(PMN) Phagocytosis and Chemotaxis after Reperfusion Injury
JULIE A. FREISCHLAG, Department Presented
at the Annual
of the Association
M.D., AND DINAH
VA and UCLA Medical of Veterans
Centers, Los Angeles, California
clear. Theories of PMN activation have been proposed once oxygen is returned to the skeletal muscle bed by the reperfusion of blood. The first step in PMN activation is chemotaxis which may allow the PMN to migrate through the injured vascular cell junctions and into the surrounding tissues [3, 41. Once present in the tissues, activated PMN than can release their lysosomal enzymes as well as superoxide anion. PMN phagocytosis is required for such lysosomal enzyme release and may cause even further tissue destruction. Previously, we have demonstrated that PMN phagocytosis and chemotaxis were significantly increased after only 2 hr of ischemia followed by 1 hr of reperfusion . Two hours of ischemia in our rabbit hindlimb model of ischemia and reperfusion did not result in a clinical reperfusion injury. Therefore, the PMN was stimulated early in ischemia prior to a reperfusion injury. Therefore, the purpose of this study was to determine PMN phagocytosis and chemotaxis after a longer period of ischemia which is followed by a severe reperfusion injury once blood flow is returned. In our model, 3 hr of ischemia followed by 1 hr of reperfusion results in such a severe clinical reperfusion injury. Specifically, our purpose was to determine if PMN phagocytosis and chemotaxis is further stimulated when ischemia is allowed to progress. The second objective of this study was to evaluate the effect of the serum after the ischemia and reperfusion periods on the PMN. Both assays for phagocytosis and chemotaxis utilize serum; therefore different sera could be easily compared.
Neutrophils (PMN) have been implicated as mediators of the reperfusion injury which occurs in skeletal muscle after &hernia. This study was performed to measure PMN phagocytosis and chemotaxis after 3 hr of ischemia followed by 1 hr of reperfusion in a model where a significant reperfusion injury occurred. Baseline blood samples were drawn from an ear artery from New Zealand white rabbits for PMN and serum. The right iliac and femoral arteries were clamped for 3 hr which resulted in a severe clinical reperfusion injury. Just prior to clamp release, blood was harvested from the right iliac vein. After 1 hr of reperfusion, blood was again harvested from the right iliac vein. Phagocytosis was measured by the percentage ingestion of zymosan beads by the PMN. The zymosan beads had been opsonized with baseline (b), ischemia (i), or reperfusion (r) serum. Results for phagocytosis revealed no difference for (b) PMN when opsonized by (b), (i), or (r) serum. A significant increase was seen in (i) PMN phagocytosis when (i) or (r) serum was present. Also, a significant increase in (r) PMN phagocytosis was seen when (i) serum was present (ANOVA: F = 14.47; P = 0.0002). Chemotaxis was evaluated by the number of PMN migrating across a filter. Serum obtained from (b), (i), and (r) blood samples served as the chemoattractants. Significant increases in chemotaxis were observed for (b) , (i), and (r) PMN when (i) serum was used as the chemoattractant (ANOVA: F = 7.11; P = 0.0026). We conclude: (1) Rabbit PMN harvested after &hernia and reperfusion demonstrated increased phagocytosis when (i) serum was present. (2) Rabbit PMN harvested at baseline, after ischemia and after reperfusion, demonstrated increased chemotaxis when (i) serum was the chemoattractant. (3) PMN functions of phagocytosis and chemotaxis are stimulated by serum obtained after 0 1992 Academic Press, Inc. severe ischemia.
Male New Zealand white rabbits (2-5 kg) were bought from Universal Rabbitry, California. The procedures performed upon the rabbits were approved by the animal research committee and met their guidelines. The rabbits were anesthetized with acepromazine (13 mgjkg) and ketamine (100 mg/kg) which were given intramuscularly. Baseline blood samples (10 ml) were drawn from an ear artery at the beginning of the experiment. PMN and serum were harvested from this blood sample as well
PMN have been identified as participants in the reperfusion injury seen in skeletal muscle after ischemia [l, 21. Their mechanism of action however remains un0022.4804/92 $1.50 Copyright 0 1992 by Academic Press, Inc. All rights of reproduction in any form reserved.
as from the subsequent blood samples. A midline laparotomy and right groin incision were performed and the aorta and both iliac and right femoral arteries were isolated. All collateral branches were ligated from the aorta to the right common femoral artery. Also, both internal iliac arteries were ligated. The right profunda femoris artery was ligated as well [ 61. Vascular clamps were then applied to the right common iliac and common femoral arteries for 3 hr. Absence of the right anterior tibialis pulse after placement of the clamps was confirmed by Doppler. In previous experiments, angiograms had been performed in rabbits undergoing this devascularization procedure and these demonstrated complete interruption of all blood supply to the ischemic limb. The rabbits were monitored by an arterial line placed in the carotid artery. The blood pressure remained constant throughout the experiment as measured by mean arterial pressure with deviation of 5-7 mm Hg change seen during blood draws. Body temperature was maintained by heating pads and warming lights. An intravenous line was placed at the beginning of the experiment in an ear vein and normal saline was administered at 5 ml/kg/hr. After the 3-hr period of ischemia had elapsed, a ZOgauge angiocatheter (Critikon) was placed in the right common iliac vein. A blood sample (10 ml) was drawn from the right common iliac vein just prior to the removal of the vascular clamps from the right iliac and femoral arteries. PMN and serum were harvested from this blood sample as well. After the clamps were removed, the presence of the anterior tibialis pulse was confirmed by Doppler. After 1 hr of reperfusion, another blood sample (10 ml) was drawn from the right iliac vein via the same angiocatheter for PMN and serum. The rabbit was then sacrificed. All three blood samples [baseline (b), ischemia (i), and reperfusion (r)] were combined with phosphate-buffered saline with glucose and gelatin (PBSG-G) in a 1:l ratio. This mixture was added and thoroughly mixed with an equal volume of 6% dextran (Sigma Chemical Company) in order to allow the erythrocytes to sediment for 45 min. The supernatant was removed and centrifuged at 1200 rpm for 15 min. The pellet was resuspended in PBSG-G (8 ml) and layered over Histopague 1077 (3 ml) (Sigma). This was spun at 1500 rpm for 30 min to allow the lymphocytes and monocytes to separate from the PMN. The bottom layer was resuspended in PBSG-G and lysed with distilled water in order to remove any remaining erythrocytes, and 3.5% saline was added to restore osmolarity. This was centrifuged at 1200 rpm for 15 min and the isolated PMN were resuspended in Hank’s balanced salt solution (Sigma). A hemochromocytometer was used to count the PMN and they were diluted to a concentration of 2 X lo6 cells/ml. Differential slides were prepared to document purity and the trypan blue exclusion test was used to document PMN viability.
Phagocytosis was determined using a modified method of Dankberg and Persidsky [7, 81. Zymosan beads taken up by the PMN were analyzed under ultraviolet light in dark field illumination. Zymosan particles (Sigma) were incubated with (b), (i), or (r) serum from each individual rabbit which served as opsonins. A concentration of 10 mg/ml was incubated at 37°C for 60 min. This combination was then spun at 1200 rpm for 10 min. The serum was aspirated and the zymosan beads were resuspended in phosphate-buffered saline with glucose but not gelatin. Fifty microliters of the zymosan solution was added to 250 ~1 of the PMN suspension. This combination was incubated at 27°C for 10 min. A fluorescein diacetate (FDA-Nutrition Biochemical) and ethidium bromide (2,7-diamino-lo-ethyl-g-phenylphenan-thridium bromide; EB, - Nutrition Biochemical) solution was then added in the following concentration: 4 ~1 FDA, 200 ~1 EB, and 9.8 ml of phosphate-buffered saline with glucose but not gelatin. This solution in a volume of 125 ~1 was added to the PMN and zymosan. This was incubated at 27°C for 10 min. Fifty microliters of this mixture was analyzed under ultraviolet light using oil immersion. A total of 100 PMN were counted for each sample and the number of PMN containing zymosan beads was determined. Statistical analysis was performed using analysis of variance. Chemotaxis was evaluated using a Neuro Probe Chamber (Neuro Probe, Cabin John, MD) [9, lo]. The (b), (i), and (r) serum from each rabbit served as the chemoattractants and were placed in the bottom wells of the chamber. The (b), (i), and (r) PMN in a concentration of 2 X lo6 cells/ml were placed in the top wells of the chamber after a 5-pm filter had been placed in the middle of the chamber. The entire chamber was then placed in a CO, incubator for 45 min. The filter was removed, rinsed, and dried. After staining, the filter was mounted on a glass microscope slide. Each sample was run in triplicate. The PMN were counted under oil immersion to determine the number of PMN which had migrated across the filter. Statistical analysis was performed using analysis of variance. RESULTS
Differential counts revealed PMN purity to be 90100%. PMN viability as determined by the trypan blue exclusion test was 95-97%. The results of phagocytosis are seen in Table 1 and Fig. 1. As a group, (b) PMN demonstrated no difference in phagocytosis whether (b), (i), or (r) serum was used as an opsonin. A significant increase in phagocytosis was seen by (i) PMN in the presence of (i) and (r) serum. A similar increase in phagocytosis was observed with (r) PMN only in the presence of (i) serum. (ANOVA: F = 14.47; P = 0.0002) The results of chemotaxis are seen in Table 2 and Fig. 2. As a group, (b) PMN demonstrated an increase in
TABLE Neutrophil n
10 10 10 10 10 10 10 10 10
Baseline Baseline Baseline Ischemia Ischemia Ischemia F&perfusion Renerfusion
Baseline Ischemia Reperfusion Baseline Ischemia Reperfusion Baseline Ischemia Reperfusion
14.8 16.7 15.2 13.1 23.2 17.0 13.1 20.9 14.6
?Z 1.1 t 0.9 t 1.1 t 0.8 t- 1.9” k 0.9” + 1.3 + 1.6” + 1.6
In the investigation of reperfusion injury in skeletal muscle, PMN have been identified as the source of free
chemotaxis when (i) serum was used as the chemoattractant but not with (r) serum. A similar but enhanced increase in chemotaxis was seen in both (i) PMN and (r) PMN when (i) serum served as the chemoattractant (ANOVA: F = 7.11; P = 0.0025); no such changes were seen with (r) serum. All rabbits demonstrated a clinically severe reperfusion injury which was manifested by a stiff limb despite a Doppler signal having been obtained. Microscopically, no PMN were demonstrated to have migrated as of yet in the areas surrounding the capillaries or in the muscle or surrounding connective tissue. After 3 hr of ischemia and 1 hr of reperfusion, edema and muscle bundle destruction was present in the microscopic sections of the muscle biopsied.
52, NO. 3, MARCH
Note. PMN, neutrophii; hpf, high-powered field. “Using ANOVA: F = 14.47; P = 0.0002 as compared PMN.
10 10 10 10 10 10 10 10 10
Baseline Baseline Baseline Ischemia Ischemia Ischemia Reperfusion Reperfusion Reperfusion
Baseline Ischemia Reperfusion Baseline Ischemia Reperfusion Baseline Ischemia Reperfusion
18.9 34.4 21.8 25.1 45.1 23.1 25.1 45.6 24.1
Note. PMN, neutrophil; hpf, high-powered field. “Using ANOVA: F = 7.11; P = 0.0025 as compared serum.
+ * + f 2 k -c + +
+ SE 1.2 4.7 2.8 3.0 6.4” 1.7 3.0 4.6” 2.6
radicals due to the fact that skeletal muscle possesses little xanthine dehydrogenase [ 111.PMN activation has been hypothesized to occur during the ischemic interval and is amplified by reperfusion with the return of oxygen to the muscle bed. Oxygen activates the superoxide anion surface NADPH system present on the PMN. Previous studies have shown a reduction in the reperfusion injury when PMN were removed prior to the ischemit interval. Korthius and colleagues demonstrated less vascular permeability and resistance in ischemic canine gracilis muscles that were reperfused with blood without PMN [ 11. In skeletal muscle, alterations in vascular permeability, vascular endothelial cell swelling, and PMN migration have been observed prior to electron microscopic evidence of skeletal muscle damage [ 121. Therefore, alterations in the microvasculature occur early in ischemia and prior to clinical skeletal muscle destruction. Other indirect evidence demonstrating that the PMN plays a major role in the reperfusion injury has utilized the administration of pharmacologic agents that specifi-
ig 50 % 3 40 FL 1 3o 2 20 i r:
FIG. 1. Bar graph demonstrating mean phagocytosis of zymosan beads by PMN (neutrophils) having been opsonized by baseline, ischemia, or reperfusion serum after 3 hr of ischemia and 1 hr of reperfusion.
10 0 Baseline PMN
FIG. 2. Bar graph demonstrating mean chemotaxis of PMN (neutrophils) with baseline, ischemia, or reperfusion serum serving as the chemoattractants after 3 hr of ischemia and 1 hr of reperfusion.
tally act upon the PMN. Using a canine acute myocardial ischemia model, Engler pretreated a group of animals with AICA-riboside (5-amino-4-imidazole carboxamide-riboside) which stimulates adenosine release from tissue cells [ 131. Adenosine is able to prevent superoxide radical formation by PMN. AICA riboside pretreatment allowed an increase in collateral blood flow, a decrease in arrhythmias usually seen after reperfusion in this model, and a lo-fold elevation in tissue adenosine. In another myocardial ischemia experiment, Simpson and colleagues administered iloprost during the ischemit interval and during reperfusion in dogs undergoing 90 min of left circumflex coronary artery occlusion followed by 2 hr of reperfusion [ 141. Myocardial infarct size was significantly smaller in the iloprost group and PMN accumulation at 6 hr after reperfusion was markedly less as measured by the myeloperoxidase activity determined from the myocardial tissue. The mechanism of action of iloprost on the PMN is unknown. PMN phagocytosis and chemotaxis are increased when PMN are activated. Many nonspecific substances can stimulate the PMN and include many which are released into the blood. These include complement, interleukin-1, tumor necrosis factor, and platelet-derived growth factor [ 15-181. These substances may be specifically released by the endothelial cells in the area of ischemia which may activate PMN which are traveling by. Normally, PMN come into contact with the endothelial cells of the capillaries. In reperfusion injury models, PMN have actually been demonstrated to become attached to the endothelial cell and obstruct small capillaries leading to more areas of “no reflow” which may worsen the reperfusion injury picture [ 191. PMN phagocytosis and chemotaxis are the first steps in PMN activation; therefore, we chose to evaluate them in an ischemia/reperfusion model. In our first study, we evaluated PMN phagocytosis and chemotaxis after 2 hr of &hernia followed by 1 hr reperfusion . In our model, 2 hr of ischemia resulted in no clinical reperfusion injury and microscopically no PMN infiltration of the muscle or tissues was seen. The architecture of the skeletal muscle was intact without edema. Significant increases in phagocytosis and chemotaxis were seen after 2 hr of ischemia as well as after 1 hr of reperfusion. The serum used as either an opsonin in the phagocytosis assay or as a chemoattractant in the chemotaxis assay did not influence the increases seen. In the present study, PMN phagocytosis and chemotaxis were studied after 3 hr of ischemia followed by 1 hr of reperfusion. Three hours of ischemia in our model results in a severe clinical reperfusion injury which is manifested by a stiff and swollen experimental limb. Microscopically, again, no PMN were identified to have migrated into the skeletal muscle or surrounding tissues. However, there was edema of the skeletal muscle bundles but the architecture was intact. The (b) PMN did not demonstrate any increase in phagocytosis when op-
sonized with (b), (i), or (r) serum which is similar to the effect seen with (b) PMN tested in the 2-hr ischemia study. After 3 hr of ischemia, the (i) PMN only demonstrated increases in phagocytosis when (i) and (r) serum were used as the opsonins. Similarly, phagocytosis by the (r) PMN was stimulated only by (i) serum. Such serum influences were not seen in the 2-hr ischemia study. After 3 hr of ischemia, chemotaxis in all groups of PMN was stimulated only by (i) serum. In the 2-hr ischemia study, PMN chemotaxis was stimulated after the ischemic and reperfusion interval without specific differences being seen between the serum used. These findings suggest that when a severe reperfusion occurs (the 3-hr ischemia group), an interaction is being observed between the PMN and the serum which was not seen in the 2-hr ischemia (no reperfusion injury) study. Some substance in the serum may be originating from the vascular endothelial cells which are also being injured during longer periods of ischemia [ 201. Clearly, one of the mechanisms of the PMN response to ischemia which leads to reperfusion injury involves the activation of PMN phagocytosis and chemotaxis prior to the PMN migration into the surrounding tissues which causes further damage. In summary, in this model of rabbit ischemia followed by reperfusion, 3 hr of ischemia followed by 1 hr of reperfusion resulted in a severe clinical reperfusion injury. Phagocytosis of (i) PMN was stimulated by (i) and (r) serum and phagocytosis of (r) PMN was stimulated by (i) serum alone. Chemotaxis of(b), (i), and (r) PMN was increased by (i) serum. This serum-PMN interaction was not seen in a shorter ischemic interval which did not result in a reperfusion injury (2-hr ischemia). Future studies should involve the identification of the factor present in the (i) serum which is modifying the PMN. Administration of a pharmacological agent after the ischemic interval prior to reperfusion in order to alter this substance may ablate the reperfusion injury seen. ACKNOWLEDGMENT The authors thank Jeffrey with the statistical analysis.
Dr. PH, for his assistance
REFERENCES Korthuis, R. J., Granger, D. N., Townsley, M. I., and Taylor, A. E. The role of oxygen-derived radicals in ischemia-induced increases in canine skeletal muscle vascular permeability. Circ. Res. 57: 599, 1985. Korthuis, R. J., Grisham, M. B., and Granger, D. N. Leukocyte depletion attenuates vascular injury in post ischemic skeletal muscle. Am. J. Physd. 254: H823,1988. Wedmore, C. V., and Williams, T. J. Control of vascular permeability by polymorphonuclear leukocytes in inflammation. Nature 289: 646, 1981. Tonnesen, M. G. Neutrophil-endothelial cell interactions: Mech-
anisms of neutrophil adherence to vascular vest. Dermatol. 93(2 suppl): 53s, 1989.
5. Freischlag, J. A., and Hanna, D. Neutrophil and chemotaxis 1991.
(PMN) phagocytosis after 2 hours of ischemia. J. Surg. Res. 50: 648,
Engler, R. Consequences of activation and adenosine-mediated inhibition of granulocytes during myocardial ischemia. Fed. Proc. 46: 2407, 1987.
14. Simpson, P. J., Mickelson,
J., Fantone, J. C., Gallagher, K. P., and Lucchesi, B. R. Iloprost inhibits neutrophil function in vitro and in vivo and limits experimental infarct size in canine heart. Circ. Res. 60: 666, 1987.
H., Smith, L., and Freischlag, J. A. The effects of one hour of ischemia on neutrophil chemotaxis. Surg. Forum 40: 295,1989.
52, NO. 2, FEBRUARY
Ward, P. A., Cochrane, C. G., and Muller-Eberhard, H. J. The role of serum complement in chemotaxis in vitro. J. Exp. Med.
Schleimer, endothelial lation with bol diesters.
9. Falk, W., Goodwin,
Seymour, S. J., Vadas, M. A., Harlan, J. M., Sparks, L. H., Gamble, J. R., Agosti, J. M., and Walterdorph, A. M. Stimulation of neutrophils by tumor necrosis factor. J. Zmmunol. 136: 4220, 1986.
Deuel, T. F., Senior, R. M., Huang, J. S., and Griffin, G. L. Chemotaxis of monocytes and neutrophils to platelet-derived growth factor. J. Clin. Inuest. 69: 1046, 1982.
Lindbom, L., and Arfors, K. E. Mechanisms and site of control for variation in the number of perfused capillaries in skeletal muscle. Znt. J. Microcirc. 4: 19, 1985.
F., and Persidsky, rity and phagocytic function.
M. D. A test of granulocyte Cryobiology 13: 340, 1976.
J. A., Backstrom, B., Kelly, D., Keehn, G., and Busuttil, R. W. Comparison of blood and peritoneal neutrophil (PMN) function in rabbit with and without peritonitis. J. Surg. Res. 40: 145, 1986. R. H., and Leonard, E. J. A 48 well micro chemotaxis assembly for rapid and accurate measurement of leukocyte migration. J. Zmmunol. Methods 33: 239, 1980.
Harvath, L., Falk, W., and Leonard, E. J. Rapid quantification of neutrophil chemotaxis: Use of a polyvinylpyrrolidone-free polycarbonate membrane in a multiwell assembly. J. Immunol. Methods 37: 39, 1980.
Suval, W. D., Duran, W. N., Boric, M. P., Hobson, R. W., III, Berendsen, P. B., and Ritter, A. B. Microvascular transport and endothelial cell alterations preceding skeletal muscle damage in ischemia and reperfusion injury. Am. J. Surg. 154: 211,1987.
B., and &ward, P. J. The histochemical oxidase. Histochem. J. 10: 615, 1978.
R. P., and Rutledge, B. K. Cultured human vascular cells acquire adhesiveness for neutrophils after stimuinterleukin 1, endotoxin and tumor-promoting phorJ. Immunol. 136: 649, 1986.
T., Jensen, M., and Chait, A. Human arterial walls secrete factors that are chemotactic for monocytes. Proc. N&l. Acad. Sci. USA 80: 5094,1983.