sepsis induces myocardial dysfunction in both experimental animals and humans (28, 33, 41, 55). However, the mechanism by which it does so remains unclear. Over 30 years ago, Lefer (28) reported that the serum of septic patients contained a myocardial depressant factor. Subsequently, Parrillo et al. (41) reported that tumor necrosis factor-α (TNF-α) and interleukin-1 (IL-1) are the dominant cytokines responsible for the depressant effect of septic human serum on cardiomyocytes. Although in vitro studies demonstrate an immediate depression in contractile function by TNF-α and IL-1 (10, 32, 38), in vivo exposure to the same agents results in a delayed cardiodepression that takes several hours to manifest (37, 39). We (34) and others (18) reported that the lipopolysaccharide (LPS)-induced increase in circulating and myocardial TNF-α does not coincide with the development of endotoxemic cardiac dysfunction (34). Despite this temporal discordance, neutralization of TNF-α attenuates LPS-induced cardiac dysfunction, suggesting that TNF-α plays a critical but indirect role in mediating contractile dysfunction during endotoxemia.

The role of TNF-α and IL-1 in depressing cardiac contractility may lie in their induction of other downstream factors. The early elaboration of these proinflammatory cytokines during sepsis stimulates the release of chemokines and the expression of cellular adhesion molecules (13, 14), ultimately resulting in neutrophil [polymorphonuclear neutrophil (PMN)] infiltration. We (49) and others (15, 58) reported that LPS-induced lung and liver injury is provoked by this PMN infiltration. However, the role of PMNs in LPS-induced myocardial dysfunction remains to be defined.

PMN infiltration requires the orderly expression of cellular adhesion molecules (2). Intercellular adhesion molecule-1 (ICAM-1) and vascular cell adhesion molecule-1 (VCAM-1) are two members of the immunoglobulin-like supergene family of adhesion molecules known to mediate tight adherence of PMNs to endothelial cells (19,52). LPS and TNF-α both increase the cell surface expression of ICAM-1 (12, 13) and VCAM-1 (13, 14,21) on both coronary endothelial cells and cardiac myocytes. The importance of ICAM-1 in mediating myocardial PMN infiltration and tissue necrosis after ischemia/reperfusion has been confirmed (29, 30, 51).

VCAM-1 was not originally recognized as an important mediator of PMN infiltration as PMNs were thought to lack very late antigen-4 (VLA-4), the integrin ligand for VCAM-1. The subsequent identification of VLA-4 on PMNs in several different species (19, 23) led to further investigation of the role of myocardial VCAM-1 in acute inflammation (13, 21). We recently reported that increased VCAM-1 expression following LPS is associated with myocardial dysfunction (44).

Although both ICAM-1 and VCAM-1 have been implicated in PMN-mediated lung and liver injury during sepsis (5, 6, 59), their individual roles in LPS-induced myocardial PMN infiltration and dysfunction remain obscure. Moreover, ICAM-1 and VCAM-1 activation has been shown to directly mediate cell signaling (27, 31). Thus it remains unclear whether their influence on myocardial function during endotoxemia is neutrophil dependent.

The purposes of this study were 1) to delineate the time course of myocardial ICAM-1/VCAM-1 expression and neutrophil accumulation after LPS, 2) to determine the role of ICAM-1 and VCAM-1 in myocardial neutrophil accumulation and myocardial dysfunction after LPS, and 3) to explore the role of neutrophils in LPS-induced myocardial dysfunction.

METHODS

Animals.

Male wild-type and ICAM-1-deficient C57BL/6 mice, 20–30 g body wt, were purchased from Jackson Laboratory (Bar Harbor, ME). Animals were acclimated for 1 wk after delivery on a 12:12-h light-dark cycle and maintained on a standard pellet diet before usage. All experiments were approved by the Animal Care and Research Committee, University of Colorado Health Sciences Center. All animals received humane care in compliance with the Guide for the Care and Use of Laboratory Animals {DHEW Publication no. [National Institutes of Health (NIH)] 85–23, revised 1985, Office of Science and Health Reports, DRR/NIH, Bethesda, MD 20205}.

Chemicals and reagents.

Hamster anti-mouse ICAM-1 monoclonal antibody (mAb; clone 3E2B) and rat anti-mouse VCAM-1 mAb (clone MVCAM.A429) were purchased from Endogen (Woburn, MA). Rat anti-mouse ICAM-1 mAb (clone KAT-1) and rat anti-mouse neutrophil p40 antigen mAb (clone 7/4) were purchased from Serotec (Oxford, UK). Rat IgG2a, Cy3-conjugated donkey anti-rat IgG, and fluorescein-conjugated goat anti-rabbit IgG were purchased from Jackson ImmunoResearch Laboratories (West Grove, PA). Hamster IgG and adsorbed rabbit anti-neutrophil Ab were purchased from Accurate Chemicals (Westbury, NY). Rabbit polyclonal Ab against human von Willebrand factor (cross-reacts with mouse antigen), Escherichia coli LPS (serotype 055:B5), and all other chemicals were purchased from Sigma Chemical (St. Louis, MO).

Experimental protocols.

To determine the effect of LPS on cardiac function, mice were injected with either LPS (0.5 mg/kg ip, n = 10) or vehicle control (normal saline intraperitoneal, n = 10). We previously monitored the hemodynamic response in rats to this dose of LPS and observed profound reduction in cardiac contractility with minimal influence on mean arterial pressure, thus eliminating shock as a potential confounding cause of myocardial dysfunction. At 6 h after treatment, animals were intraperitoneally anesthetized with ketamine/xylazine (75 mg/kg, Phoenix Pharmaceutical, St. Joseph, MO; and 10 mg/kg, Fort Dodge Animal Health, Fort Dodge, IA, respectively) and simultaneously heparinized (2,000 U/kg, Elkins-Sinn, Cherry Hill, NJ). Hearts were then isolated, and cardiac contractility was assessed by the Langendorff technique. We previously examined the time course of endotoxin-induced myocardial dysfunction in mice and observed maximal dysfunction at 6 h (44). Thus we examined cardiac dysfunction in this study at 6 h post-LPS.

The effects of LPS on myocardial ICAM-1/VCAM-1 expression and neutrophil accumulation were examined by treating mice with LPS (0.5 mg/kg ip, n = 13) or vehicle control (normal saline intraperitoneal, n = 5) and euthanizing the animals at 2, 4, and 6 h. Hearts were excised and divided in half transversely after the atria were removed. The apical portion was embedded in tissue-freezing media and frozen in dry ice-chilled isopentane and stored at −70°C for immunofluorescent staining.

To examine the role of ICAM-1 and VCAM-1 in myocardial neutrophil accumulation and cardiac function during endotoxemia, mice were administered neutralizing mAb to either ICAM-1 (5 mg/kg iv,n = 6) or VCAM-1 (5 mg/kg iv, n = 5) 5 min after LPS injection. Similarly, mice genetically deficient in ICAM-1 (n = 5) were injected with LPS. Hearts were isolated 6 h after LPS, and contractility was compared with that of mice treated with control Ab (n = 3 each, nonspecific rat IgG2a and hamster IgG, 5 mg/kg iv) and LPS. No difference in myocardial neutrophil accumulation or cardiac function was observed in wild-type mice that received the two different control Abs. Thus these two groups were combined during subsequent comparisons.

To examine the role of neutrophils in endotoxin-induced cardiac dysfunction, mice were depleted of neutrophils using two different methods. Vinblastine (5 mg/kg iv, n = 5) was administered at 24 and 72 h before LPS. Neutropenia was also produced in a separate group of mice by treating them with adsorbed rabbit anti-neutrophil Ab (n = 5) (1). In these experiments, mice were treated intraperitoneally with anti-neutrophil Ab diluted 1:15 in 0.2 ml of PBS on days 3and 4 before LPS, followed by 2 days of treatment with 0.2 ml of undiluted Ab. At the time of heart isolation for determination of contractility, blood was obtained by aortotomy, and complete blood count and cell differential were determined.

Isolated heart perfusion.

Cardiac function was determined by an isovolumic, nonrecirculating Langendorff technique, as described previously (35). Isolated hearts were perfused with normothermic Krebs-Henseleit solution containing (in mmol/l) 11.0 glucose, 1.2 CaCl₂, 4.7 KCl, 25 NaHCO₃, 119 NaCl, 1.17 MgSO₄, and 1.18 KH₂PO₄. Perfusion pressure was maintained at 70 mmHg. The perfusion buffer was saturated with a gas mixture of 92.5% O₂-7.5% CO₂ to achieve a Po₂ of 450 mmHg, a Pco₂of 40 mmHg, and pH 7.4. A latex balloon was inserted in the left ventricle via the left atrium and inflated with water to achieve a left ventricular end-diastolic pressure (LVEDP) of 10 mmHg. Pacing wires were attached to the right atrium, and hearts were paced at 300 beats/min. Coronary flow was assessed by quantifying the effluent from the pulmonary arteries. Myocardial temperature was maintained at 37°C. Left ventricular developed pressure (LVDP), its first derivatives (+dP/dt, − dP/dt), and LVEDP were continuously recorded by a computerized pressure amplifier-digitizer (Maclab 8, AD Instrument, Cupertino, CA). After a 20-min equilibration period, LVDP and ±dP/dt were determined at varied LVEDP levels (10, 15, and 20 mmHg). The degree of reduction in LVDP and ±dP/dt in LPS-treated hearts was not influenced by varying LVEDP from 10 to 20 mmHg. Therefore, all data represented herein were obtained at an LVEDP of 10 mmHg.

Immunofluorescent staining.

Indirect immunofluorescent detection and localization of ICAM-1, VCAM-1, and neutrophils were performed as described previously (53). Transverse sections (5-μm thick) of ventricular myocardium were cut with a cryotome (International Equipment, Needham Heights, MA) and then dried at room temperature for 2 h. Sections were treated with a mixture of 30% methanol and 70% acetone at room temperature for 10 min and washed with PBS. The sections were then fixed in PBS-buffered 3% paraformaldehyde at room temperature for 10 min and washed with PBS. All subsequent incubations were performed at room temperature. To block nonspecific binding sites, sections were incubated for 30 min with 10% donkey serum in PBS. Sections were then incubated with primary Ab (diluted 1:200 for ICAM-1 and VCAM-1 and 1:500 for neutrophils in PBS containing 1% BSA) for 60 min. After three washes with PBS, sections were incubated for 45 min with Cy3-labeled donkey anti-rat IgG (1:250 dilution with PBS containing 1% BSA). After thorough washes with PBS, sections were counterstained with fluorescein-labeled wheat germ agglutinin (5 μg/ml for cell surface staining) and bis-benzimide (2.5 μg/ml for nuclear staining). The sections were mounted with aqueous anti-quenching media. To ascertain the specificity of the primary Ab, adjacent sections were incubated with nonspecific rat IgG (diluted 1:200 in PBS containing 1% BSA) in replacement of the primary Ab and then processed in identical conditions. Microscopic observation and photography were performed with a Leica DMRXA microscope (Solms, Germany). Slidebook 2.6 software (I. I. I., Denver, CO) was used for image quantitation.

Colocalization.

Colocalization of ICAM-1 or VCAM-1 with von Willebrand factor was performed to distinguish endothelial vs. cardiomyocyte expression. Sections were incubated with rabbit anti-human von Willebrand factor Ab (cross-reacts with mouse antigen, diluted 1:500 in PBS) along with ICAM-1 or VCAM-1 mAb and processed as described underImmunofluorescent staining with the addition of fluorescein-labeled goat anti-rabbit IgG.

Image quantitation.

ICAM-1 and VCAM-1 images were quantitated with Slidebook software. Eight random images were taken from each myocardial section. Imaging was performed at ×40 magnification (1,000 × 1,000 pixels/image). All images were taken while blinded to both the specimen and Cy3 channel. Images were masked to exclude 95% of nonspecific fluorescence as determined from images of myocardium incubated with nonspecific rat IgG. The mean area (μm²), mean intensity, and integrated intensity were determined for each image.

Myocardial neutrophil accumulation was determined by counting all nucleated cells with Cy3 fluorescence. The number of neutrophils per section was divided by the calculated area of the myocardial section and reported as number of neutrophils per millimeters squared. We previously reported that this method of assessing tissue neutrophil accumulation closely correlates with myeloperoxidase activity (54).

Statistical analysis.

All data are expressed as means ± SE. ANOVA was performed, and differences between groups were considered significant whenP < 0.05, as verified by Bonferroni/Dunn post hoc test.

RESULTS

Cardiac function after LPS exposure.

Isolated heart perfusion was used to determine the effect of LPS on cardiac function. At 6 h following LPS treatment, LVDP was reduced by nearly 40% compared with saline controls (35.7 ± 1.5 vs. 56.9 ± 2.7 mmHg, P < 0.05; Fig.1A). Similarly, +dP/dt and −dP/dt were reduced by LPS treatment (Fig. 1B). To determine whether LPS-induced myocardial dysfunction is due to a change in left ventricular compliance, LVDP was recorded at LVEDPs ranging from 10 to 20 mmHg. The degree of reduction in LVDP in LPS-treated hearts compared with saline controls was not influenced by increasing LVEDP (data not shown). Coronary flow was not different between the LPS and control groups (3.3 ± 0.2 vs. 2.9 ± 0.2 ml/min, respectively; Fig. 1C).

Myocardial ICAM-1 and VCAM-1 expression after LPS exposure.

To determine the influence of LPS on myocardial ICAM-1 and VCAM-1 expression, mice were treated with LPS, and myocardial sections were examined by immunofluorescence. Hearts incubated with nonimmune rat IgG showed no immunoreactivity (not shown), whereas incubation with anti-mouse ICAM-1 and VCAM-1 mAb demonstrated both adhesion molecules in control hearts (Fig. 2, A,B, E, and F, respectively). LPS induced a time-dependent increase in both ICAM-1 (Fig. 2, Cand D) and VCAM-1 (Fig. 2, G and H) expression. Quantitative data are shown in Fig.3, A and B. Both the mean area and mean intensity of VCAM-1 increased. However, only the mean area of ICAM-1 increased. The maximal increase in expression of both ICAM-1 and VCAM-1 occurred 6 h after LPS.

Localization of myocardial ICAM-1 and VCAM-1.

To localize ICAM-1 and VCAM-1 within the myocardium, colocalization was performed with von Willebrand factor to identify vascular endothelial cells. The images shown in Fig. 4 are of myocardial sections from LPS-treated mice depicting similarly sized vessels with surrounding cardiomyocytes. VCAM-1 exclusively colocalized with von Willebrand factor, whereas ICAM-1 colocalized with von Willebrand factor on endothelium but was also present on other cell types including cardiomyocytes.

Time course of LPS-induced myocardial neutrophil accumulation.

Myocardial neutrophil accumulation, as determined by immunofluorescence, also demonstrated a time-dependent increase after LPS with a maximal sixfold increase over saline controls observed at 6 h (13.0 ± 1.7 vs. 1.9 ± 1.3 PMN/mm²,P < 0.05; Fig. 5). The majority of neutrophils in LPS-treated hearts were localized in the interstitial space.

Role of ICAM-1 and VCAM-1 in mediating myocardial neutrophil accumulation.

Having demonstrated the association between myocardial ICAM-1/VCAM-1 expression and neutrophil accumulation following LPS, we sought to determine whether blockade of ICAM-1 or VCAM-1 would attenuate LPS-induced myocardial neutrophil accumulation. Administration of VCAM-1-neutralizing mAb to LPS-treated mice reduced myocardial neutrophil accumulation by 75% compared with LPS-treated mice that received control Ab (3.6 ± 0.9 vs. 16.0 ± 2.9 PMN/mm², P < 0.05; Fig.6). In contrast, neutralization of ICAM-1 using mAb or genetic absence of ICAM-1 did not influence LPS-induced neutrophil accumulation (14.8 ± 2.8 PMN/mm² in ICAM Ab and 14.3 ± 2.7 PMN/mm² in ICAM knockout; Fig. 6).

Role of ICAM-1 and VCAM-1 in LPS-induced myocardial dysfunction.

After demonstrating the differential roles of ICAM-1 and VCAM-1 in mediating LPS-induced myocardial neutrophil accumulation, we sought to determine the role of each adhesion molecule in mediating LPS-induced myocardial dysfunction. Nonspecific control Ab did not alter cardiac function in saline controls and had no effect on LPS-induced cardiac dysfunction. Genetic absence of ICAM-1 or treatment with either ICAM-1 or VCAM-1 mAb nearly abrogated LPS-induced depression of both LVDP (Fig. 7) and the positive and negative first derivatives of LVDP (dP/dt, data not shown). Coronary flow was not significantly different among groups.

Role of neutrophils in LPS-induced myocardial dysfunction.

Because neutralization of ICAM-1 abrogated LPS-induced myocardial dysfunction without reducing myocardial neutrophil accumulation, we sought to determine whether neutrophils were obligatory for LPS-induced myocardial dysfunction. Mice were effectively depleted of neutrophils (>95% reduction) using either vinblastine or Ab (absolute blood neutrophil count: 0.04 ± 0.02 × 10⁹/l in vinblastine-treated mice, 0.03 ± 0.03 × 10⁹/l in Ab-treated mice vs. 1.6 ± 0.1 × 10⁹/l in LPS-treated mice). Immunofluorescent staining revealed complete absence of neutrophils in myocardial sections of neutrophil-depleted mice following LPS. Neutrophil depletion had no effect on normal cardiac function in saline controls and did not attenuate LPS-induced myocardial dysfunction at 6 h (Fig.8).

DISCUSSION

This study examined the effects of LPS on myocardial ICAM-1/VCAM-1 expression and neutrophil accumulation and sought to determine the role of ICAM-1, VCAM-1, and neutrophils in LPS-induced myocardial dysfunction. LPS induced a time-dependent increase in myocardial ICAM-1/VCAM-1 expression and neutrophil accumulation. Neutralization or genetic absence of ICAM-1 as well as neutralization of VCAM-1 all attenuated LPS-induced myocardial dysfunction. Neutralization of VCAM-1 reduced myocardial neutrophil accumulation. In contrast, neutralization or genetic absence of ICAM-1 protected myocardial function without reducing neutrophil accumulation. Lastly, prior neutrophil depletion by two different methods did not influence LPS-induced myocardial dysfunction. Thus LPS-induced myocardial dysfunction occurs independent of neutrophil accumulation and requires both ICAM-1 and VCAM-1.

Intraperitoneal injection of a nonlethal dose of E. coli LPS was used to induce myocardial dysfunction. Myocardial contractile function was depressed by nearly 40% at 6 h (Fig. 1A), which is similar to previous reports (9, 18, 35, 57). Both +dP/dt and −dP/dt were decreased to a similar degree, suggesting that both systolic and diastolic functions were impaired by LPS. Altering LVEDP in a range of 10–20 mmHg did not influence the degree of myocardial dysfunction in LPS-treated mice. This suggests that LPS-induced myocardial dysfunction is not simply due to a decrease in ventricular compliance. LPS-treated hearts exhibited similar coronary flow compared with saline controls, suggesting that LPS-induced dysfunction is not a result of significant changes in myocardial perfusion.

Our findings confirm in vitro reports of the effects of LPS on cardiomyocyte ICAM-1 expression (14). However, this study extends these findings by demonstrating a time-dependent increase in myocardial ICAM-1 expression in an in vivo model. In addition, colocalization with von Willebrand factor established that ICAM-1 expression was increased on both cardiomyocytes and vascular endothelium, whereas VCAM-1 was expressed exclusively on endothelium. This difference in cellular expression may explain why both the mean intensity and area of VCAM-1 expression increased during endotoxemia, whereas only the area of ICAM-1 expression increased. Cardiomyocytes represent the majority of the area of a myocardial section. Thus the increase in the area but not mean intensity of ICAM-1 expression suggests that the majority of the increase in ICAM-1 expression during endotoxemia occurs as a result of an increase in the number of cardiomyocytes expressing ICAM-1 rather than an increase in the expression of ICAM-1 on individual cardiomyocytes. Because VCAM-1 is expressed on endothelial cells, the observation that both the mean intensity and mean area of VCAM-1 increased following LPS suggests that the increased expression of VCAM-1 during endotoxemia is due to both an increase in the number of endothelial cells expressing VCAM-1 and an increase in the level of expression of VCAM-1 on individual endothelial cells.

Neutralization of ICAM-1 has been shown to reduce myocardial neutrophil accumulation following ischemia and reperfusion (12,40). Surprisingly, neither genetic absence nor Ab blockade of ICAM-1 attenuated LPS-induced myocardial neutrophil accumulation. Despite the failure to reduce myocardial neutrophil accumulation, both genetic absence and Ab blockade of ICAM-1 abrogated LPS-induced myocardial dysfunction. Que and colleagues (43) reported that neutralization or absence of ICAM-1 failed to reduce pulmonary neutrophil accumulation following cecal ligation and puncture in mice. Our study constitutes an initial report that myocardial neutrophil accumulation following LPS is independent of ICAM-1.

The time course of myocardial VCAM-1 expression following LPS paralleled that of ICAM-1. In contrast to ICAM-1 expression, VCAM-1 closely colocalized with von Willebrand factor, suggesting that its expression increased predominantly on endothelial cells. This finding may explain the lower myocardial VCAM-1 protein concentrations compared with ICAM-1 reported in other studies (14). To determine the role of VCAM-1 in neutrophil accumulation, VCAM-1 was blocked with a neutralizing Ab. VCAM-1 blockade reduced myocardial neutrophil accumulation by 75% following LPS. Blockade of VCAM-1 also abrogated LPS-induced myocardial dysfunction. Similarly, VCAM-1 has been found to be involved in hepatic neutrophil accumulation and necrosis following LPS (6). Our findings suggest that VCAM-1 is the predominant adhesion molecule responsible for LPS-induced myocardial neutrophil accumulation, although both ICAM-1 and VCAM-1 are obligatory for myocardial dysfunction.

Neutrophils have been implicated in myocardial necrosis and dysfunction following ischemia and reperfusion (45) and in LPS-induced lung and liver injury (1, 15). However, the role of neutrophils in LPS-induced myocardial dysfunction remains obscure. Our results demonstrate a time-dependent increase in myocardial neutrophil accumulation. The majority of neutrophils in the myocardium were localized to the interstitial space by immunofluorescent staining. Despite a sixfold increase in myocardial neutrophil accumulation at 6 h following LPS, the absolute neutrophil number in myocardium (13.0 ± 1.7 PMN/mm²) is quite small compared with that in lungs (1,049 ± 72 PMN/mm²) following the same dose of LPS (53). Moreover, blockade or absence of ICAM-1 abrogated LPS-induced myocardial dysfunction without decreasing myocardial neutrophil accumulation. These observations prompted us to examine whether neutrophils were obligatory for LPS-induced myocardial dysfunction. Prior neutrophil depletion by two different methods failed to prevent LPS-induced myocardial dysfunction. These findings suggest that LPS induces myocardial dysfunction independent of neutrophil accumulation. It is possible that other leukocytes play a role in LPS-induced myocardial dysfunction (50). In the liver, depletion/inactivation of Kupffer cells attenuates LPS-induced cytokine responses (47) and hepatocyte injury (20). Further investigation of the role of other leukocytes in LPS-induced myocardial dysfunction is warranted.

Our findings suggest that LPS-induced myocardial dysfunction requires both ICAM-1 and VCAM-1. The temporal increase in myocardial ICAM-1 expression following LPS closely parallels that of VCAM-1. Cross-linking of ICAM-1 has been shown to increase the expression of both ICAM-1 (4) and VCAM-1 (26). Thus it is possible that ICAM-1 and VCAM-1 expression is synergistic. It is interesting that VCAM-1 is expressed primarily by vascular endothelial cells yet blockade of VCAM-1 protected myocardial function. Furthermore, we observed that intravenously administered ICAM-1 Ab exclusively colocalized with von Willebrand factor (data not shown), which suggests that the ICAM-1 and VCAM-1 Ab block these adhesion molecules only on endothelial cells. Despite the endothelial site-specific action of the ICAM-1 and VCAM-1 Ab, both attenuated LPS-induced myocardial dysfunction. In fact, ICAM-1 knockout did not provide significant additional protection, which suggests that myocyte ICAM-1 plays a minor role in endotoxemic myocardial dysfunction. These observations suggest that LPS-induced myocardial function may involve cross talk between the coronary endothelium and cardiomyocyte. Such cross talk has been implicated by Shah et al. (48) who demonstrated that cultured vascular endothelial cells release soluble factors with negative inotropic effects on isolated cardiomyocytes.

Recent investigations of the role of ICAM-1 and VCAM-1 in signal transduction have broadened the potential function(s) of these molecules far beyond their original description as simply adhesion molecules (17). The novel finding that LPS-induced myocardial dysfunction requires both ICAM-1 and VCAM-1 but occurs independent of neutrophils further emphasizes the importance of examining the role of ICAM-1 and VCAM-1 in cellular signaling. In the absence of neutrophils, ligation of ICAM-1 and/or VCAM-1 could occur via soluble fibrinogen (11, 25), other leukocytes (50), or potentially by soluble integrins (60).

The mechanism by which ICAM-1 and VCAM-1 mediate LPS-induced myocardial dysfunction is unclear. However, several in vitro studies in noncardiac cells have demonstrated that many potentially cardiodepressant factors are influenced by activation of ICAM-1. For instance, activation of ICAM-1 may induce the production of cytokines such as TNF-α and IL-1β, which have been implicated in LPS-induced myocardial dysfunction (3, 10, 32, 34). Activation of ICAM-1 by Ab cross-linking has been shown to increase the production of TNF-α and IL-1β in astrocytes (7, 27) and synovial cells (22). This ICAM-1-mediated induction of TNF-α and IL-1β appears to be mediated by activation of mitogen-activated protein kinase (MAPK). Activation of ICAM-1 by Ab cross-linking or soluble fibrinogen provokes activation of MAPK and extracellular signal-regulated kinase (ERK) in astrocytes (7) and lymphoid cells (11, 16). Inhibition of ERK activation attenuates ICAM-1-mediated TNF-α (7) and IL-1β (27) production in astrocytes. Although TNF-α and IL-1β contribute to endotoxemic myocardial dysfunction (3, 32,34), the roles of adhesion molecules and stress kinases (MAPK, ERK) in LPS-induced myocardial production of these cytokines remain to be determined.

Another mechanism by which ICAM-1 may mediate LPS-induced myocardial dysfunction is via alteration in intracellular calcium concentrations. Decreases in cardiomyocyte calcium sensitivity have been implicated in the depression of cardiac contractility by LPS (36, 56). Antibody cross-linking of ICAM-1 has been reported to alter intracellular calcium concentrations in endothelial cells (4) and astrocytes (8); thus alteration of calcium flux via activation of ICAM-1 may contribute to LPS-induced contractile dysfunction. Antibody cross-linking of ICAM-1 has also been shown to stimulate an oxidative burst in monocytes (46). Less has been published on VCAM-1-mediated signal transduction. However, Matheny and colleagues (31) reported that ligation of VCAM-1 on endothelial cells also induced production of reactive oxygen species as well as actin restructuring. Activation of ICAM-1 and/or VCAM-1, therefore, may generate oxygen-free radicals that have been shown to contribute to LPS-induced myocardial dysfunction (42). It is likely that ligation of both ICAM-1 and VCAM-1 leads to the production of myocardial depressive factors. However, the role of ICAM-1/VCAM-1 signaling in myocardial cytokine production, calcium handling, and the generation of reactive oxygen species awaits investigation.

FOOTNOTES

• Address for reprint requests and other correspondence: C. D. Raeburn, Dept. of Surgery, Univ. of Colorado Health Sciences Center, 4200 E. Ninth Ave., Campus Box C-320, Denver, CO 80262 (E-mail: christopher.[email protected]edu).

• The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

• 10.1152/ajpregu.00034.2002