- Email: [email protected]
Serum proteins facilitate neutrophil induction of endothelial leukocyte adhesion molecule 1
Serum proteins facilitate neutrophil induction of endothelial leukocyte adhesion molecule 1 James M. Van den Bogaerde, BS, Karen L. Hynes, BS, Elizabeth T. Clark, MD, and Bruce L. Gewertz, MD, Chicago, Ill.
Background. Although the individual actions of neutrophils and serum proteins such as complement in acute inflammation are well characterized, less is known about their effects in combination. We investigated the combined effects of neutrophil contact and active serum proteins on the expression of endothelial leukocyte adhesion molecule 1 (ELAM-1). Methods. Confluent monolayers of human umbilical vein endothelial cells were incubated with neutrophils in the presence and absence of fresh human serum. Flow cytometry was used to assess expression of endothelial intercellular adhesion molecule 1 (ICAM-1) and ELAM-1. In addition, neutrophils were retained in a semipermeable insert, which allowed their secretions to contact the endothelium but restricted neutrophil-endothelial contact. Results. ELAM-1 expression was significantly increased on the cells coincubated with neutrophils and fresh human serum (25.8%; p < 0.01). There was no significant change in ELAM-1 expression on endothelial cells incubated with fresh human serum alone (3.9%; p > 0.01) or in those incubated with neutrophils and heat-inactivated serum (9.3%; p > 0.01). In the absence of neutrophil contact, ELAM1 expression was increased only in the presence of fresh human serum (9.6%; p < 0.05). Conclusions. These findings suggest that serum proteins may potentiate the volume or potency of neutrophil-derived diffusable mediators of ELAM-1 expression. These effects are eliminated with the heat inactivation of serum proteins, implicating a heat sensitive mediator such as the complement cascade. (Surgery 1998;123:199-204.) From the Department of Surgery, University of Chicago, Chicago, Ill.
IN ISCHEMIA/REPERFUSION INJURY the tissue damage caused by inflammatory factors is often more extensive than the damage caused by the original ischemic insult.1-3 In vivo experiments suggest that such tissue damage is caused largely by infiltrating neutrophils because reperfusion of a wide range of ischemic organs with neutropenic blood greatly reduces the extent of tissue necrosis.4-8 Other studies have shown that using monoclonal antibodies against neutrophil adhesion molecules to block the molecular interactions between the endothelium and neutrophils also increases the survival of tissue after an ischemic insult.9-11 A great deal of recent information has further implicated the complement cascade in ischemia-reperfusion injury. With recombinant soluble complement receptor 1 (sCR1) to inhibit the activation of alterAccepted for publication May 20, 1997. Reprint requests: Bruce L. Gewertz, MD, Department of Surgery, University of Chicago, 5841 S. Maryland Ave., MC 5028, Chicago, IL 60637. Copyright © 1998 by Mosby, Inc. 0039-6060/98/$5.00 + 0 11/56/85438
native and classic complement cascades, investigators were able to reduce tissue damage in the ischemia-reperfused myocardium to an extent comparable to the studies using neutropenic blood.12,13 These findings raise the possibility that neutrophil-mediated damage and serum proteinmediated damage in acute inflammation may be related. We investigated the differential contributions of neutrophils and serum proteins to endothelial cell adhesion molecule expression in vitro. Endothelial cell adhesion molecule expression was measured to provide insight into the propensity of neutrophils and serum proteins to activate the inflammatory cascade, alone and in combination. METHODS Experiment protocol. In brief, endothelial cells in culture were coincubated with neutrophils in the presence or absence of fresh human serum. Flow cytometry was used to assess the expression of endothelial intercellular adhesion molecule 1 (ICAM-1) and endothelial leukocyte adhesion SURGERY 199
200 Van den Bogaerde et al.
Fig. 1. Schematic representation of the coincubation apparatus.
molecule 1 (ELAM-1) after 5 hours of the coincubation assay. Specific methods are described below. Endothelial cells. Human umbilical vein endothelial cells (HUVECs) were cultured from fresh (less than 24 hours old) human umbilical vein after a 10-minute incubation at 37° C with 0.2% collagenase (Sigma, St. Louis, Mo.).14 HUVECs were grown to confluence on gelatin-coated T25 flasks (Becton-Dickinson, Franklin Lakes, N.J.) in M-199 media with 10% human serum and 10% fetal bovine serum (Sigma) at pH 7.4. Next, confluent cells were passed to a gelatin-coated T75 flask with 0.1% collagenase, 0.1% ethylenediaminetetraacetic acid (EDTA), 0.25% bovine serum albumin and expanded. These cells showed typical endothelial cell structure in vitro and stained positively for factor VIII. For experimentation, the cells were passed to a 24-well plate (Becton-Dickinson) and a 60 mm dish and were used when they reached confluence. HUVECs from different umbilical veins were used for each experimental trial. Neutrophils. Venous blood was drawn from consenting volunteers into heparinized Vacutainer tubes (Becton-Dickinson) and brought to room temperature. A human erythrocyte agglutination reagent (Red Out; Robbins Scientific, Sunnyvale, Calif.) was added to aid in cell separation. The blood was layered over a commercial sodium metrizoate/dextran 500 sedimentation gradient (PMN; Robbins Scientific), and centrifuged. The second band of cells, corresponding to the polymorphonuclear neutrophils (PMNs), was isolated and washed twice with room temperature phosphate-buffered saline solution (PBS). PMN purity was found to be approximately 95% of leukocytes by using Wright’s stain; more than 98% of the cells excluded trypan blue. The cells were then counted and suspended in serum-free M-199 at a concentration of 3 × 106/ml and used immediately in the coincubation assay (below).
Surgery February 1998 Serum. Venous blood was drawn from 20 consenting volunteers into uncoated glass Vacutainer tubes and allowed to clot at room temperature for 45 minutes. The clot was contracted for 2 hours at 4° C, and the blood was centrifuged at 1000 g for 30 minutes at 4° C. Fresh human serum (FHS) was aspirated off the pellet and pooled. Half the FHS was heat inactivated (iHS) at 56° C for 30 minutes. The FHS and iHS were immediately frozen and stored in 10 ml aliquots at –70° C to ensure freshness for experimentation. Coincubation assay. HUVECs grown to confluence on a 24-well plate were rinsed with PBS. M-199 with 25% FHS was added to 12 of the wells, and M199 with 25% iHS was added to the other 12 wells. Semipermeable tissue culture inserts (BectonDickinson) containing approximately 1 × 106 PMNs each were added to 12 of the wells (Fig. 1). In the first series of experiments these inserts had pore diameters of 3.0 µm, which allowed PMN transmigration and HUVEC-PMN contact. Four experimental groups were studied on the basis of the contents of the wells: (1) HUVEC + PMN + FHS; (2) HUVEC + PMN + iHS; (3) HUVEC + FHS; (4) HUVEC + iHS. The same protocol was repeated in a second series of experiments using inserts with 0.45 µm pores, which blocked the migration of PMN across the membrane, eliminating PMNHUVEC contact. Each 24-well plate was then incubated at 37° C for 1, 3, or 5 hours. The semipermeable inserts containing the PMNs were discarded at the end of the coincubation period, and the HUVECs were rinsed with PBS and harvested with 0.1% collagenase and 0.25% EDTA for antibody labeling and flow cytometric analysis of ELAM-1 and ICAM-1 expression. Baseline HUVEC. HUVECs from the same umbilical vein as those used in the coincubation assay were grown to confluence on a 60 mm dish, rinsed with PBS, and incubated in M-199 with 25% FHS for 1 hour at 37° C. The cells were then rinsed with PBS and harvested with 0.1% collagenase and 0.25% EDTA for flow cytometric analysis of baseline HUVEC fluorescence and ELAM-1 and ICAM1 expression. Antibody labeling. HUVECs were incubated with monoclonal mouse antihuman ELAM-1 or ICAM-1 antibody (R&D Systems, Minneapolis, Minn.) for 20 minutes at 4° C and washed with PBS. The cells were then incubated with fluorescein isothiocyanate–conjugated monoclonal goat antimouse immunoglobin G antibody (Caltag Laboratories, South San Francisco, Calif.) for 20 minutes at 4° C
Surgery Volume 123, Number 2
Van den Bogaerde et al. 201
Fig. 2. Flow cytometry overlay plots of the ELAM-1 fluorescence of control HUVEC with that of HUVECs from the coincubation assays (shaded curve). The ordinate shows the number of fluorescent events and the abscissa shows the log fluorescence. A, HUVECs coincubated with PMNs and FHS show a large increase in ELAM-1 fluorescence, translating to an increase in ELAM-1 expression. HUVECs incubated with PMNs and iHS (B), FHS alone (C), or iHS alone (D) show no significant shift in fluorescence.
and washed with PBS. The cells were fixed in PBS with 0.01% paraformaldehyde. Flow cytometry. Flow cytometric analysis was performed with a FACScanner (fluorescence-activated cell sorter; Becton-Dickinson) with Lysis II software to quantitate ELAM-1 and ICAM-1 receptor expression. Forward and side scatter fluorescence data identified 10,000 viable HUVECs in each experimental group for unlabeled cells, nonspecific-antibody-labeled cells, ELAM-1-labeled cells, and ICAM-1-labeled cells. Fluorescence data were then accumulated on 10,000 cells at 530 nm. These fluorescence data were expressed in a dot histogram form of events versus log fluorescence and analyzed in comparison to the autofluorescence of unlabeled cells, as well as to the baseline fluorescence of ELAM-1- or ICAM-1-labeled cells. Group comparisons of percentage of shift from baseline were evaluated with Student’s paired t tests. Significance was assumed for p < 0.05. All data are expressed as percentage shift from baseline ELAM1 and labeled cells ± standard error of the mean. RESULTS Experiments using 3.0 µm inserts. Light microscopic observation of the HUVECs after 1, 3, and 5 hours of coincubation revealed that PMNs had migrated through the 3.0 µm pores of the inserts and adhered to the HUVECs in both wells containing FHS and those containing iHS. Flow cytometric analysis showed that ELAM-1 is not constitutively expressed on HUVECs. HUVECs coincubated with PMNs and FHS and stained for ELAM-1 had a large and statistically significant shift in fluorescence from baseline HUVECs. The other groups in the coincubation assay did not
Fig. 3. ELAM-1 percentage shift fluorescence at 3 and 5 hours for HUVECs coincubated with PMNs and FHS, compared with baseline HUVEC. *p < 0.05; **p < 0.01; n = 10.
show this shift (Fig. 2). Quantifying this increase in fluorescence revealed that after 3 hours of incubation with PMNs and FHS, HUVEC ELAM-1 fluorescence shifted 10.6% ± 3.1% compared with baseline autofluorescence. By 5 hours of coincubation with PMNs and FHS the ELAM-1 fluorescence shift was 25.8% ± 6.8% (p < 0.01; Fig. 3). Even after 5 hours of coincubation, ELAM-1 expression was not significantly increased in the PMN + iHS (9.3% ± 3.9%; p > 0.1), FHS alone (3.9% ± 0.8%; p > 0.1), or iHS alone (4.7% ± 1.2%; p > 0.1) groups (Fig. 4). ICAM-1 expression did not change in any of the experimental groups of HUVECs within the 5-hour time frame of these assays. Experiments using 0.45 µm inserts. Light microscopic observation of the HUVEC after 5 hours of coincubation confirmed that the 0.45 µm inserts prevented PMN transmigration, thereby eliminating HUVEC-PMN contact. In experiments using the smaller pore inserts
202 Van den Bogaerde et al.
Fig. 4. ELAM-1 percentage shift fluorescence at 5 hours for all four experimental groups, compared with baseline. HUVECs coincubated with PMNs and FHS show a significant increase in ELAM-1 expression over baseline cells. *p < 0.01; n = 10. All other groups show no significant shift.
Fig. 5. The percentage shift in ELAM-1 fluorescence at 5 hours for HUVECs coincubated with PMNs and FHS using inserts with 3.0 µm pores (PMN-HUVEC contact) or 0.45 µm pores (no contact). *p < 0.05; **0.1 < p < 0.05; n = 6.
there was an increase in ELAM-1 induction over baseline in HUVECs incubated with PMNs and FHS, although the magnitude was less than that observed in experiments in which PMN-HUVEC contact was permitted. ELAM-1 fluorescence shifted 9.6% ± 2.2% (p < 0.002) after 5 hours of coincubation with PMNs and FHS in the 0.45 µm wells, compared with a 21.8% ± 5.8% shift (p < 0.004) in the 3.0 µm wells (Fig. 5). DISCUSSION Extravasation of neutrophils into the interstitium is a prerequisite for tissue damage during acute inflammation. This process is initiated by endothelial surface alterations that cause neutrophils to roll along venule walls. Rolling is the result of an up-regulation of endothelial selectin molecules (ELAM-1), which interact with oligosaccharides constitutively present on the surface of neutrophils.15,16 ELAM-1 is expressed on the surface of endothelial cells after de novo synthesis; transport to the cell surface is stimulated by cytokines interleukin 1 (IL-1) and tumor necrosis
Surgery February 1998 factor–α (TNF-α).17 Interfering with these interactions can prevent later firm neutrophil adhesion to endothelial ICAM-l and subsequently activation and diapedesis.18-20 Data from this ex vivo study suggest that PMNs can directly induce endothelial ELAM-1 expression only in the presence of serum proteins. Because HUVECs incubated with fresh human serum in the absence of PMNs did not express ELAM-1, we hypothesize that a serum protein potentiates PMN secretion of diffusible mediators that induce ELAM-1 expression. The fact that the time course of ELAM-1 expression in this experiment was similar to that of endothelial cells stimulated with IL-1 and TNF-α raises the possibility that PMN induction of ELAM1 is mediated by these cytokines.21 In fact, PMNs are known to produce IL-l and TNF-α when stimulated.22 PMN production of reactive oxygen radicals after serum protein stimulation is another potential stimulus for the observed induction of ELAM-l. Morita et al.23 used intravital microscopy to show that leukocyte rolling could be attenuated with antioxidants superoxide dismutase or catalase. Significant ELAM-1 induction by PMN seems to be dependent on PMN proximity to the endothelium. When PMNs were placed in inserts with 0.45 µm pores, separating them from the endothelium by 2 µm, the magnitude of ELAM-1 expression was significantly reduced in comparison to the coincubation assay in which PMNs were free to migrate to the endothelium. Lack of PMN-endothelial contact, however, did not ablate ELAM-1 induction. This suggests that up-regulation of ELAM-1 does not depend on PMN-endothelial contact per se but is enhanced by the concentrating effects of PMN proximity to the endothelium. By extension, clinical situations involving stasis in the microcirculation would predispose the endothelium to PMNinduced activation. This same mechanism could also contribute to the “no-reflow” phenomenon in ischemia/reperfusion injury in which aggregating PMNs form microvascular plugs.24 A general pathway of PMN activation by serum proteins is furthered by the observation that infusion of active complement components into animals causes neutrophil aggregation, microvascular “plugging,” and complications consistent with PMN tissue infiltration such as adult respiratory distress syndrome and multiple organ failure.25 Preliminary clinical studies support the importance of both complement and neutrophils in these processes. For example, it has been shown in several studies that the expression of complement
Van den Bogaerde et al. 203
Surgery Volume 123, Number 2 receptors on PMN is up-regulated in clinical states of inflammation.26-28 This study further defines the relationship between the complement cascade and PMN induction of ELAM-l. Heating serum to 56° C for 30 minutes, a standard method of denaturing and inactivating complement proteins, ablated PMN induction of ELAM-1. In support of this observation, Varsano et al.29 recently showed that neutrophil adherence to respiratory epithelium is dependent on a heat labile factor in serum. Furthermore, Vaporciyan and Ward30 have shown that the C3 fragment iC3b activates PMN on interaction with PMN receptor CD11b/CD18. Small amounts of active C3 are produced in serum by spontaneous hydrolysis, which is then converted to iC3b.31 This alternative pathway of complement could therefore provide potent PMN activators in our coincubation assay. Like ELAM-1, ICAM-1 has been shown to be upregulated on the endothelium after IL-1 and TNFα stimulation. The shorter time course used in our coincubation experiments (5 hours) prevented any conclusions about induction of ICAM-1 because maximal expression of ICAM-1 occurs 12 hours after stimulation.21 It is also possible that heat labile serum proteins other than complement or complement fragments facilitate neutrophil induction of ELAM-1. Future studies involving the use of complement antagonists such as sCR1 or incubation media supplemented with serum deficient in specific complement fractions will help to further delineate the role of complement and specific complement fractions in the inflammatory response. Should these be confirmatory, it would support a trial of complement antagonists in decreasing tissue damage during acute inflammation. Theoretically, such pharmacotherapy could be applied to a wide variety of clinically important inflammatory processes including reperfusion injury, myocardial infarction, and hypovolemic shock.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16. 17. 18.
19.
20.
REFERENCES 1. Hearse DJ. Reperfusion of ischemic myocardium. J Mol Cell Cardiol 1977;9:605-15. 2. Parks DA, Granger DN. Contributions of ischemia and reperfusion to mucosal lesion formation. Am J Physiol 1986;250:G749-53. 3. Rauen U, Viebahn R, Lauchart W, De Groot H. The potential role of reactive oxygen species in liver ischemia/reperfusion injury following liver surgery. Hepatogastroenterology 1994;41:333-6. 4. Hernandez LA, Grisham MB, Twohig B, Arfors KE, Granger DN. Role of neutrophils in ischemia-reperfusion injury. Am J Physiol 1987;253:H699-708. 5. Hohn DC, Meyers AJ, Gherini ST, Beckman A, Markison
21. 22.
23.
24.
25.
RE, Churg AM. Production of acute pulmonary injury by leukocytes and activated complement. Surgery 1980;88:4857. Korthius RJ, Grishman MB, Granger DN. Leukocyte depletion attenuates vascular injury in postischemic skeletal muscle. Am J Physiol 1988;254:H823-7. Litt MR, Jeremy RW, Weisman HF, Winkelstein JA, Becker JC. Neutrophil depletion limited to reperfusion reduces myocardial infarct size after 90 minutes of ischemia: evidence for neutrophil mediated reperfusion injury. Circulation 1989;80:1816-27. Sisley AC, Desai T, Harig JM, Gewertz BL. Neutrophil depletion attenuates human intestinal reperfusion injury. J Surg Res 1994;57:192-6. Ma XL, Lefer DJ, Lefer AM, Rothlein R. Coronary endothelial and cardiac protective effects of a monoclonal antibody to intercellular adhesion molecule-1 in myocardial ischemia and reperfusion. Circulation 1992;86:937-46. Perry MA, Granger DN. Leukocyte adhesion in local versus hemorrhage-induced ischemia. Am J Physiol 1992;263:H810-5. Simpson PJ, Todd RF III, Fantone JC, Mickelson JK, Griffin JD, Lucchesi BR. Reduction of experimental canine myocardial reperfusion injury by a monoclonal antibody (anti-Mo-1, anti-CD11b) that inhibits leukocyte adhesion. J Clin Invest 1988;81:624-9. Smith EF, Griswold DE, Egan JW, Hillegass LM, Smith RA, Hibbs MJ, et al. Reduction of myocardial reperfusion injury with human soluble complement receptor type 1 (BRL 55730). Eur J Pharmacol 1993;236:477-81. Weisman HF, Bartow T, Leppo MK, et al. Soluble human complement receptor type 1: in vivo inhibitor of complement, suppressing post-ischemic myocardial inflammation and necrosis. Science 1990;249:146-51. Jaffe EA, Nachman RL, Becker CG, Minick CR. Culture of human endothelial cells derived from umbilical veins. J Clin Invest 1973;52:2745-56. Abbassi O, Kishimoto TK, McIntire LV, Anderson DC, Smith CW. E-Selectin supports neutrophil rolling in vitro under conditions of flow. J Clin Invest 1993;92:2719-30. McEver RP. Selectins. Curr Opin Immunol 1994;6:75-84. Albelda SM, Smith CW, Ward PA. Adhesion molecules and inflammatory injury. FASEB J 1994;8:504-12. Seekamp A, Till GO, Mulligan MS. Role of selectins in local and remote tissue injury following ischemia and reperfusion. Am J Pathol 1994;144:592-8. von Adrian UH, Hansell P, Chambers JD, et al. L-selectin function is required for beta-2 integrin-mediated neutrophil adhesion at physiologic shear rates in vivo. Am J Physiol 1992;263:H1034-44. Springer TA. Adhesion receptors of the immune system. Nature 1990;346:425-34. Pober JS, Cotran RS. The role of endothelial cells in inflammation. Transplantation 1990;50:537-44. Lloyd AR, Oppenheim JJ. Poly’s lament: the neglected role of the polymorphonuclear neutrophil in the afferent limb of the immune response. Immunol Today 1992;13:169-72. Morita Y, Clemens MG, Miller LS, et al. Reactive oxidants mediate TNF-α-induced leukocyte adhesion to rat mesenteric venular endothelium. Am J Physiol 1995;269:H183342. Kloner RA, Ganote CE, Jennings RB. The “no-reflow” phenomenon after temporary coronary occlusion in the dog. J Clin Invest 1974;54:1496-508. Jacob HS. The role of activated complement and granulo-
204 Van den Bogaerde et al.
cytes in shock states and myocardial infarction. J Lab Clin Med 1981;98:645-53. 26. Buyon JP, Shadick N, Berkman R, et al. Surface expression of Gp 165/95, the complement receptor type 3, CR3, as a marker of disease activity in systemic lupus erythematosus. Clin Immunol Immunopathol 1988;46:141-9. 27. Nelson RD, Hasslen SR, Ahrenholz DH, Solem LD. Influence of minor thermal injury on expression of complement receptor CR3 on human neutrophils. J Burn Care Rehabil 1987;8:496-502. 28. Tellado JM, Christou NV. Activation of polymorphonuclear
Surgery February 1998 leukocytes in surgical patients: characterization of surface receptor expression. Surgery 1993;113:624-30. 29. Varsano S, Joseph-Lerner N, Reshef T, Frolkis I. Normal serum increases adhesion of neutrophils to tracheal epithelial cells by a CD11b/CD18-dependent mechanism. Am J Respir Cell Mol Biol 1994;10:298-305. 30. Vaporciyan AA, Ward PA. Enhanced generation of superoxide anion by human neutrophils via a complement iC3b/Mac-1 interaction. Biol Signals 1993;2:126-35. 31. Fearon DT, Austen KF. Initiation of C3 cleavage in the alternative complement pathway. J Immunol 1975;115:1357-61.
ANNOUNCEMENT Effective October 1, 1997, please send all manuscripts and other submissions for Surgery to the following address: Kim Fons Managing Editor Surgery 11830 Westline Industrial Drive St. Louis, MO 63146-3318
Background. Although the individual actions of neutrophils and serum proteins such as complement in acute inflammation are well characterized, less is known about their effects in combination. We investigated the combined effects of neutrophil contact and active serum proteins on the expression of endothelial leukocyte adhesion molecule 1 (ELAM-1). Methods. Confluent monolayers of human umbilical vein endothelial cells were incubated with neutrophils in the presence and absence of fresh human serum. Flow cytometry was used to assess expression of endothelial intercellular adhesion molecule 1 (ICAM-1) and ELAM-1. In addition, neutrophils were retained in a semipermeable insert, which allowed their secretions to contact the endothelium but restricted neutrophil-endothelial contact. Results. ELAM-1 expression was significantly increased on the cells coincubated with neutrophils and fresh human serum (25.8%; p < 0.01). There was no significant change in ELAM-1 expression on endothelial cells incubated with fresh human serum alone (3.9%; p > 0.01) or in those incubated with neutrophils and heat-inactivated serum (9.3%; p > 0.01). In the absence of neutrophil contact, ELAM1 expression was increased only in the presence of fresh human serum (9.6%; p < 0.05). Conclusions. These findings suggest that serum proteins may potentiate the volume or potency of neutrophil-derived diffusable mediators of ELAM-1 expression. These effects are eliminated with the heat inactivation of serum proteins, implicating a heat sensitive mediator such as the complement cascade. (Surgery 1998;123:199-204.) From the Department of Surgery, University of Chicago, Chicago, Ill.
IN ISCHEMIA/REPERFUSION INJURY the tissue damage caused by inflammatory factors is often more extensive than the damage caused by the original ischemic insult.1-3 In vivo experiments suggest that such tissue damage is caused largely by infiltrating neutrophils because reperfusion of a wide range of ischemic organs with neutropenic blood greatly reduces the extent of tissue necrosis.4-8 Other studies have shown that using monoclonal antibodies against neutrophil adhesion molecules to block the molecular interactions between the endothelium and neutrophils also increases the survival of tissue after an ischemic insult.9-11 A great deal of recent information has further implicated the complement cascade in ischemia-reperfusion injury. With recombinant soluble complement receptor 1 (sCR1) to inhibit the activation of alterAccepted for publication May 20, 1997. Reprint requests: Bruce L. Gewertz, MD, Department of Surgery, University of Chicago, 5841 S. Maryland Ave., MC 5028, Chicago, IL 60637. Copyright © 1998 by Mosby, Inc. 0039-6060/98/$5.00 + 0 11/56/85438
native and classic complement cascades, investigators were able to reduce tissue damage in the ischemia-reperfused myocardium to an extent comparable to the studies using neutropenic blood.12,13 These findings raise the possibility that neutrophil-mediated damage and serum proteinmediated damage in acute inflammation may be related. We investigated the differential contributions of neutrophils and serum proteins to endothelial cell adhesion molecule expression in vitro. Endothelial cell adhesion molecule expression was measured to provide insight into the propensity of neutrophils and serum proteins to activate the inflammatory cascade, alone and in combination. METHODS Experiment protocol. In brief, endothelial cells in culture were coincubated with neutrophils in the presence or absence of fresh human serum. Flow cytometry was used to assess the expression of endothelial intercellular adhesion molecule 1 (ICAM-1) and endothelial leukocyte adhesion SURGERY 199
200 Van den Bogaerde et al.
Fig. 1. Schematic representation of the coincubation apparatus.
molecule 1 (ELAM-1) after 5 hours of the coincubation assay. Specific methods are described below. Endothelial cells. Human umbilical vein endothelial cells (HUVECs) were cultured from fresh (less than 24 hours old) human umbilical vein after a 10-minute incubation at 37° C with 0.2% collagenase (Sigma, St. Louis, Mo.).14 HUVECs were grown to confluence on gelatin-coated T25 flasks (Becton-Dickinson, Franklin Lakes, N.J.) in M-199 media with 10% human serum and 10% fetal bovine serum (Sigma) at pH 7.4. Next, confluent cells were passed to a gelatin-coated T75 flask with 0.1% collagenase, 0.1% ethylenediaminetetraacetic acid (EDTA), 0.25% bovine serum albumin and expanded. These cells showed typical endothelial cell structure in vitro and stained positively for factor VIII. For experimentation, the cells were passed to a 24-well plate (Becton-Dickinson) and a 60 mm dish and were used when they reached confluence. HUVECs from different umbilical veins were used for each experimental trial. Neutrophils. Venous blood was drawn from consenting volunteers into heparinized Vacutainer tubes (Becton-Dickinson) and brought to room temperature. A human erythrocyte agglutination reagent (Red Out; Robbins Scientific, Sunnyvale, Calif.) was added to aid in cell separation. The blood was layered over a commercial sodium metrizoate/dextran 500 sedimentation gradient (PMN; Robbins Scientific), and centrifuged. The second band of cells, corresponding to the polymorphonuclear neutrophils (PMNs), was isolated and washed twice with room temperature phosphate-buffered saline solution (PBS). PMN purity was found to be approximately 95% of leukocytes by using Wright’s stain; more than 98% of the cells excluded trypan blue. The cells were then counted and suspended in serum-free M-199 at a concentration of 3 × 106/ml and used immediately in the coincubation assay (below).
Surgery February 1998 Serum. Venous blood was drawn from 20 consenting volunteers into uncoated glass Vacutainer tubes and allowed to clot at room temperature for 45 minutes. The clot was contracted for 2 hours at 4° C, and the blood was centrifuged at 1000 g for 30 minutes at 4° C. Fresh human serum (FHS) was aspirated off the pellet and pooled. Half the FHS was heat inactivated (iHS) at 56° C for 30 minutes. The FHS and iHS were immediately frozen and stored in 10 ml aliquots at –70° C to ensure freshness for experimentation. Coincubation assay. HUVECs grown to confluence on a 24-well plate were rinsed with PBS. M-199 with 25% FHS was added to 12 of the wells, and M199 with 25% iHS was added to the other 12 wells. Semipermeable tissue culture inserts (BectonDickinson) containing approximately 1 × 106 PMNs each were added to 12 of the wells (Fig. 1). In the first series of experiments these inserts had pore diameters of 3.0 µm, which allowed PMN transmigration and HUVEC-PMN contact. Four experimental groups were studied on the basis of the contents of the wells: (1) HUVEC + PMN + FHS; (2) HUVEC + PMN + iHS; (3) HUVEC + FHS; (4) HUVEC + iHS. The same protocol was repeated in a second series of experiments using inserts with 0.45 µm pores, which blocked the migration of PMN across the membrane, eliminating PMNHUVEC contact. Each 24-well plate was then incubated at 37° C for 1, 3, or 5 hours. The semipermeable inserts containing the PMNs were discarded at the end of the coincubation period, and the HUVECs were rinsed with PBS and harvested with 0.1% collagenase and 0.25% EDTA for antibody labeling and flow cytometric analysis of ELAM-1 and ICAM-1 expression. Baseline HUVEC. HUVECs from the same umbilical vein as those used in the coincubation assay were grown to confluence on a 60 mm dish, rinsed with PBS, and incubated in M-199 with 25% FHS for 1 hour at 37° C. The cells were then rinsed with PBS and harvested with 0.1% collagenase and 0.25% EDTA for flow cytometric analysis of baseline HUVEC fluorescence and ELAM-1 and ICAM1 expression. Antibody labeling. HUVECs were incubated with monoclonal mouse antihuman ELAM-1 or ICAM-1 antibody (R&D Systems, Minneapolis, Minn.) for 20 minutes at 4° C and washed with PBS. The cells were then incubated with fluorescein isothiocyanate–conjugated monoclonal goat antimouse immunoglobin G antibody (Caltag Laboratories, South San Francisco, Calif.) for 20 minutes at 4° C
Surgery Volume 123, Number 2
Van den Bogaerde et al. 201
Fig. 2. Flow cytometry overlay plots of the ELAM-1 fluorescence of control HUVEC with that of HUVECs from the coincubation assays (shaded curve). The ordinate shows the number of fluorescent events and the abscissa shows the log fluorescence. A, HUVECs coincubated with PMNs and FHS show a large increase in ELAM-1 fluorescence, translating to an increase in ELAM-1 expression. HUVECs incubated with PMNs and iHS (B), FHS alone (C), or iHS alone (D) show no significant shift in fluorescence.
and washed with PBS. The cells were fixed in PBS with 0.01% paraformaldehyde. Flow cytometry. Flow cytometric analysis was performed with a FACScanner (fluorescence-activated cell sorter; Becton-Dickinson) with Lysis II software to quantitate ELAM-1 and ICAM-1 receptor expression. Forward and side scatter fluorescence data identified 10,000 viable HUVECs in each experimental group for unlabeled cells, nonspecific-antibody-labeled cells, ELAM-1-labeled cells, and ICAM-1-labeled cells. Fluorescence data were then accumulated on 10,000 cells at 530 nm. These fluorescence data were expressed in a dot histogram form of events versus log fluorescence and analyzed in comparison to the autofluorescence of unlabeled cells, as well as to the baseline fluorescence of ELAM-1- or ICAM-1-labeled cells. Group comparisons of percentage of shift from baseline were evaluated with Student’s paired t tests. Significance was assumed for p < 0.05. All data are expressed as percentage shift from baseline ELAM1 and labeled cells ± standard error of the mean. RESULTS Experiments using 3.0 µm inserts. Light microscopic observation of the HUVECs after 1, 3, and 5 hours of coincubation revealed that PMNs had migrated through the 3.0 µm pores of the inserts and adhered to the HUVECs in both wells containing FHS and those containing iHS. Flow cytometric analysis showed that ELAM-1 is not constitutively expressed on HUVECs. HUVECs coincubated with PMNs and FHS and stained for ELAM-1 had a large and statistically significant shift in fluorescence from baseline HUVECs. The other groups in the coincubation assay did not
Fig. 3. ELAM-1 percentage shift fluorescence at 3 and 5 hours for HUVECs coincubated with PMNs and FHS, compared with baseline HUVEC. *p < 0.05; **p < 0.01; n = 10.
show this shift (Fig. 2). Quantifying this increase in fluorescence revealed that after 3 hours of incubation with PMNs and FHS, HUVEC ELAM-1 fluorescence shifted 10.6% ± 3.1% compared with baseline autofluorescence. By 5 hours of coincubation with PMNs and FHS the ELAM-1 fluorescence shift was 25.8% ± 6.8% (p < 0.01; Fig. 3). Even after 5 hours of coincubation, ELAM-1 expression was not significantly increased in the PMN + iHS (9.3% ± 3.9%; p > 0.1), FHS alone (3.9% ± 0.8%; p > 0.1), or iHS alone (4.7% ± 1.2%; p > 0.1) groups (Fig. 4). ICAM-1 expression did not change in any of the experimental groups of HUVECs within the 5-hour time frame of these assays. Experiments using 0.45 µm inserts. Light microscopic observation of the HUVEC after 5 hours of coincubation confirmed that the 0.45 µm inserts prevented PMN transmigration, thereby eliminating HUVEC-PMN contact. In experiments using the smaller pore inserts
202 Van den Bogaerde et al.
Fig. 4. ELAM-1 percentage shift fluorescence at 5 hours for all four experimental groups, compared with baseline. HUVECs coincubated with PMNs and FHS show a significant increase in ELAM-1 expression over baseline cells. *p < 0.01; n = 10. All other groups show no significant shift.
Fig. 5. The percentage shift in ELAM-1 fluorescence at 5 hours for HUVECs coincubated with PMNs and FHS using inserts with 3.0 µm pores (PMN-HUVEC contact) or 0.45 µm pores (no contact). *p < 0.05; **0.1 < p < 0.05; n = 6.
there was an increase in ELAM-1 induction over baseline in HUVECs incubated with PMNs and FHS, although the magnitude was less than that observed in experiments in which PMN-HUVEC contact was permitted. ELAM-1 fluorescence shifted 9.6% ± 2.2% (p < 0.002) after 5 hours of coincubation with PMNs and FHS in the 0.45 µm wells, compared with a 21.8% ± 5.8% shift (p < 0.004) in the 3.0 µm wells (Fig. 5). DISCUSSION Extravasation of neutrophils into the interstitium is a prerequisite for tissue damage during acute inflammation. This process is initiated by endothelial surface alterations that cause neutrophils to roll along venule walls. Rolling is the result of an up-regulation of endothelial selectin molecules (ELAM-1), which interact with oligosaccharides constitutively present on the surface of neutrophils.15,16 ELAM-1 is expressed on the surface of endothelial cells after de novo synthesis; transport to the cell surface is stimulated by cytokines interleukin 1 (IL-1) and tumor necrosis
Surgery February 1998 factor–α (TNF-α).17 Interfering with these interactions can prevent later firm neutrophil adhesion to endothelial ICAM-l and subsequently activation and diapedesis.18-20 Data from this ex vivo study suggest that PMNs can directly induce endothelial ELAM-1 expression only in the presence of serum proteins. Because HUVECs incubated with fresh human serum in the absence of PMNs did not express ELAM-1, we hypothesize that a serum protein potentiates PMN secretion of diffusible mediators that induce ELAM-1 expression. The fact that the time course of ELAM-1 expression in this experiment was similar to that of endothelial cells stimulated with IL-1 and TNF-α raises the possibility that PMN induction of ELAM1 is mediated by these cytokines.21 In fact, PMNs are known to produce IL-l and TNF-α when stimulated.22 PMN production of reactive oxygen radicals after serum protein stimulation is another potential stimulus for the observed induction of ELAM-l. Morita et al.23 used intravital microscopy to show that leukocyte rolling could be attenuated with antioxidants superoxide dismutase or catalase. Significant ELAM-1 induction by PMN seems to be dependent on PMN proximity to the endothelium. When PMNs were placed in inserts with 0.45 µm pores, separating them from the endothelium by 2 µm, the magnitude of ELAM-1 expression was significantly reduced in comparison to the coincubation assay in which PMNs were free to migrate to the endothelium. Lack of PMN-endothelial contact, however, did not ablate ELAM-1 induction. This suggests that up-regulation of ELAM-1 does not depend on PMN-endothelial contact per se but is enhanced by the concentrating effects of PMN proximity to the endothelium. By extension, clinical situations involving stasis in the microcirculation would predispose the endothelium to PMNinduced activation. This same mechanism could also contribute to the “no-reflow” phenomenon in ischemia/reperfusion injury in which aggregating PMNs form microvascular plugs.24 A general pathway of PMN activation by serum proteins is furthered by the observation that infusion of active complement components into animals causes neutrophil aggregation, microvascular “plugging,” and complications consistent with PMN tissue infiltration such as adult respiratory distress syndrome and multiple organ failure.25 Preliminary clinical studies support the importance of both complement and neutrophils in these processes. For example, it has been shown in several studies that the expression of complement
Van den Bogaerde et al. 203
Surgery Volume 123, Number 2 receptors on PMN is up-regulated in clinical states of inflammation.26-28 This study further defines the relationship between the complement cascade and PMN induction of ELAM-l. Heating serum to 56° C for 30 minutes, a standard method of denaturing and inactivating complement proteins, ablated PMN induction of ELAM-1. In support of this observation, Varsano et al.29 recently showed that neutrophil adherence to respiratory epithelium is dependent on a heat labile factor in serum. Furthermore, Vaporciyan and Ward30 have shown that the C3 fragment iC3b activates PMN on interaction with PMN receptor CD11b/CD18. Small amounts of active C3 are produced in serum by spontaneous hydrolysis, which is then converted to iC3b.31 This alternative pathway of complement could therefore provide potent PMN activators in our coincubation assay. Like ELAM-1, ICAM-1 has been shown to be upregulated on the endothelium after IL-1 and TNFα stimulation. The shorter time course used in our coincubation experiments (5 hours) prevented any conclusions about induction of ICAM-1 because maximal expression of ICAM-1 occurs 12 hours after stimulation.21 It is also possible that heat labile serum proteins other than complement or complement fragments facilitate neutrophil induction of ELAM-1. Future studies involving the use of complement antagonists such as sCR1 or incubation media supplemented with serum deficient in specific complement fractions will help to further delineate the role of complement and specific complement fractions in the inflammatory response. Should these be confirmatory, it would support a trial of complement antagonists in decreasing tissue damage during acute inflammation. Theoretically, such pharmacotherapy could be applied to a wide variety of clinically important inflammatory processes including reperfusion injury, myocardial infarction, and hypovolemic shock.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16. 17. 18.
19.
20.
REFERENCES 1. Hearse DJ. Reperfusion of ischemic myocardium. J Mol Cell Cardiol 1977;9:605-15. 2. Parks DA, Granger DN. Contributions of ischemia and reperfusion to mucosal lesion formation. Am J Physiol 1986;250:G749-53. 3. Rauen U, Viebahn R, Lauchart W, De Groot H. The potential role of reactive oxygen species in liver ischemia/reperfusion injury following liver surgery. Hepatogastroenterology 1994;41:333-6. 4. Hernandez LA, Grisham MB, Twohig B, Arfors KE, Granger DN. Role of neutrophils in ischemia-reperfusion injury. Am J Physiol 1987;253:H699-708. 5. Hohn DC, Meyers AJ, Gherini ST, Beckman A, Markison
21. 22.
23.
24.
25.
RE, Churg AM. Production of acute pulmonary injury by leukocytes and activated complement. Surgery 1980;88:4857. Korthius RJ, Grishman MB, Granger DN. Leukocyte depletion attenuates vascular injury in postischemic skeletal muscle. Am J Physiol 1988;254:H823-7. Litt MR, Jeremy RW, Weisman HF, Winkelstein JA, Becker JC. Neutrophil depletion limited to reperfusion reduces myocardial infarct size after 90 minutes of ischemia: evidence for neutrophil mediated reperfusion injury. Circulation 1989;80:1816-27. Sisley AC, Desai T, Harig JM, Gewertz BL. Neutrophil depletion attenuates human intestinal reperfusion injury. J Surg Res 1994;57:192-6. Ma XL, Lefer DJ, Lefer AM, Rothlein R. Coronary endothelial and cardiac protective effects of a monoclonal antibody to intercellular adhesion molecule-1 in myocardial ischemia and reperfusion. Circulation 1992;86:937-46. Perry MA, Granger DN. Leukocyte adhesion in local versus hemorrhage-induced ischemia. Am J Physiol 1992;263:H810-5. Simpson PJ, Todd RF III, Fantone JC, Mickelson JK, Griffin JD, Lucchesi BR. Reduction of experimental canine myocardial reperfusion injury by a monoclonal antibody (anti-Mo-1, anti-CD11b) that inhibits leukocyte adhesion. J Clin Invest 1988;81:624-9. Smith EF, Griswold DE, Egan JW, Hillegass LM, Smith RA, Hibbs MJ, et al. Reduction of myocardial reperfusion injury with human soluble complement receptor type 1 (BRL 55730). Eur J Pharmacol 1993;236:477-81. Weisman HF, Bartow T, Leppo MK, et al. Soluble human complement receptor type 1: in vivo inhibitor of complement, suppressing post-ischemic myocardial inflammation and necrosis. Science 1990;249:146-51. Jaffe EA, Nachman RL, Becker CG, Minick CR. Culture of human endothelial cells derived from umbilical veins. J Clin Invest 1973;52:2745-56. Abbassi O, Kishimoto TK, McIntire LV, Anderson DC, Smith CW. E-Selectin supports neutrophil rolling in vitro under conditions of flow. J Clin Invest 1993;92:2719-30. McEver RP. Selectins. Curr Opin Immunol 1994;6:75-84. Albelda SM, Smith CW, Ward PA. Adhesion molecules and inflammatory injury. FASEB J 1994;8:504-12. Seekamp A, Till GO, Mulligan MS. Role of selectins in local and remote tissue injury following ischemia and reperfusion. Am J Pathol 1994;144:592-8. von Adrian UH, Hansell P, Chambers JD, et al. L-selectin function is required for beta-2 integrin-mediated neutrophil adhesion at physiologic shear rates in vivo. Am J Physiol 1992;263:H1034-44. Springer TA. Adhesion receptors of the immune system. Nature 1990;346:425-34. Pober JS, Cotran RS. The role of endothelial cells in inflammation. Transplantation 1990;50:537-44. Lloyd AR, Oppenheim JJ. Poly’s lament: the neglected role of the polymorphonuclear neutrophil in the afferent limb of the immune response. Immunol Today 1992;13:169-72. Morita Y, Clemens MG, Miller LS, et al. Reactive oxidants mediate TNF-α-induced leukocyte adhesion to rat mesenteric venular endothelium. Am J Physiol 1995;269:H183342. Kloner RA, Ganote CE, Jennings RB. The “no-reflow” phenomenon after temporary coronary occlusion in the dog. J Clin Invest 1974;54:1496-508. Jacob HS. The role of activated complement and granulo-
204 Van den Bogaerde et al.
cytes in shock states and myocardial infarction. J Lab Clin Med 1981;98:645-53. 26. Buyon JP, Shadick N, Berkman R, et al. Surface expression of Gp 165/95, the complement receptor type 3, CR3, as a marker of disease activity in systemic lupus erythematosus. Clin Immunol Immunopathol 1988;46:141-9. 27. Nelson RD, Hasslen SR, Ahrenholz DH, Solem LD. Influence of minor thermal injury on expression of complement receptor CR3 on human neutrophils. J Burn Care Rehabil 1987;8:496-502. 28. Tellado JM, Christou NV. Activation of polymorphonuclear
Surgery February 1998 leukocytes in surgical patients: characterization of surface receptor expression. Surgery 1993;113:624-30. 29. Varsano S, Joseph-Lerner N, Reshef T, Frolkis I. Normal serum increases adhesion of neutrophils to tracheal epithelial cells by a CD11b/CD18-dependent mechanism. Am J Respir Cell Mol Biol 1994;10:298-305. 30. Vaporciyan AA, Ward PA. Enhanced generation of superoxide anion by human neutrophils via a complement iC3b/Mac-1 interaction. Biol Signals 1993;2:126-35. 31. Fearon DT, Austen KF. Initiation of C3 cleavage in the alternative complement pathway. J Immunol 1975;115:1357-61.
ANNOUNCEMENT Effective October 1, 1997, please send all manuscripts and other submissions for Surgery to the following address: Kim Fons Managing Editor Surgery 11830 Westline Industrial Drive St. Louis, MO 63146-3318