plaques compared to matched adjacent arterial tissue; and comparison of the protein expression in plaques obtained from symptomatic and asymptomatic patients. Data are presented as mean ± SD. Immunohistochemical staining of atherosclerotic plaque and control arteries showing localization of (C) LAMA4 and (D) LAMA5. The red color signifies positive staining. Pearson correlations showing the relationship between the ratio of LAMA5-to-LAMA4 gene expression and gene expression of neutrophil markers (E) CD177 and (F) Myeloperoxidase (MPO) as well as (G) CD3 (lymphocyte marker) in human carotid atherosclerotic plaques (n = 127). TMT = Tandem mass tag.
Immunohistochemical analyses revealed that LAMA5 staining was predominantly seen in subendothelial and shoulder regions of atherosclerotic plaques. The LAMA4 staining, on the other hand, was restricted to the medial layer and in areas of neo-angiogenesis within the atherosclerotic plaque. In control vessel, however, LAMA4 staining was evenly distributed in subendothelial and medial layers while LAMA5 staining was mainly seen in medial layer (Fig. 5C and D). We further investigated whether expression of laminin alpha chains is associated with abundance of granulocytes and lymphocytes in atherosclerotic plaques. For this, we used the gene expression of neutrophil markers CD177 and Myeloperoxidase (MPO) and T-lymphocyte marker (CD3). As shown in Fig. 5E and F, the ratio of LAMA5-to-LAMA4 gene expression in plaques was negatively correlated to the gene expression of CD177 and MPO. Concurrently, the ratio of LAMA5-to-LAMA4 gene expression in plaques was positively correlated to the gene expression of CD3 (Fig. 5G). Altogether, these findings suggest that the expression of LAMA4 and LAMA5 is altered in atherosclerotic plaques compared to control vessels, and that the alteration appears to be associated with stability of the atherosclerotic plaques. Furthermore, our data imply that there is an interplay between the regulation of endothelial laminin expression and immune cell composition of human atherosclerotic plaques.
4. Discussion
In this study, we show that IL-6 trans-signaling in ECs alters endothelial laminin composition by downregulating LAMA4 while upregulating LAMA5 expression. We found that the switch in laminin alpha chain expression induced by IL-6 trans-signaling is accompanied with secretion of several inflammatory mediators. Both these changes contribute to reduced trans-endothelial migration of granulocytic cells. Further, in the context of human atherosclerosis, we show that expression of laminin alpha chains is altered compared to control vessels. In addition, our study reports that the ratio of LAMA5-to-LAMA4 is correlated to markers of granulocytic and lymphocytic cells signifying an interplay between laminins and leukocyte composition of human atherosclerotic plaques.
Our study shows that IL-6 trans-signaling induces a shift in endothelial laminin composition by downregulating LAMA4 and upregulating LAMA5 expression. These findings complement previous observations which suggested inflammatory regulation of endothelial laminin expression [
Endothelial cell laminin isoforms, laminins 8 and 10, play decisive roles in T cell recruitment across the blood-brain barrier in experimental autoimmune encephalomyelitis.
Endothelial cell laminin isoforms, laminins 8 and 10, play decisive roles in T cell recruitment across the blood-brain barrier in experimental autoimmune encephalomyelitis.
]. However, the functional impact of such a shift in laminin chains has not been thoroughly investigated. Recently, a shift in laminin beta chain from LAMB2 to LAMB1 has been demonstrated to reduce adhesion, migration and tube formation ability of ECs while promoting endothelial-to-mesenchymal transition [
]. In this study, we found that ECs are morphologically similar when cultured on LAMA5 or LAMA4 containing isoforms. Meanwhile, ECs cultured on LAMA5 containing isoform have increased adhesion and expression of VE-cadherin compared to ECs cultured on LAMA4 containing isoform. These findings suggest that IL-6 trans-signaling induced LAMA5 upregulation could lead to enhanced tightness of endothelial inter-cellular junctions [
]. In turn, enhanced inter-cellular junction that is mediated by endothelial LAMA5 has previously been linked to inhibition of neutrophil trans-migration via downregulation of CD99L2, an adhesion molecule that is vital in neutrophil diapedesis [
Concomitant to tipping the balance between LAMA4 and LAMA5, we found that IL-6 trans-signaling alters the release of several inflammatory mediators which collectively were predicted to inhibit binding and migration of granulocytes. These alterations included downregulated release of the potent neutrophil chemoattractant CXCL8, and Urokinase plasminogen activator (PLAU), a protease involved in promoting binding and trans-migration of neutrophils [
]. Moreover, we demonstrated that both secretion of inflammatory mediators and LAMA4-to-LAMA5 switch induced by IL-6 trans-signaling contributed to the inhibition of granulocyte trans-migration. Simultaneously, IL-6 trans-signaling upregulated potent lymphocyte chemoattractants including CXCR3 binding chemokines CXCL9, CXCL10 and CXCL11 [
]. This transition requires trans-endothelial migration of certain subsets of leukocytes from the circulation into tissue. IL-6 trans-signaling plays pivotal role in the recruitment, proliferation, and survival of T-lymphocytes [
]. In mice models of acute inflammation, interfering with the IL-6 trans-signaling abolished recruitment of T-lymphocytes mainly through inhibition of lymphocyte-chemoattractants secretion from stromal cells [
]. Complementary to these findings, we provided evidence that IL-6 trans-signaling contributes to limiting trans-migration of granulocytic cells in vitro. In addition, we demonstrated that endothelial cells, like stromal and parenchymal cells, actively respond to IL-6 trans-signaling and take part in regulation of leukocyte trans-migration via release of inflammatory mediators and upregulating LAMA5 (Fig. 6).
Fig. 6Diagram showing proposed mechanisms through which IL-6 trans-signaling in vascular ECs inhibits trans-migration of granulocytic cells such as neutrophils (Created with BioRender.com).
IL-6 trans-signaling alters the release of proteins involved in granulocyte trans-migration including an upregulation in release of CXCL9, CXCL10, and CXCL11; and a downregulation in release of CXCL8, PLAU, and AXL. In addition, IL-6 trans-signaling in ECs causes simultaneous downregulation of LAMA4 and upregulation in LAMA5 expression. Both alterations in secreted chemoattractant proteins and basement membrane laminins contribute to IL-6 trans-signaling induced inhibition of granulocyte trans-migration. Red color = upregulated protein, Green color = downregulated protein.
In atherosclerotic plaques, we found that expression of both LAMA4 and LAMA5 was significantly downregulated compared to macroscopically healthy vessels, and LAMA5 expression was significantly lower in plaques from symptomatic patients than in those from asymptomatic patients. These general reductions in laminin expression in plaques could partly be due to degradation of basement membrane proteins because of increased activity of matrix metalloproteases (MMPs) [
Matrix metalloproteinase 2 is associated with stable and matrix metalloproteinases 8 and 9 with vulnerable carotid atherosclerotic lesions: a study in human endarterectomy specimen pointing to a role for different extracellular matrix metalloproteinase in.
]. Recently, a strong correlation between MMPs and circulating level of laminin chain fragment has been shown in patients undergoing endarterectomy highlighting the relationship between MMPs and breakdown of laminin chains in plaques [
]. However, in this study the downregulation of laminin chains was also evident on mRNA level suggesting specific transcriptional regulations rather than general matrix protein breakdown. In addition, immunohistochemical analyses revealed that the distribution of laminin alpha chains was altered in atherosclerotic plaques compared to control vessel. While LAMA5 was restricted to the medial layer of control vessel, it was abundantly expressed in the subendothelial layer of atherosclerotic plaques suggesting that LAMA5 synthesis is upregulated in luminal ECs during vascular diseases. A similar observation has been made in other chronic inflammatory conditions such as inflammatory bowel disease where LAMA5 was upregulated in subepithelial layer [
]. Yet, a contribution of migrating medial smooth muscle cells to the increased LAMA5 content around the subendothelial area of atherosclerotic plaques cannot be ruled out [
]. Meanwhile, LAMA4 was evenly distributed in control vessel while its expression appeared to be limited to the medial layer and in areas of neo-vessels within the atherosclerotic plaque. However, given the physiological variations in the distribution of laminins across the vasculature, the comparisons of atherosclerotic carotid vessels to that of iliac and radial arteries used as control in this study requires cautious interpretation.
In line with our in vitro data, we found a negative relationship between the ratio of LAMA5-to-LAMA4 expression and markers of granulocytic cells such as CD177 and MPO indicating that lower expression of LAMA4 or higher expression of LAMA5 in atherosclerotic plaques is associated with lower abundance of granulocytic cells in plaques. In addition, LAMA5-to-LAMA4 expression was positively correlated with T-lymphocyte marker CD3. These imply that the shift in laminin alpha chain might contribute to the recruitment, retention, and survival of certain subsets of leukocytes in atherosclerotic plaques [
]. Recently, we also reported, using computational estimation of trans-signaling activity, that higher plasma IL-6:sIL-6R binary complex level is strongly associated with myocardial infarction [
]. Moreover, because IL-6 is abundantly expressed in human atherosclerotic plaques together with the soluble IL-6R, it is likely that IL-6 trans-signaling might contribute to the intra-plaque laminin regulation and immune cell composition [
Inflammatory markers at the site of ruptured plaque in acute myocardial infarction: locally increased interleukin-6 and serum amyloid a but decreased C-reactive protein.
]. Nevertheless, to clearly ascertain the role of laminins in regulating immune cell composition of plaques, understanding the temporal and spatial changes in expression of laminins during the development and progression of atherosclerosis is essential. This is particularly important as sites of leukocyte entry into atherosclerotic plaques appears to vary depending on the stage of plaque development [
Cytokine Expression in Advanced Human Atherosclerotic Plaques: Dominance of Pro-inflammatory (Th1) and Macrophage-Stimulating Cytokines. vol. 145. 1999
Vascular Medicine Intravital Microscopy on Atherosclerosis in Apolipoprotein E-Deficient Mice Establishes Microvessels as Major Entry Pathways for Leukocytes to Advanced Lesions.
]. Moreover, investigating the long-term impact of laminin switch on overall endothelial function and its relevance for the pathogenesis of atherosclerosis is of paramount significance.
In conclusion, our findings show that IL-6 trans-signaling alters endothelial laminin composition by downregulating LAMA4 while upregulating LAMA5 expression. Further, we demonstrate that the switch in laminin alpha chain expression induced by IL-6 trans-signaling is accompanied with secretion of several inflammatory mediators that regulate trans-endothelial migration of leukocytes. We show that both the secreted mediators and the laminin alpha chain switch induced by IL-6 trans-signaling contribute to the reduced trans-endothelial migration of granulocytic cells. In the context of human atherosclerosis, we show that expression of laminin alpha chains is altered compared to control vessels. In addition, our study reports that an increase in the ratio of LAMA5-to-LAMA4 is negatively correlated to markers of granulocytic cells while positively correlating to T-lymphocyte marker signifying an interplay between laminins and the leukocyte composition of human atherosclerotic plaques.
Financial support
This study was funded by Knowledge Foundation (Dnr 2018-0035) and Foundation for Old Servants (Dnr 2019-00851).
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Acknowledgments
We thank BioLamina for providing Biolaminins (LN521/LN511 and LN421/LN411) and Dr Therése Kallur for helpful comments and discussions.
Appendix A. Supplementary data
The following is the Supplementary data to this article.
Endothelial cell laminin isoforms, laminins 8 and 10, play decisive roles in T cell recruitment across the blood-brain barrier in experimental autoimmune encephalomyelitis.
Activation of the JAK/STAT3 and PI3K/AKT pathways are crucial for IL-6 trans-signaling-mediated pro-inflammatory response in human vascular endothelial cells.
Matrix metalloproteinase 2 is associated with stable and matrix metalloproteinases 8 and 9 with vulnerable carotid atherosclerotic lesions: a study in human endarterectomy specimen pointing to a role for different extracellular matrix metalloproteinase in.
Inflammatory markers at the site of ruptured plaque in acute myocardial infarction: locally increased interleukin-6 and serum amyloid a but decreased C-reactive protein.
Cytokine Expression in Advanced Human Atherosclerotic Plaques: Dominance of Pro-inflammatory (Th1) and Macrophage-Stimulating Cytokines. vol. 145. 1999
Vascular Medicine Intravital Microscopy on Atherosclerosis in Apolipoprotein E-Deficient Mice Establishes Microvessels as Major Entry Pathways for Leukocytes to Advanced Lesions.
00112-0&id=ga1.jpg){kind=link}
00112-0&id=gr1.jpg){kind=link}
00112-0&id=gr2.jpg){kind=link}
00112-0&id=gr3.jpg){kind=link}
00112-0&id=gr4.jpg){kind=link}
00112-0&id=gr5.jpg){kind=link}
00112-0&id=gr6.jpg){kind=link}