UBC Faculty Research and Publications

Fibrinogen decreases cardiomyocyte contractility through an ICAM-1-dependent mechanism Boyd, John H; Chau, Edmond H; Tokunanga, Chiho; Bateman, Ryon M; Haljan, Greg; Davani, Ehsan Y; Wang, Yinjin; Walley, Keith R Jan 3, 2008

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


52383-13054_2007_Article_5646.pdf [ 1.28MB ]
JSON: 52383-1.0223247.json
JSON-LD: 52383-1.0223247-ld.json
RDF/XML (Pretty): 52383-1.0223247-rdf.xml
RDF/JSON: 52383-1.0223247-rdf.json
Turtle: 52383-1.0223247-turtle.txt
N-Triples: 52383-1.0223247-rdf-ntriples.txt
Original Record: 52383-1.0223247-source.json
Full Text

Full Text

Available online http://ccforum.com/content/12/1/R2Open AccessVol 12 No 1ResearchFibrinogen decreases cardiomyocyte contractility through an ICAM-1-dependent mechanismJohn H Boyd, Edmond H Chau, Chiho Tokunanga, Ryon M Bateman, Greg Haljan, Ehsan Y Davani, Yinjin Wang and Keith R WalleyUniversity of British Columbia Critical Care Research Laboratories, St. Paul's Hospital, 1081 Burrard Street, Vancouver, BC, V6Z 1Y6, CanadaCorresponding author: John H Boyd, jboyd@mrl.ubc.caReceived: 18 Jul 2007 Revisions requested: 5 Sep 2007 Revisions received: 14 Oct 2007 Accepted: 3 Jan 2008 Published: 3 Jan 2008Critical Care 2008, 12:R2 (doi:10.1186/cc6213)This article is online at: http://ccforum.com/content/12/1/R2© 2008 Boyd et al.; licensee BioMed Central Ltd. This is an open access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.AbstractIntroduction Cardiomyocytes exposed to inflammatoryprocesses express intracellular adhesion molecule-1 (ICAM-1).We investigated whether fibrinogen and fibrinogen degradationproducts, including D-dimer, could alter cardiomyocytecontractile function through interaction with ICAM-1 found oninflamed cardiomyocytes.Methods In vivo, rats were injected with endotoxin to modelsystemic inflammation, whereas isolated rat cardiomyocyteswere treated with tumor necrosis factor-alpha to model theinflammatory environment seen following exposure to bacterialproducts such as lipopolysaccharide.Results In vivo, endotoxin administration profoundly decreasedcardiac contractile function associated with a large increase inintracardiac ICAM-1 and perivascular fibrinogen. Confocalmicroscopy with double-staining of isolated rat cardiomyocytesdemonstrated colocalization of ICAM-1 and fibrinogen. Thisinteraction was disrupted through pre-treatment of the cells withan ICAM-1-blocking antibody. Functionally, isolated ratcardiomyocyte preparations exhibited decreased fractionalshortening when incubated with fibrinogen, and through the useof synthetic peptides, we determined that residues 117–133 ofthe fibrinogen gamma chain are responsible for this interactionwith ICAM-1. Despite having crosslinked gamma chains, D-dimer retained the ability to decrease cardiomyocytecontractility.Conclusion Site 117–133 of the fibrinogen gamma chain isable to depress cardiomyocyte contractility through bindingICAM-1.IntroductionBoth local and systemic inflammation impair cardiac contrac-tility, although the precise mechanism behind this is stillunclear [1-3]. It is now recognized that high levels of inflamma-tory biomarkers such as C-reactive protein and D-dimer areassociated with an increased incidence of, and worse progno-sis for, cardiovascular disease [4-8]. However, whether thesemolecules are simply markers of the inflammatory process ormight actually play a causative role in the resultant organ dys-function is not known. We previously reported a novel two-step regulatory mechanism of cardiomyocyte contractilitywhereby systemic inflammation induces cardiomyocyte-expressed intracellular adhesion molecule-1 (ICAM-1), whosesubsequent ligation results in decreased contractility by sign-aling via the cytoskeleton [9]. While studies using cardiomyo-cyte/leukocyte co-culturing methods demonstrate thatactivated leukocytes can bind to ICAM-1 with a resultantdecrease in myocardial contractility [9-11], we and othershave noted a paucity of intramyocardial leukocytes in wholeanimal models of inflammation [12,13]. Therefore, we postu-lated that in the more complex environment of an in vivo model,ICAM-1 ligands other than the CD11/CD18 receptors foundon activated leukocytes [14-16] play a greater role in ICAM-1-dependent decreases in cardiomyocyte contractility.Fibrinogen, a 340-kDa plasma glycoprotein with a physiologi-cal plasma concentration of 1.5 to 4.5 g/L, as well as itsrelated protein fragments D-dimer and other fibrinogen degra-BSA = bovine serum albumin; EDV = end diastolic volume; Ees = end-systolic elastance; EF = ejection fraction; ELISA = enzyme-linked immunosorb-Page 1 of 10(page number not for citation purposes)ent assay; Emax = maximal end-systolic elastance; FDP = fibrinogen degradation product; FITC = fluorescein isothiocyanate; ICAM-1 = intracellular adhesion molecule-1; Ig = immunoglobulin; LAD = left anterior descending; LPS = lipopolysaccharide; PBS = phosphate-buffered saline; TNF-α = tumor necrosis factor-alpha; vWF = von Willebrand factor; XL = crosslinking.Critical Care    Vol 12 No 1    Boyd et al.dation products (FDPs) represent potential ICAM-1-bindingmyocardial depressant substances. The amino acid sequence117–133 of fibrinogen gamma chain (fg-γ-117–133) binds tothe amino acid sequence 8–22 (ICAM-1-8–22) within the firstimmunoglobulin (Ig) domain of ICAM-1 [17]. The functionalrole of the fibrinogen-ICAM-1 interaction includes adhesion ofleukocytes to endothelial cells [18,19], leukocyte transmigra-tion [20], and promotion of endothelial cell survival [21]. Inter-estingly, fibrinogen-ICAM-1 ligation leads to cytoskeleton-dependent ERK1/2 (extracellular signal-regulated kinase-1/2)phosphorylation in endothelial cells [22]. In view of our previ-ous observation in cardiomyocytes that ICAM-1 ligation by leu-kocytes reduces contractility via focal adhesion kinasephosphorylation at the cytoskeleton [9], fibrinogen-ICAM-1ligation ultimately could lead to alteration in cardiomyocytecontractile function. As fragments of polymerized fibrinogensuch as D-dimer are markedly elevated in most inflammatorystates [4-8], it is of particular interest to determine whetherthese molecules are able to influence cardiac physiologythrough interaction with ICAM-1.The goal of our study, therefore, was to determine whetherexposure to fibrinogen and FDPs altered the contractile func-tion of cardiomyocytes. To simulate systemic inflammation,rats were injected with endotoxin, and through immunohisto-chemistry, we confirmed an increase in both cardiomyocyte-expressed ICAM-1 as well as increased intramyocardial fibrin-ogen deposition. In isolated cardiomyocytes exposed to aninflammatory environment, we established the specificity of thefibrinogen-ICAM-1 interaction and went on to determine theactive site on fibrinogen responsible for ICAM-1-mediatedalterations in contractility.Materials and methodsThis study was approved by the University of British ColumbiaAnimal Care Committee and adheres to the Canadian andNational Institutes of Health guidelines for animalexperimentation.In vivo experimental modelsFor the endotoxin model of inflammation, male Sprague-Daw-ley rats 350 to 450 g in weight were injected intraperitoneallywith lipopolysaccharide (LPS) (10 mg/kg) or vehicle control(normal saline). The LPS dosage was selected as a midrangedosage of endotoxemic models that result in a hemodynamiceffect [23,24]. The heart was excised 6 hours after injection,embedded in Optimal Cutting Temperature compound (Elec-tron Microscopy Sciences, Hatfield, PA, USA), frozen in dry-ice-chilled isopentane, and stored at -80°C.Measurement of left ventricular contractility and cardiac functionLeft ventricular contractility and other measures of ventricularfunction were determined from pressure-volume measure-Millar Instruments Inc., Houston, TX, USA). Six to ten pressure-volume loops during a vena cava occlusion were sampled andused to measure end-systolic elastance (Ees), which is theslope of the end-systolic pressure-volume relationship rela-tively insensitive to changes in preload and afterload [25]. Emaxis defined as the maximal Ees. The volume axis intercept, Vd,was considered zero volume for the steady-state measure-ments. Pressure-volume loops measured during steady-stateconditions were used to measure the maximum rate of changeof intraventricular pressure during isovolumic systole dividedby end diastolic volume (EDV), (dP/dtmax)/EDV, which is a sen-sitive isovolumic phase measure of left ventricular contractility[26], as well as to calculate ejection fraction (EF). End-systolicpressure during steady state was used as a measure of sys-temic arterial pressure afterload.Immunofluorescent imaging with quantificationFrozen heart sections (6 μm) were acetone-fixed and incu-bated with universal blocking agent (DakoCytomation, Glos-trup, Denmark). Fibrinogen and von Willebrand factor (vWF),a marker for the endothelium, were stained together to assesswhether infiltration of fibrinogen into the myocardiumoccurred. Sections were incubated with 1:20 fluorescein iso-thiocyanate (FITC)-conjugated goat anti-mouse fibrinogen(Nordic Immunological Laboratories, Tilburg, The Netherlands)and 1:200 rabbit anti-mouse vWF (DakoCytomation) primaryantibody and then labeled with Alexa Fluor 594 goat anti-rab-bit antibody (Invitrogen Corporation, Carlsbad, CA, USA).Nuclei were stained with the Hoechst stain (Invitrogen Corpo-ration). Control sections were incubated with FITC-conju-gated non-specific goat IgG (Santa Cruz Biotechnology, Inc.,Santa Cruz, CA, USA) and non-specific rabbit IgG (DakoCy-tomation) and processed in identical conditions.Immunofluorescent ICAM-1 staining was carried out by incu-bating sections with 1:500 mouse anti-rat ICAM-1 monoclonalantibody 1A29 (BD Biosciences, San Jose, CA, USA) andthen with Alexa Fluor 594-labeled goat anti-mouse antibody(Invitrogen Corporation). Control sections were incubatedwith non-specific mouse IgG (Invitrogen Corporation). Afterdrying, the slides were mounted with DABCO (1,4-diazabicy-clo[2.2.2]octane) to prevent photobleaching.Images were captured using a laser scanning confocal micro-scope with a 63× water immersion lens (Leica SP2; Leica,Wetzler, Germany). Samples were imaged using fluorescencewith wavelength excitations and emissions of 488 nm and 495to 580 nm (respectively) for fibrinogen and 594 nm and 600to 700 nm (respectively) for vWF and ICAM-1. The scan for-mat was 512 × 512 pixels. Image capturing was performedsequentially using a three-frame average. All imaging was per-formed under identical microscope settings (for example, laserintensity and photomultiplier tube gain).Page 2 of 10(page number not for citation purposes)ments using Pressure-Volume Analysis software (PVAN 2.9;Available online http://ccforum.com/content/12/1/R2Cross-sections of 15 randomly selected blood vessels, identi-fied via vWF staining, were imaged. Two ellipses were tracedaround each vessel: a small ellipse positioned closely alongthe vessel boundary and a large ellipse with proportional majorand minor axes but three times the area of the small ellipse.Fibrinogen staining present in the annulus between the twoellipses was identified as perivascular fibrinogen. The sum flu-orescence intensity per annulus area was measured using theLeica software.To measure myocardial ICAM-1 expression, heart sectionsfrom the endotoxemic and control groups were imaged asdescribed above and fluorescent intensity measures weretaken using traced field areas containing myocardial tissue.The sum fluorescence intensity per unit area was measuredusing the Leica software.Isolation of rat ventricular myocytesMale Sprague-Dawley rats were injected with heparin andanesthetized using isofluorane. The heart was excised,mounted on a modified Langendorff apparatus, and digestedwith 281 U/mL collagenase (Worthington Biochemical Corpo-ration, Lakewood, NJ, USA). After digestion, the cells wereresuspended in modified Eagle's medium containing increas-ing Ca2+ concentrations (200 μM, 500 μM, and 1 mM). Fivehundred thousand cells in M199 with bovine serum albumin(BSA) were loaded into a laminin-coated Petri dish 6 cm indiameter (BD Biosciences) and the cardiomyocytes wereincubated for 12 hours to allow them to become relatively qui-escent. After 24 hours, cells were considered viable if theydemonstrated a characteristic rod shape without cytoplasmicblebbing.Measurement of cardiomyocyte fractional shorteningCells were paced at 1 Hz using a Grass S48 stimulator(Grass-Telefactor, Warwick, RI, USA) with a voltage set at120% of the threshold capture voltage. Images were capturedusing a Myocam video camera (IonOptix Corporation, Milton,MA, USA) and analyzed using an IonOptix Softedge detectionpackage (IonOptix Corporation). Fractional shortening wascalculated as the difference between diastolic and systoliclengths, divided by diastolic length.Coating of fibrinogen to polystyrene beadsPolystyrene beads 8 μm in diameter (Bangs Laboratories, Inc.,Fishers, IN, USA) were washed twice with acetate buffer (pH5.4). Beads were mixed with rat fibrinogen (Enzyme ResearchLaboratories, South Bend, IN, USA) at a concentration of300,000 beads per microgram of fibrinogen in a 500-μLEppendorf tube. A micromagnetic stir bar was placed in thetube, and the mixture was gently stirred for 2 hours at roomtemperature. The beads were then washed three times withfresh acetate buffer. Clumps of fibrinogen-coated beads werebroken apart by passing them through a syringe with a 27.5-ICAM-1 peptide-fibrinogen binding assayNinety-six-well Corning Costar 9018 enzyme-linked immuno-sorbent assay (ELISA) plates (eBioscience, Inc., San Diego,CA, USA) were coated for 2 hours at room temperature withfibrinogen concentrations ranging from 0.01 to 100 nM inbicarbonate/carbonate coating buffer (3.03 g of Na2CO3, 6.0g of NaHCO3 per 1,000 mL of distilled water) (pH 9.6). Wellswere washed with phosphate-buffered saline (PBS) and thenblocked overnight with 1% BSA. One hundred micromolarbiotinylated ICAM-1 (8–22) sequence EAFLPRGGS-VQVNCS or biotinylated scrambled peptide sequence SCN-VQVSGGRPLFAE (University of British Columbia PeptideFacility, Vancouver, BC, Canada) was then added to the wellsand incubated for 2 hours at room temperature before wash-ing three times with PBS. HRP-linked anti-biotin antibody (1μg/mL) (Invitrogen Corporation) was then added and incu-bated for 2 hours, followed by three washes. One hundredmicroliters of ABTS (2,2'-azino-di(3-ethylbenzthiazoline sul-fonate) solution (Chemicon International, Temecula, CA, USA)was added to each well and incubated for 60 minutes.Absorbance values were measured at 405 nm and at 492 nmfor reference.IncubationsTwenty-four hours after cardiomyocyte isolation, cells wereactivated with tumor necrosis factor-alpha (TNF-α) (20 ng/mL)for 4 hours to upregulate ICAM-1 expression [9]. In studiesusing fibrinogen-coated beads, 25,000 beads were added tocells in laminin-coated 96-well plates (500 cells per well). Abead that moved with the contracting cardiomyocyte andmaintained a contact relative location on the membrane duringcontraction was considered to be adherent. In studies usingrat fibrinogen (Enzyme Research Laboratories), human fibrino-gen, D-dimer fragments D and E (HYPHEN BioMed, Neuville-sur-Oise, France), and fibrinogen gamma chain (AnaSpec,Inc., San Jose, CA, USA) cells were incubated at concentra-tions of 0.03 μM, 0.1 μM, 0.3 μM, and 1 μM, respectively, for4 hours at 37°C. After 4 hours at 37°C, cardiomyocyte frac-tional shortening was measured as described above.In studies using the ICAM-1 (8–22) sequence EAFLPRGGS-VQVNCS or scrambled peptide sequence SCNVQVSG-GRPLFAE (University of British Columbia Peptide Facility), 1μM rat fibrinogen and 100 μM ICAM-1 (8–22) or scrambledpeptide were mixed and incubated for 4 hours at 37°C prior toincubation with cardiomyocytes as described above. In stud-ies using anti-ICAM-1-blocking monoclonal antibody 1A29(BD Biosciences), the antibody was added to cardiomyocytesat a concentration of 200 ng/mL 4 hours prior to the additionof fibrinogen.Colocalization of ICAM-1 with fibrinogenUpon TNF-α activation, cardiomyocytes were co-cultured withOregon Green-labeled fibrinogen (Invitrogen Corporation) forPage 3 of 10(page number not for citation purposes)guage needle. 4 hours. They were then fixed with 3% paraformaldehyde forCritical Care    Vol 12 No 1    Boyd et al.20 minutes and blocked with universal blocking agent. Immun-ofluorescent ICAM-1 staining was carried out by incubatingthe cardiomyocytes with 1:500 mouse anti-rat ICAM-1 anti-body (BD Biosciences) followed by Alexa Fluor 594-labeledgoat anti-mouse antibody. The cardiomyocytes were examinedusing a confocal microscope (Leica) as described in the pre-vious section. The scan format was 1,024 × 1,024 pixels, and2× zoom was applied.Statistical analysisAll data are expressed as mean ± standard error. For eachexperimental condition and time point, at least four independ-ent replicate analyses were performed, unless otherwisenoted. Differences between groups were tested using a one-way analysis of variance and the post hoc Bonferroni test toidentify specific differences between groups. Differenceswere considered significant for P values of less than 0.05.ResultsSystemic inflammation depresses cardiac contractility and is associated with intracardiac extravasation of fibrinogen and increased expression of ICAM-1 by cardiomyocytesWe determined whether endotoxin would create an environ-ment within the heart which would allow ICAM-1 expressed oncardiomyocytes to interact with extravasated fibrinogen andwhether this would be associated with alterations in cardiaccontractility. Six hours after LPS injection, the endotoxemicgroup of rats exhibited decreased left ventricular contractilitycompared with the saline-treated controls (Figure 1). Cardiaccycle pressure-volume loops using mean data clearly demon-strate an LPS-induced rightward shift along the volume axis.This shift reflects left ventricular dilation represented byincreased EDV, maintenance of stroke volume (SV), andresultant marked reduction of left ventricular EF (EF = SV/EDV) (Figure 1 and Table 1). The preload-independent Emax isdramatically decreased in LPS-treated versus saline-treatedrats, whereas (dP/dT)/EDV reflects isovolemic contractilityand is also depressed with LPS (Table 1). Immunostaining ofheart tissue from these rats demonstrated a dramatic increasein both intramyocardial ICAM-1 expression and perivascularfibrinogen in the myocardium of LPS-treated rats (Figure 2a,b).Image quantification demonstrated a 5.5 ± 1.6-fold increase inICAM-1 expression and a 2.1 ± 0.6-fold increase in perivascu-lar fibrinogen in the myocardium of LPS-treated rats (Figure2c).ICAM-1 expressed on activated isolated cardiomyocytes specifically binds to fibrinogenBy means of confocal microscopy, ICAM-1 was found to bepresent on the cell surface of TNF-α-activated cardiomyo-cytes. Co-immunostaining with fluorescently labeled fibrino-gen demonstrated a high degree of colocalization (Figure 3a),supporting previous reports of interaction between these twomolecules [17,21]. To confirm that fibrinogen specificallybound ICAM-1, we pre-treated isolated cardiomyocytes witheither with an ICAM-1-blocking antibody or non-specific IgG.Compared with the IgG control group, cardiomyocytes treatedwith blocking antibody to ICAM-1 exhibited significantly loweradherence to fibrinogen-coated beads (Figure 3b). ICAM-1 isknown to interact with fibrinogen via ICAM-1 peptides 8–22.To confirm this specific interaction between ICAM-1 andfibrinogen, we performed an ELISA binding assay using immo-bilized fibrinogen incubated with biotinylated ICAM-1 (8–22)peptide. In dose-finding experiments, as expected with recep-tor-ligand binding, there is a dose-response curve that reachessaturation at a fibrinogen concentration of 100 μM (Figure 4a).When group mean absorbance data are taken at this plateaufibrinogen concentration of 100 μM, there is a large increasein ICAM-1 (8–22) peptide binding with fibrinogen comparedwith scrambled peptide (Figure 4b). Thus, fibrinogen specifi-cally binds to ICAM-1 through interaction with ICAM-1 pep-tides 8–22.Fibrinogen mediates decreased cardiomyocyte contractility via an ICAM-1-dependent mechanismWe next examined whether ICAM-1 ligation by fibrinogenalters cardiomyocyte contractility. In one series of experiments,isolated cardiomyocytes were pre-treated with either ICAM-1-blocking antibody or isotype control antibody before addingfibrinogen-coated polystyrene beads. Whereas cardiomyo-cytes pre-treated with isotype control antibody demonstrateda dose-dependent decrease in contractility upon exposure tofibrinogen-coated beads, cardiomyocytes pre-treated withICAM-1-blocking antibody demonstrated no reduction in con-tractility (Figure 5). To verify that this interaction is mediatedTable 1Hemodynamic data from endotoxemic and control animalsTreatment Heart rate (beats per minute) Systolic pressure (mm Hg) Ejection fraction (percentage) (dP/dT)/EDV EmaxLPS 392 ± 152 108 ± 8 38 ± 2 44 ± 3 2.5 ± 0.4Controla 313 ± 30 112 ± 6 52 ± 8 121 ± 11 7.7 ± 0.6P = NS P = NS P < 0.05 P < 0.05 P < 0.05aSaline-treated rats. dP, derivative of pressure; dT, derivative of time; EDV, end diastolic volume; Emax, maximal end-systolic elastance; LPS, Page 4 of 10(page number not for citation purposes)lipopolysaccharide-treated rats; NS, not significant.Available online http://ccforum.com/content/12/1/R2specifically via the ICAM-1 (8–22) fibrinogen binding site, weperformed a competitive assay in which a peptide containingthe ICAM-1 (8–22) fibrinogen binding site was pre-incubatedwith soluble fibrinogen before addition of this mixture toactivated isolated cardiomyocytes. Pre-incubation of the fibrin-ogen with excess ICAM-1 (8–22) peptide abolished the fibrin-ogen-mediated decrease in cardiomyocyte contractility,whereas pre-incubation of fibrinogen with a 'scrambled' pep-tide containing the same residues showed no such effect (Fig-ure 6).Fibrinogen chain D mediates decreased cardiomyocyte contractilityFibrinogen is composed of three major subunits, a central Echain linked to two D chains (Figure 7a). The smaller gammachain is always found associated with the D chain and thus isnot generally considered to be a distinct subunit. To determinewhether the fibrinogen-mediated contractile dysfunctionresults from interaction of ICAM-1 with the intact whole mole-cule or whether a single chain contains the binding site, cardi-omyocytes were incubated with whole fibrinogen, fibrinogenchain D, or fibrinogen chain E. There was a significantdecrease in cardiomyocyte contractility following incubationwith whole fibrinogen and fibrinogen chain D, but no effectwas seen with fibrinogen chain E (Figure 7b).Figure 1LPS decreases cardiac contractilityes cardiac contractility. Cardiac cycle pressure-volume loops obtained 6 hours after intraperitoneal injection of lipopolysaccha-ride (LPS) or saline into rats. Acquired with group mean data, the curves demonstrate an LPS-induced increased end diastolic volume (EDV), maintenance of stroke volume (SV), and therefore a marked reduction of left ventricular ejection fraction (SV/EDV). Ees, end-systolic elastance.Figure 2LPS increases intracardiac ICAM-1 and perivascular fibrinogenes intracardiac ICAM-1 and perivascular f brinogen. (a) Frozen cardiac sections from lipopolysaccharide (LPS)-treated and saline-treated rats demonstrate that intracardiac intracellular adhesion molecule-1 (ICAM-1) (red) is dramatically increased in the former. (b) Fibrinogen (green) was greatly increased outside the endothelium (von Willebrand factor labeled red) in the LPS group compared with rats treated with saline. (c) Group mean data of the fold increases in myo-cardial ICAM-1 expression and perivascular fibrinogen deposition in LPS-treated versus saline-treated animals. *p < 0.05 versus saline.Page 5 of 10(page number not for citation purposes)Critical Care    Vol 12 No 1    Boyd et al.Amino acid sequence 117–133 of the fibrinogen gamma chain is the active site, and the crosslinked gamma chains of D-dimer retain the ability to interact with ICAM-1We next determined whether the ICAM-1 binding gammachain site 117–133 [17] of the D chain (Figure 8a) wasresponsible for the observed contractile dysfunction. Further-more, as dimerization of fibrinogen chain D (commonly knownas D-dimer) is accomplished in part via crosslinking the XLsites of the gamma chain (Figure 8a,b) [27], we testedwhether dimerization resulted in attenuation of the ICAM-1-mediated effect. The gamma peptide 117–133 resulted in asignificant reduction in fractional shortening compared withcontrol, whereas scrambled peptide had no effect (Figure 8c).Incubation with D-dimer also resulted in a significant reductionin fractional shortening (Figure 8c), demonstrating that thefunctional site 117–133 remains active despite crosslinking ofgamma chains.DiscussionIn this study, we propose a novel mechanism linking two phe-nomena that occur as a result of inflammation: dysregulation ofthe coagulation cascade and myocardial dysfunction. The keyfibrinogen gamma chain decreases cardiomyocyte contractil-ity through binding ICAM-1. Of great interest to clinicians isthat D-dimer, a product of fibrinogen polymerization and sub-sequent digestion which includes 117–133 of the gammachain, in addition to its important role in diagnosis of throm-boembolism, can decrease cardiomyocyte contractility.It has long been recognized that seemingly disparate causesof local or systemic inflammation, such as ischemia reper-fusion, inflammatory cardiomyopathy, orthotopic heart trans-plant rejection, or sepsis [1-3], all culminate in myocardialdysfunction. While each disorder undoubtedly poses uniquechallenges to the maintenance of myocardial homeostasis,there could be a factor that is common to all. Endothelial dam-age with subsequent capillary leakage represents a final com-mon pathway of inflammatory disorders [28]. Increasedpermeability of the endothelium leads to a shift of circulatingelements from the plasma into the organs. Should this fluid fluxcontain circulating substances capable of depressing myocar-dial contractility, this may be the link between myocardial dys-function and inflammatory states. This depressant not onlymust reach the cardiac myocytes but must have a receptorFigure 3Fibrinogen binds specifically to cardiomyocyte ICAM-1i ogen binds specifically to cardiomyocyte ICAM-1. Colocalization of fibrinogen and cardiomyocyte intracellular adhesion molecule-1 (ICAM-1). (a) Isolated cardiomyocytes were incubated with Oregon Green-labeled fibrinogen (green) and fluorescently stained for ICAM-1 (red). A multi-photon dual-excitation image of the contour of the cell of interest is shown. By means of an overlay of images, strong colocalization of fibrinogen and ICAM-1 was indicated by a yellow color. (b) A representative image of a rat cardiomyocyte with adherent fibrinogen-coated polystyrene beads (white arrows) is shown to the left of the graph. The specificity of the ICAM-1-fibrinogen interaction is demonstrated as anti-ICAM-1 antibody pre-treatment results in significantly less fibrinogen-coated polystyrene beads adherent to the cardiomyocytes (*p < 0.05 versus control). Ab, antibody; CL, con-trol; IgG, immunoglobulin G.Page 6 of 10(page number not for citation purposes)finding reported is that amino acid sequence 117–133 of the capable of mediating changes in contractility. We have previ-Available online http://ccforum.com/content/12/1/R2ously shown that ICAM-1 expressed on cardiomyocytes isinduced by inflammatory mediators and, upon activation, iscapable of decreasing cardiomyocyte contractility [9,13].Any circulating ICAM-1 ligands could be candidates for caus-ing myocardial dysfunction provided that they permeate theheart. CD11a/CD18 (LFA-1) and Cd11b/CD18 (Mac-1)expressed on the surface of polymorphonuclear leukocytesare ICAM-1 ligands capable of reducing myocyte contractilityin vitro [11] and have been proposed to be the link betweeninflammation and cardiac dysfunction. However, we and oth-ers have noted a striking paucity of intramyocardial leukocytesin whole animal models of inflammation [12,13]. Fibrinogen, aswell as its related protein fragments D-dimer and other FDPs,represents potential ICAM-1 binding myocardial depressantbasic science, fibrinogen and FDPs satisfy two major criteriafor causality. Clinically, not only are fibrinogen and D-dimermarkedly increased in inflammatory disorders, but their levelsare inversely correlated with favorable outcome [4-8]. As forbiologic plausibility, the amino acid sequence 117–133 offibrinogen gamma chain (fg-γ-117–133) is capable of bindingthe amino acid sequence 8–22 (ICAM-1-8–22) within the firstIg domain of ICAM-1 [17]. While the fibrinogen-ICAM-1 inter-action facilitates adhesion of leukocytes to endothelial cells[18,19], leukocyte transmigration [20], and promotion ofendothelial cell survival [21], there is no information regardingits role in cardiac physiology.In this study, we show for the first time that fibrinogen is capa-ble of mediating a reduction in cardiomyocyte contractilitythorough activation of ICAM-1. It is important to note that sig-nificant reduction in fractional shortening was achieved at afibrinogen concentration of 0.2 mg/mL, approximately oneorder of magnitude less than its physiological concentration inplasma [27]. Fibrinogen is a large 340-kDa plasma glycopro-tein and, as such, would be expected to have limited tissuepenetration compared with smaller plasma proteins. Despiteits large size, however, we showed a significant increase inperivascular fibrinogen deposition in an in vivo model ofsystemic inflammation. FDPs, notably fragment D (100 kDa) orD-dimer at roughly double that size [27], could potentially infil-trate deeper into the myocardium and exert their depressanteffect in areas that fibrinogen could not access. Importantly,not only did we find that systemic injury increased intracardiacFigure 4ICAM-1 (8–22) binds to fibrinogen to fibrinogen. Two-step enzyme-linked immuno-sorbent assay in which fibrinogen was immobilized on the 96-well plate and then incubated with biotinylated intracellular adhesion molecule-1 (ICAM-1) (8–22) peptide or biotinylated 'scrambled' control peptide. Anti-biotin antibody was added, and colorimetric absorbance quantified the peptide-fibrinogen interaction. (a) Representative ICAM-1 (8–22) dose-response curve showing absorbance plateau at a fibrinogen con-centration of approximately 100 μM. (b) Group mean absorbance data taken at the plateau fibrinogen concentration (100 μM) demonstrating a strong interaction between the ICAM-1 (8–22) peptide and fibrino-gen compared with a small non-specific interaction between the scram-bled peptide and fibrinogen. *p < 0.05 versus scrambled peptide.Figure 5Fibrinogen decreases cardiomyocyte fractional shortening via ICAM-1i ogen decreases cardiomyocyte fractional shortening via ICAM-1. Cardiomyocytes were pre-treated with either a blocking anti-ICAM-1 antibody or isotype control antibody prior to the addition of fibrinogen-coated beads. Fractional shortening was then measured. Whereas pre-treatment with immunoglobulin G (IgG) isotype antibody was no different than fibrinogen alone, treatment with blocking anti-ICAM-1 antibody prevented the fibrinogen-induced decrease in frac-tional shortening. *p < 0.05 versus fibrinogen beads alone. Ab, anti-body; FBG, fibrinogen; ICAM-1, intracellular adhesion molecule-1.Page 7 of 10(page number not for citation purposes)substances. Through both human epidemiologic data and fibrinogen, there was a dramatic increase in ICAM-1 expres-Critical Care    Vol 12 No 1    Boyd et al.sion. Thus, our animal models of disease provide in vivo evi-dence to support the hypothesis that fibrinogen and FDPsmight be the circulating myocardial depressant factors.Determining whether the previously identified interactionbetween fibrinogen and ICAM-1 [17-21] is responsible foralterations in cardiac physiology is fundamental informationrequired both to understand the mechanism and to designpotential therapeutics. Fibrinogen consists of two majorchains, E and D, as illustrated in Figure 7a. The gamma chainis a subunit of chain D, as shown in Figures 7a and 8a, andcontains a crosslinking (XL) site through which two D chainsdimerize. The putative active site of fibrinogen is 117–133 onthe gamma chain [17] and, though remote from the XL site asshown in Figure 8b, might be altered or allosterically interferedwith upon dimerization. We show that fibrinogen chain Dcauses cardiomyocyte contractile dysfunction whereas chainE had no biologic effect. This agrees with findings by otherinvestigators that it is the D chain responsible for ICAM-1-mediated vasoconstriction [29]. Furthermore, we went on toshow that previously identified [17] site 117–133 of the fibrin-ogen gamma chain was responsible for the ICAM-1-mediatedphysiologic effects. As the gamma chain is linked to the Dchain, it is important to consider it in the context of polymerizedD chains. Despite its crosslinked gamma chains, D-dimer alsospecificity of interaction of D-dimer with fibrinogen was nottested through antibody blockade or competing peptide, webelieve that, as it is composed of two D subunits, the mecha-nism of action is nearly certainly analogous to the individual Dchain. This last finding is particularly exciting given that, whilelong touted as a biomarker for both the presence and progno-sis of inflammatory disease [4-8], D-dimer has been perceived,to date, as a disease marker with no intrinsic biologic effect.Not only do we now know that vascular tone can be altered bythe fibrinogen D chain's activation of ICAM-1 [29], but here wedemonstrate that the key contractile function of cardiac mus-cle cells is impaired via the same mechanism.ConclusionSite 117–133 of the fibrinogen gamma chain is able todepress cardiomyocyte contractility through binding ICAM-1.The implication of the reported mechanism extends beyondthe realm of our models of inflammation to other pathologiescharacterized by inflammation and heart failure, such as post-ischemia reperfusion injury, inflammatory cardiomyopathy, andFigure 6ICAM-1 (8–22) mediates decreased contractilityiates decreased contractility. Before fibrinogen was added to activated cardiomyocytes, soluble intracellular adhesion molecule-1 (ICAM-1) (8–22) peptide, which binds fibrinogen at pep-tides 117–133, is added in excess to the fibrinogen. Activated cardio-myocytes incubated with fibrinogen alone demonstrate a 40% reduction in contractility. Pre-incubation of fibrinogen with the ICAM-1 (8–22) peptide results in competition between cardiomyocyte-expressed ICAM-1 and the ICAM-1 (8–22) peptide for the fibrinogen active site (117–133). Pre-incubation with the ICAM-1 (8–22) peptide abolishes the reduced contractility seen with fibrinogen alone, whereas pre-incubation with 'scrambled' ICAM-1 peptide had no effect. *p < 0.05 versus control. Fbg, fibrinogen.Figure 7Fibrinogen subunit D decreases contractilityi ogen subunit D decreases contrac ility. (a) Schematic diagram of the fibrinogen molecule, showing two D chains each containing a gamma chain linked to a central E chain. (b) Cardiomyocytes were incubated with whole fibrinogen as well as the major subunits D and E of fibrinogen. Whole fibrinogen and subunit D resulted in significant decreases in fractional shortening (FS), whereas subunit E had no sig-nificant effect. *,†p < 0.05 versus control.Page 8 of 10(page number not for citation purposes)caused significant reductions in contractility. Although the orthotopic heart transplant rejection.Available online http://ccforum.com/content/12/1/R2Competing interestsThe authors declare that they have no competing interests.Authors' contributionsJHB drafted the manuscript. EHC performed the initial con-tractility measurements. EYD helped design the contractilityexperiments. CT performed the peptide experiments. RMBperformed immunohistochemistry. GH designed the peptides.YW performed the animal work. KRW conceived of the study,participated in its design and coordination, and helped to draftthe manuscript. All authors read and approved the finalmanuscript.AcknowledgementsThis work was supported by the Canadian Institutes of Health Research. KRW is a Michael Smith Foundation for Health Research (MSFHR) Dis-Figure 8Fibrinogen gamma chain and D-dimer decrease contractilityi ogen gamma chain and D-dimer d crease contractility. (a) Schematic diagram of D-dimer with two D subunits linked in part via interaction of the gamma chains. For simplicity, we have not shown the areas on the D chain itself which participate in dimerization. (b) Expanded diagram of the gamma subunit showing the crosslinking (XL) site of the C-terminal regions which results in amine donor lysine 406 of one gamma chain and a glutamine acceptor at residue 398 or 399. The intracellular adhesion molecule-1 (ICAM-1) binding site is shown as residue 117–133, far removed from the XL site. (c) Cardiomyocytes were incubated with a peptide with the sequence 117–133, D-dimer, or scrambled peptide. D-dimer and the gamma (117–133) peptide resulted in significant decreases in fractional shortening (FS), whereas scrambled peptide had no significant effect. *,†p < 0.05 versus control.Key messages• Intracardiac intracellular adhesion molecule-1 (ICAM-1) and fibrinogen are increased as a result of systemic inflammation.• Amino acid sequence 117–133 of the fibrinogen gamma chain is responsible for binding ICAM-1, func-tionally decreasing cardiomyocyte contractility.• D-dimer contains the fibrinogen gamma chain and also decreases cardiomyocyte contractility.Page 9 of 10(page number not for citation purposes)tinguished Scholar. JHB is an IMPACT (Integrated and mentored pulmo-nary and cardiovascular training) postdoctoral fellow. RMB is an MSFHR postdoctoral fellow.Critical Care    Vol 12 No 1    Boyd et al.References1. Van Eyk JE, Powers F, Law W, Larue C, Hodges RS, Solaro RJ:Breakdown and release of myofilament proteins duringischemia and ischemia/reperfusion in rat hearts: identificationof degradation products and effects on the pCa-force relation.Circ Res 1998, 82:261-271.2. Reilly JM, Cunnion RE, Burch-Whitman C, Parker MM, ShelhamerJH, Parrillo JE: A circulating myocardial depressant substanceis associated with cardiac dysfunction and peripheral hypop-erfusion (lactic acidemia) in patients with septic shock.  Chest1989, 95:1072-1080.3. Yano M, Ono K, Ohkusa T, Suetsugu M, Kohno M, Hisaoka T,Kobayashi S, Hisamatsu Y, Yamamoto T, Noguchi N, et al.: Alteredstoichiometry of FKBP12.6 versus ryanodine receptor as acause of abnormal Ca(2+) leak through ryanodine receptor inheart failure.  Circulation 2000, 102:2131-2136.4. Lowe GD, Rumley A, McMahon AD, Ford I, O'Reilly DS, PackardCJ: Interleukin-6, fibrin D-dimer, and coagulation factors VIIand XIIa in prediction of coronary heart disease.  ArteriosclerThromb Vasc Biol 2004, 24:1529-1534.5. Lowe GD: Circulating inflammatory markers and risks of cardi-ovascular and non-cardiovascular disease.  J Thromb Haemost2005, 3:1618-1627.6. McDermott MM, Ferrucci L, Liu K, Criqui MH, Greenland P, GreenD, Guralnik JM, Ridker PM, Taylor LM, Rifai N, et al.: D-dimer andinflammatory markers as predictors of functional decline inmen and women with and without peripheral arterial disease.J Am Geriatr Soc 2005, 53:1688-1696.7. Heuer JG, Sharma GR, Gerlitz B, Zhang T, Bailey DL, Ding C, BergDT, Perkins D, Stephens EJ, Holmes KC, et al.: Evaluation of pro-tein C and other biomarkers as predictors of mortality in a ratcecal ligation and puncture model of sepsis.  Crit Care Med2004, 32:1570-1578.8. Sagastagoitia JD, Sáez Y, Vacas M, Narváez I, Sáez de LafuenteJP, Molinero E, Magro A, Lafita M, Santos M, Escobar A, et al.:Association between inflammation, lipid and hemostatic fac-tors in patients with stable angina.  Thromb Res 2007,120:53-59.9. Davani EY, Dorscheid DR, Lee CH, van Breemen C, Walley KR:Novel regulatory mechanism of cardiomyocyte contractilityinvolving ICAM-1 and the cytoskeleton.  Am J Physiol Heart CircPhysiol 2004, 287:H1013-1022.10. Poon BY, Ward CA, Cooper CB, Giles WR, Burns AR, Kubes P:Alpha(4)-integrin mediates neutrophil-induced free radicalinjury to cardiac myocytes.  J Cell Biol 2001, 152:857-866.11. Entman ML, Youker K, Shoji T, Kukielka G, Shappell SB, Taylor AA,Smith CW: Neutrophil induced oxidative injury of cardiac myo-cytes. A compartmented system requiring CD11b/CD18-ICAM-1 adherence.  J Clin Invest 1992, 90:1335-1345.12. Raeburn CD, Calkins CM, Zimmerman MA, Song Y, Ao L, BanerjeeA, Harken AH, Meng X: ICAM-1 and VCAM-1 mediate endotox-emic myocardial dysfunction independent of neutrophil accu-mulation.  Am J Physiol Regul Integr Comp Physiol 2002,283:R477-486.13. Davani EY, Boyd JH, Dorscheid DR, Wang Y, Meredith A, Chau E,Singhera GK, Walley KR: Cardiac ICAM-1 mediates leukocyte-dependent decreased ventricular contractility in endotoxemicmice.  Cardiovasc Res 2006, 72:134-142.14. Landis RC, McDowall A, Holness CL, Littler AJ, Simmons DL,Hogg N: Involvement of the "I" domain of LFA-1 in selectivebinding to ligands ICAM-1 and ICAM-3.  J Cell Biol 1994,126:529-537.15. Staunton DE, Marlin SD, Stratowa C, Dustin ML, Springer TA: Pri-mary structure of ICAM-1 demonstrates interaction betweenmembers of the immunoglobulin and integrin supergenefamilies.  Cell 1988, 52:925-933.16. Clayton A, Evans RA, Pettit E, Hallett M, Williams JD, Steadman R:Cellular activation through the ligation of intercellular adhe-sion molecule-1.  J Cell Sci 1998, 111(Pt 4):443-453.17. Duperray A, Languino LR, Plescia J, McDowall A, Hogg N, CraigAG, Berendt AR, Altieri DC: Molecular identification of a novelfibrinogen binding site on the first domain of ICAM-1 regulat-ing leukocyte-endothelium bridging.  J Biol Chem 1997,272:435-441.cular endothelium through an ICAM-1-dependent pathway.Cell 1993, 73:1423-1434.19. Languino LR, Duperray A, Joganic KJ, Fornaro M, Thornton GB,Altieri DC: Regulation of leukocyte-endothelium interactionand leukocyte transendothelial migration by intercellularadhesion molecule 1-fibrinogen recognition.  Proc Natl AcadSci USA 1995, 92:1505-1509.20. Sans E, Delachanal E, Duperray A: Analysis of the roles of ICAM-1 in neutrophil transmigration using a reconstituted mamma-lian cell expression model: implication of ICAM-1 cytoplasmicdomain and Rho-dependent signaling pathway.  J Immunol2001, 166:544-551.21. Gardiner EE, D'Souza SE: Sequences within fibrinogen andintercellular adhesion molecule-1 (ICAM-1) modulate signalsrequired for mitogenesis.  J Biol Chem 1999,274:11930-11936.22. Pluskota E, D'Souza SE: Fibrinogen interactions with ICAM-1(CD54) regulate endothelial cell survival.  Eur J Biochem 2000,267:4693-4704.23. Parker JL: Contractile function of heart muscle isolated fromendotoxin-shocked guinea pigs and rats.  Adv Shock Res1983, 9:133-145.24. Heneka MT, Loschmann PA, Osswald H: Polymerized hemo-globin restores cardiovascular and kidney function in endo-toxin-induced shock in the rat.  J Clin Invest 1997, 99:47-54.25. Nozawa T, Yasumura Y, Futaki S, Tanaka N, Uenishi M, Suga H:Efficiency of energy transfer from pressure-volume area toexternal mechanical work increases with contractile state anddecreases with afterload in the left ventricle of the anesthe-tized closed-chest dog.  Circulation 1988, 77:1116-1124.26. Kass DA, Maughan WL, Guo ZM, Kono A, Sunagawa K, SagawaK: Comparative influence of load versus inotropic states onindexes of ventricular contractility: experimental and theoreti-cal analysis based on pressure-volume relationships.  Circula-tion 1987, 76:1422-1436.27. Walker JB, Nesheim ME: The molecular weights, mass distribu-tion, chain composition, and structure of soluble fibrin degra-dation products released from a fibrin clot perfused withplasmin.  J Biol Chem 1999, 274:5201-5212.28. Lindbom L: Regulation of vascular permeability by neutrophilsin acute inflammation.  Chem Immunol Allergy 2003,83:146-166.29. Lominadze D, Tsakadze N, Sen U, Falcone JC, D'Souza SE: Fibrin-ogen and fragment D-induced vascular constriction.  Am JPhysiol Heart Circ Physiol 2005, 288:H1257-1264.Page 10 of 10(page number not for citation purposes)18. Languino LR, Plescia J, Duperray A, Brian AA, Plow EF, GeltoskyJE, Altieri DC: Fibrinogen mediates leukocyte adhesion to vas-


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items