Normalization of Brain Death—Induced Injury to Rat Renal Allografts by Recombinant Soluble P-Selectin Glycoprotein Ligand
Martin Gasser*,
Ana Maria Waaga,,
Joana E. Kist-Van Holthe,
Susanne M. Lenhard,
Igor Laskowski*,
Gray D. Shaw,
Wayne W. Hancock|| and
Nicholas L. Tilney*
*Surgical Research Laboratory, Harvard Medical School, Department of Surgery, Department of Nephrology, Brigham and Women’s Hospital, and Children’s Hospital, Boston, Massachusetts; and Genetics Institute/Wyeth Research and ||Millennium Inc., Cambridge, Massachusetts.
Correspondence to Dr. Nicholas L. Tilney, Department of Surgery, Brigham and Women’s Hospital, 75 Francis Street, Boston, MA 02115; Phone: 617-732-6817; Fax: 617-232-9576; E-mail: bhayslett{at}rics.bwh.harvard.edu
ABSTRACT. Donor brain death has been considered a significantrisk factor for both early and late organ allograft dysfunction.This central injury not only evokes an upsurge of catecholamineswith resultant peripheral tissue vasoconstriction and ischemiabut also promotes release of hormones and inflammatory mediatorsthat may also affect the organs directly. One of the resultantinfluences of these events is the rapid upregulation of theacute-phase adhesion molecules, the selectins. These initiateleukocyte adhesion to vascular endothelium and trigger subsequentcellular and molecular changes in the compromised tissues. Anestablished F344 LEW rat model of chronic rejection was usedto examine (1) whether the initial inflammatory events thatdevelop within kidney allografts from brain-dead donors couldbe normalized using a recombinant soluble form of P-selectinglycoprotein ligand and (2) whether amelioration of these earlychanges would alter the inexorable progression of chronic allograftrejection. Untreated living donor controls experienced unrelentingchronic rejection over time. This complex process was acceleratedin brain-dead donor kidneys. Treatment with P-selectin glycoproteinligand prevented the early inflammatory changes in the transplantedorgans and their subsequent (200 d) functional and morphologicmanifestations, particularly when the soluble ligand was administeredboth to the donor before organ removal and to the recipientafter engraftment. This strategy of using a naturally occurringselectin ligand to prevent donor-associated chronic graft dysfunctionmay be of special clinical interest in cadaver donor transplantation.
The clinical findings that kidneys from living unrelated donorsperform as well over the short and long term as those from livingrelated sources and that grafts from both donor groups are invariablysuperior to those from cadavers have emphasized that antigen-independentevents, which include increased donor age and intercurrent disease,the state of brain death, and the period of ischemia, influencethe quality of solid organs after transplantation and are importantrisk factors for their eventual outcome (1,2). Brain death,a central catastrophe unique to the cadaver organ donor, producesprofound physiologic and structural derangements in the peripheraltissues of experimental animals both before and after placementin the recipient (3,4). These include massive upregulation ofmajor histocompatibility antigens, adhesion molecules, cytokines,and other acute-phase proteins. Ischemia with systemic vasoconstrictionis an important facet of brain death, secondary to the initialburst of catecholamines released into the circulation (5). Theearly injury and the associated reperfusion after transplantationevoke nonspecific inflammatory changes in the affected organs.However, additional factors may contribute to the peripheraleffects of brain injury. Important changes in the dynamics ofa series of hormones have been identified (6,7). Because braindeath may also influence an organ differently than global ischemiaalone, patterns of infiltration of circulating leukocytes throughthe tissues may also vary between the two conditions (8,9).
One result of these physiologic shifts is to cause prompt upregulationof selectins, cell-surface glycoproteins responsible for earlyrecruitment of leukocytes to sites of injury (10,11). P-selectinis translocated within minutes from intracellular stores tothe surface of vascular endothelial cells and/or platelets inresponse to inflammatory stimuli. E-selectin is then expressedon endothelial surfaces after transcriptional induction of itsmRNA. These adhesion molecules react with their ligands on circulatingpolymorphonuclear leukocytes (PMN) to promote transient sticking(tethering) and slowing (rolling) along vessel walls. With strongerattachment to endothelium and diapedesis into the tissues viathe sequential activity of other adhesion molecules, PMN becomeactivated by locally produced chemokines and cytokines and triggera further cascade of inflammatory/immunologic events.
A recombinant soluble form of P-selectin glycoprotein ligand(rPSGL-Ig) has been shown to inhibit the initial selectin activityand subsequent cellular and humoral events in global ischemia/reperfusion(12). In the present study, we examined its effects on kidneyallografts from brain-dead (BD) donors, a putatively more complexmodel. Blockade of this initial step has allowed assessmentof the influence of the agent on both early and late changesin long-surviving renal transplants in rats.
Animals and Operative Technique
Inbred male rats (Harlan Sprague-Dawley, Indianapolis, IN),8 to 10 wk of age and weighing 200 to 250 g, were used throughoutthe experiments. Lewis rats (LEW, RT1l) served as recipientsof renal isografts from LEW donors or allografts from Fisher(F344, RT1lvl) donors. The organs were flushed with 3 ml ofcold lactated Ringers solution before removal and stored transientlyin the cold solution (4°C) before heterotopic transplantationto the host abdominal great vessels. The left native kidneyof the recipients was removed during the engraftment procedureand the right was removed after 10 d to provide time for thegraft to recover from any brain death-associated and/or ischemicinsult and to allow examination of the integrity of the ureteralanastomosis. All experiments were conducted in accordance withthe National Institutes of Health Guide for the Care and Useof Laboratory Animals and the Harvard Research Committee.
Model of Brain Death
F344 donor rats were anesthetized and tracheotomized. Braindeath was induced by slow inflation (200 ± 250 µlof saline) of a 3F Fogarty arterial embolectomy catheter (BaxterHealthcare, Irvine, CA) inserted into the subdural space viaan occipital burr hole (3). The balloon was inflated over c.15 min under continuous BP and electroencephalographic monitoring.Herniation of the brain stem was confirmed by flat-line tracings,plus the physical signs of apnea, absent reflexes, and maximallydilated and fixed pupils. The animals were then connected toa rodent respirator (Harvard Rodent Ventilator, model 683, HarvardApparatus, South Natick, MA) and mechanically ventilated ata rate of 100 breaths/min with a tidal volume of 2.0 ml. Intra-arterialBP was monitored continuously with a PE 50 catheter insertedinto the left femoral artery and connected via a transducer(P23ID, Gould, Cleveland, OH) to a BP monitor (Recorder 2200S,Gould). For avoiding peripheral effects of hypotension-associatedischemia, BD rats with a mean arterial BP <80 mmHg sustainedduring the follow-up period were excluded from the study. After6 h, the kidneys were removed and transplanted.
Experimental Groups
Six groups (eight animals/group) were studied. Recipients ofisologous kidneys from living donors (LD) served as syngeneiccontrols (LEWLEW, group 1). LEW animals that received graftsfrom F344 LD (group 2) or BD F344 donors (group 3) served asallogeneic controls.
Several rPSGL-Ig treatment protocols were used in animals bearingkidney allografts from BD donors. In group 4 rats, the leftkidney was perfused slowly (c. 70 to 90 s) in situ with 3 mlof cold Ringers solution 6 h after brain death via the clampedaortic segment, followed by installation of a solution of rPSGL-Ig(5 µg in 0.5 ml of phosphate-buffered saline [PBS]). Anadditional dose (5 µg in 0.3 ml of PBS) was administeredafter occlusion of the renal vein. As the ureter was also clampedto prevent any loss of material via the ureteric vessels, thesolution was retained in the isolated kidney during the timeneeded to prepare the recipient (15 min). This protocol allowedrPSGL-Ig to inhibit P- and E-selectin expressed in the organduring the 6 h after donor brain death and during the briefperiod of cold ischemia (3).
Donors in group 5 received 3 h after induction of brain deathan intravenous injection of 50 µg of rPSGL-Ig in 0.8 mlof PBS to diminish P- and E-selectin upregulation in the inflamedperipheral organs before their removal. In group 6 animals,a similar dose of rPSGL-Ig was injected intravenously into thedonor 3 h after the central injury while a second dose of 50µg was given to the transplant recipient intravenouslyat the time of unclamping the vessels after transplantation.Control kidneys (groups 1 to 3) were perfused with a nonspecificmurine monoclonal antibody (L-6 mAb, 50 µg in cold solution)3 h after induction of brain death (Bristol Myers Squibb, courtesyof Dr. Robert Peach, Princeton, NJ). No control animals receivedrPSGL-Ig. Low-dose cyclosporine (Novartis Pharmaceuticals Corporation,E. Hanover, NJ) was administered to all allograft recipientsin groups 2 to 6 (1.5 mg/kg/d x 10 d, subcutaneously) to inhibitearly host immunologic activity and to allow long-term graftsurvival (13). Group 1 animals did not receive the agent aswe have shown previously that it did not influence isograftbehavior.
rPSGL-Ig
The cell surface glycoprotein adhesion molecule PSGL-1 is expressedby virtually all subsets of leukocytes and recognized by P-,E-, and L-selectins (14–16). The recombinant soluble formused in this study (Genetics Institute/Wyeth Research, Cambridge,MA) consists of the first extracellular 47 amino acids of maturehuman PSGL-1 fused with a human IgG1Fc (17). Two "hinge-proximal"amino acids at positions 234 and 237 within the IgG1Fc portionare mutated to alanine to reduce both complement activationand Fc receptor binding (18). The rPSGL-Ig molecule is secretedas a disulfide bounded dimer with a half-life in rats of c.100 h (19).
Physiologic Studies
Systolic BP was recorded at 2-wk intervals in representativerecipients in groups 1, 2, 3, and 6 (n = 4 to 6 animals/group)using a tail cuff method. For determining functional changesoccurring over time, urine (24 h) was collected every 4 wk fromrecipients in all groups (n = 8/group). Protein excretion wasdetermined by measuring precipitation after interaction with3% sulfosalicylic acid. Turbidity was assessed by absorbanceat a wavelength of 595 nm using a Coleman Junior II Spectrophotometer.Serum creatinines were measured by a modified Jaffe’sreaction on an autoanalyzer (911; Hitachi, Indianapolis, IN).
Histology and Immunohistology
Representative portions of kidney grafts of groups 1, 2, 3,and 6 (n = 4 to 6/group) were fixed in 10% buffered formalinfor histologic examination after 6 h of donor brain death andat 3 and 200 d after transplantation for immunohistology andmolecular biology. Paraffin sections were evaluated using hematoxylinand eosin, periodic acid-Schiff (PAS), trichrome, and elastinstains. Additional pieces were quick frozen and stored at -80°Cfor immunohistology. Cryostat sections were fixed in paraformaldehyde-lysine-periodateto stain cell-surface antigens or in acetone for localizationof cytokines using a peroxidase-antiperoxidase method as described(13). Monoclonal antibodies used for staining were purchasedfrom Serotec (Harlan Bioproducts for Science, Indianapolis,IN), except where noted, and were directed against all rat leukocytes(CD 45, OX-1), T cells (T cell receptor-/, R73), CD4 (W3/25)and CD8 (OX-8), B cells (CD45RA, OX-33), natural killer cells(CD161, 10/78), mononuclear phagocytes (CD68, ED1), and neutrophils(PMN, RP3, F. Sendo, Yamagata, Japan). Cell activation was assessedusing mAb to rat MHC class II antigens (RT1B, OX-3), P-selectin(CD62P; BD PharMingen, San Diego, CA), and E-Selectin (CD62E,BBA-1; British Biotechnology, Cowley, UK). Cytokine expressionwas determined using mAb to rat interleukin-1 (IL-1), IL-2 (1D10),IL-4 (OX-81), IL-10 (A5-4; R&D, Minneapolis, MN), tumornecrosis factor (TNF-; MAB 510, R&D), transforming growthfactor (TGF-; AB-100-NA, R&D), and interferon- (IFN-; D-10,P. van der Meide, Rijswijk, Holland). Control mAb and secondaryantibodies were purchased from BD PharMingen. Isotype-matchedmAb or purified IgG1 and controls for residual endogenous peroxidaseactivity were included in each experiment.
The numbers of labeled cells within 20 consecutive high-powerfields (x40 magnification) were determined in three kidneysper group. Expression of cytokines and chemokines within thesefields is reported on the basis of semiquantitative assessment.
RNAse Protection Assays
RNase protection was performed using the Riboquant Multi-ProbeRNase Protection assay system (BD PharMingen) to analyze a panelof cytokines relevant for initial nonspecific inflammatory eventsoccurring in the organ before (6 h after brain death) and after(3 d) engraftment and those associated with chronic rejection(CR) at 200 d. RNA was isolated from kidney grafts using Trizol.32P-labeled probes were synthesized from the rCK-1 Multi-ProbeTemplate Set (PharMingen) and were hybridized overnight withRNA samples in hybridization buffer according to the manufacturer’sinstructions. Samples were digested with RNase and T1 mixedin RNase buffer, and protected probes were purified and runon a 5% acrylamide gel in 0.5% TBE buffer. Control RNA fromkidney grafts and a dilution of the probe set (serving as sizemarkers) were run in parallel. The gel was absorbed onto filterpaper, dried, and exposed onto Kodak photographic paper at -70°Cfor 24 h. The RNA was analyzed by a phosphoimager using Imagequantsoftware, allowing accurate quantification of mRNA.
Reverse Transcriptase-PCR
mRNA expression of representative adhesion molecules, chemoattractants,and growth factors (intracellular adhesion molecule-1, monocytechemotactic protein-1, and TGF-) that were not measured by theRNAse protection assay were analyzed together with IL-1 andTNF- by reverse transcriptase-PCR (RT-PCR) in renal allograftsfrom 6 h after induction of BD (n = 4 to 6/group) (3,4). Densitiesof competitive mimic and target gene DNA bands were measuredby scanning densitometry using ScanJet 4c (Hewlett Packard,Corvallis, OR) with NIH Image software. The ratios of the densitiesof the respective bands were plotted to establish a linear relationship.Thus, absolute amounts of DNA from unknown samples were calculatedfrom the known amount of the mimic in the starting reaction.Specimens were run in duplicate, and the average value was used.This assay is capable of detecting a twofold difference in targetgene concentration and is as accurate as scintillation countingof radiolabeled PCR products (20). Results were expressed asa ratio to glyceraldehyde-3-phosphate dehydrogenase.
Statistical Analyses
Results are expressed as mean ± SEM. Each of the abovementioned studies was performed in eight rats per group. Comparisonswere performed by analysis of variance or paired and unpairedt test when appropriate. Bonferroni’s correction for multiplecomparisons was used to determine the level of significance.P < 0.05 was considered significant.
Physiologic Changes after BD and Transplantation
With the gradual-onset brain death technique used in these studies,the systolic BP increased sharply over 15 to 25 min from a baselinemean arterial pressure of 98 ± 12 mmHg before injuryto 214 ± 36 mmHg (n = 25, P < 0.0001). This graduallydiminished to normotensive levels (90 to 110 mmHg) during the6-h period before the kidney was removed for transplantation.Electroencephalographic monitoring showed flat-line tracingsin BD animals versus physiologic activity in the controls.
All recipients survived during the 200 d of observation. SystolicBP was assessed serially in representative animals throughoutthe follow-up period. Recipients of isografts in group 1 remainednormotensive (110 ± 12 mmHg) with levels similar to thoseof ungrafted animals (n = 9). Recipients of LD allografts (group2) varied in systolic BP between normal and c. 120 mmHg. Thatof rats bearing BD donor allografts, in contrast, increasedprogressively to 160 ± 15 mmHg by 20 wk (P < 0.001).Treated animals in group 6 were normotensive throughout thetime of observation.
Renal function varied between the groups. Isografted animals(group 1) never manifested renal dysfunction (Figure 1). Proteinuriaincreased progressively in group 2 rats bearing chronicallyrejecting LD grafts after c. 12 wk. It became manifest earlierand reached higher levels in group 3 hosts with grafts fromBD donors. Urinary protein loss in recipients of BD donor kidneysperfused in situ with rPSGL-Ig (group 4) or from donors treatedintravenously (group 5) was decreased and delayed in onset (P< 0.0001). When both donor and recipient were treated withrPSGL-Ig (group 6), renal function remained at baseline throughoutthe follow-up period.
Figure 1. The development of proteinuria in the different groups. Proteinuria was significantly reduced in all treatment groups: --------, syngeneic control, group 1; , allogeneic living donor (LD) control, group 2; -X- allogeneic brain-dead (BD) donor control, group 3; ——, allogeneic BD donor + P-selectin glycoprotein ligand (rPSGL-Ig) perfused into the graft in situ 6 h after induction of brain death, group 4; -*- rPSGL-Ig injected intravenously into the allogeneic donor 3 h after induction of brain death, group 5; , rPSGL-Ig injected intravenously into both BD donor and allograft recipient, group 6; n = 8 animals/group, *groups 2 and 3 versus groups 4 to 6, P < 0.0001.
As an additional assessment of graft function, levels of serumcreatinine were monitored serially after transplantation inall groups (Figure 2). Creatinine was significantly increasedin animals with kidneys from BD donors in group 3 at 200 d atlevels consistently higher than in group 2 recipients of LDgrafts (P < 0.0001). No increase in plasma creatinine levelsoccurred in treated animals in groups 4 to 6 over time, comparableto the syngeneic controls.
Figure 2. Serial serum creatinine levels show comparable patterns to those of proteinuria in the various groups. *Groups 2 and 3 versus groups 4 to 6, P < 0.0001; **group 3 versus group 2, P < 0.001.
Histology
Except for minor tubular swelling, nongrafted kidneys in theBD animals seemed morphologically normal 6 h after the centralinjury. Isologous (group 1) and allogeneic LD (group 2) kidneyswere also unchanged 3 d after transplantation. In contrast,allografts from BD donors (group 3) already showed severe tubularnecrosis with widespread mixed perivascular PMN and mononuclearcell infiltration. The kidneys of group 6 recipients showedonly focal tubular necrosis at the corticomedullary junctionand rare polymorphonuclear leukocyte (PMN) (Figure 3).
Figure 3. Histology of rat renal transplants (group 6) showing the effects of rPSGL-Ig treatment early and late after transplantation. (a and b) Periodic acid-Schiff-stained sections from day 3 allografts. Subsequent panels (c through f) show trichrome-stained sections of grafts harvested at day 200. Whereas allografts from group 3 at day 3 show widespread tubular vacuolization and sloughing in conjunction with a marked neutrophil-rich infiltrate (a), rPSGL-Ig therapy (group 6) led to minimal tubular injury and only minor focal mononuclear cell infiltration (b). At day 200, isografts (group 1) show some glomerular hypertrophy but are otherwise well preserved (c); allografts from LD (group 2) and much more intensified in those from BD donors (group 3) show advanced injury (d and e). In contrast, when both BD donors and recipients were treated with rPSGL-Ig (group 6), the allografts show almost normal architecture comparable to that of isograft controls. Sections are representative of four to six grafts/group/time point. Magnification, x300.
By 200 d after transplantation, group 2 kidneys showed progressivechanges of CR, with widespread (>50%) glomerulosclerosis,obliteration of capillary loops, increased mesangial matrix,and marked mesangial cell proliferation (Figure 3). Tubularatrophy, fibrosis, and focal interstitial mononuclear cell infiltrationwere obvious. Kidney grafts from BD donor group 3 animals wereend stage, with severe glomerulosclerosis (>90%), interstitialfibrosis, tubular atrophy and dilation, and dense cellular infiltration.In contrast, allografts in rPSGL-Ig-treated animals in group6 resembled isografts (group 1), with <10% exhibiting minimalchanges. These grafts were essentially normal.
Immunohistology
No immunohistologic changes were noted in any graft before transplantation,regardless of donor group. Infiltration of leukocyte populationswas minimal after 3 d in kidneys of groups 1 and 2. Allograftsin group 3 were infiltrated by relatively large numbers of PMN(c. 50%) and macrophages (c. 40%) (Table 1). Few lymphocyteswere present. Cell populations infiltrating allografts of group6 were sparse. Similarly, lymphocyte- and macrophage-associatedcytokines were highly expressed in group 3 kidneys but absentin those of group 6 animals.
Table 1. Immunohistologic results of kidney allografts at day 3 after transplantationa
At 200 d, patterns of infiltration of leukocyte populationsand their products generally mirrored the morphologic changes(Table 2). Cellular infiltration of group 1 isografts was minimal,and cell products were not expressed. The chronically rejectingLD allografts in group 2 showed a preponderance of macrophages(>75%) combined with c. 10% of each CD4+ and CD8+ T lymphocytes.PMN were absent. Interstitial and mononuclear cell stainingfor representative cytokines was moderate. Grafts from BD donorsin group 3 showed increased numbers of macrophages and T cellsversus group 2. Staining of interstitium and glomeruli withTGF- was more intense in group 3 kidneys than in the other groups;>50% leukocytes were positive for TNF-. In contrast, thegrafts of rPSGL-Ig-treated animals (group 6) remained relativelybland, with minimal cell infiltration and features of immuneactivation resembling those in isografts.
Table 2. Immunohistologic results of kidney allografts at day 200 after transplantationa
Molecular Changes
As a corollary to the above findings, the kidneys in all animalgroups were examined at intervals by RNAse protection assayand semiquantitative RT-PCR for mRNA levels of a representativeseries of lymphocyte- and macrophage-associated inflammatorymediators. Gene expression of all factors examined in the kidneysof untreated BD donor controls (group 3), except IL-1, wereupregulated significantly 6 h after the central injury and beforetransplantation (P < 0.0001), whereas levels in treated donorsin group 6 remained at baseline, comparable to those from naiveanimals (Figures 4 and 5a). These cytokines remained upregulatedin kidneys of group 3 at 3 d but were only minimally expressedin group 6 kidneys (Figure 5b). Those of groups 4 and 5 werealso at baseline (data not shown).
Figure 4. mRNA expression of representative cytokines 6 h after induction of BD (group 3) was highly upregulated versus baseline values. Values of the rPSGL-Ig-treated group (group 6) were comparable to those of naive controls. Data are expressed as cytokine/glyceraldehyde-3-phosphate dehydrogenase ratio (n = 4 animals/group; *P < 0.0001 for each cytokine).
Figure 5. Gene expression of a panel of factors assessed by RNAse protection assay was elevated in kidneys from BD donors (group 3) 6 h (A) and at 3 days after transplantation (B) after the central injury. These were essentially negative when the donor had been treated with rPSGL-Ig (group 6) (n = 4 animals/group, normalized optic density ratio; group 3 versus group 6, *P < 0.0001 for each cytokine).
The activity of cytokines at 200 d was clearly associated withthe severity of the CR process as assessed morphologically (Figure 6).End-stage BD donor kidneys (group 3) expressed predominantlyIL-1, IL-2, TNF-, and IL-4. In contrast, cytokines in kidneysfrom animals treated with rPSGL-Ig and without structural changes(group 6) resembled those of syngeneic controls. Expressionin the other treated groups (groups 4 and 5) was in the samerange (IL-1 and IL-5), minimally increased compared with kidneysof group 6 animals (IL-2 and IL-4), or upregulated marginally(TNF-). This indicates that initial treatment of both the donorand the recipient, in particular, prevents significant expressionof cytokines at a late period when the controls are experiencingovert changes of CR.
Figure 6. Kidneys of groups 2 and 3 show high expression of interleukin (IL)-2, tumor necrosis factor , and IL-4 200 d after transplantation, assessed by RNAse protection assay. The treatment groups (groups 4 to -6) resemble isografts (n = 4 animals/group, normalized optic density ratio; group 3 versus 6, *P < 0.0001).
Nonspecific peripheral injuries that occur secondary to donorbrain death as well as those that surround organ removal, perfusion,and storage may initiate early renal dysfunction after transplantation.The associated inflammatory response, in turn, may trigger andamplify acute host immunologic activity against the graft (4).This observation would explain the apparent synergy noted clinicallybetween the effects of delayed renal graft function and acuterejection episodes (1). At some later stage after resolutionof the acute insults, the progression of chronic changes developingwithin a compromised allograft may become autonomous from anyhost-mediated processes, leading to its increasing deteriorationand failure over time (21). Appreciation of the influence ofthese initial events on the quality of organs accepted for transplantationhas stimulated examination of novel therapeutic approaches designedto normalize the affected tissue even before its placement inthe recipient.
The relationship between donor brain death and inflammatorychanges in peripheral organs is not fully defined. The autonomicstorm produced from the central insult evokes chaotic changesin BP, with transient hypertension often followed by hypotension(5). As blood flow through a given tissue may not be uniformafter brain death, the pattern of reperfusion may be differentto that after the global ischemia associated with transplantation(9). The dynamics of the responsible catecholamines, well documentedin model using large animals and in rats, may produce intenseperipheral vasoconstriction and ischemia, with reduced redistributionof blood flow perfusion despite highly increased perfusion pressures(5,9). As vasodilation may then occur with decline in catecholaminelevels and hypotension, low perfusion pressures may reduce regionalblood flow further. This pattern seems different from the "lowreflow phenomenon" in ischemic organs reperfused after transplantation,in which capillary swelling and platelet aggregation may impedeblood flow despite persistent normotension. Resultant oxidativestresses may contribute to early organ dysfunction, as studiedin models of ischemia but not after brain death (22–24).
Important endocrine changes that follow the central injury mayalso produce additional changes in peripheral organs that donot occur in pure ischemic models. As only 38% to 87% of cadaverorgan donors, for instance, develop diabetes insipidus fromfailure of vasopressin release from the hypothalamus and posteriorpituitary gland, it is thought that viable cells in these structuresmay continue to function as long as 2 wk after death (25). Similarly,cells preserved in the periphery of the anterior pituitary,despite extensive central necrosis, may produce thyroid-stimulatinghormone, prolactin, growth hormone, and other factors, as detectedin BD patients 1 wk later (6,26). These substances have beenimplicated in metabolic injury to peripheral organs, althoughthe subject still remains controversial.
Inflammatory factors are released into the circulation aftera central insult or brain death. Clinically, expression of transcriptionallevels of Il-1 and TNF has been associated with episodes offocal cerebral ischemia, whereas Il-6 has been identified inthe serum of BD patients (6,27). TNF-, IL-1, IL-2, IL-6, andIL-10 have been found in high levels in the circulation of brain-injuredand BD rats (3,28). Among the earliest events that develop afterbrain injury, as in other insults, is expression of a seriesof adhesion molecules in sequence by activated vascular endothelium(3,14,15). Upregulation of selectins mediates the adhesion ofplatelets, PMN, monocytes, and some lymphocytes to the vascularwall (29). The expression of P-selectin during the transplantationevent is associated primarily with platelets, although immunohistologicassessment of these formed elements was not carried out in thepresent studies (30).
rPSGL-Ig, a recombinant soluble form of PSGL-1, inhibits selectin-mediatedadhesion events through competitive binding and prevents theinteraction between vascular endothelial cells, platelets, andPMN (16,17). Perfusion of an ischemic kidney with the agentdramatically reduces later acute cellular and molecular eventsassociated with reperfusion, preventing resultant chronic changesover time (12). As brain death accelerates the inflammatoryprocess, the timing of administration of rPSGL-Ig in these studieswas designed to be relevant to the clinical situation by providingan adequate concentration of the blocking ligand in the kidneyduring the period of injury surrounding transplantation. A lowdose of rPSGL-Ig was added directly to the cold renal perfusate(group 4) or systemically to the donor 3 h after induction ofbrain death (group 5) to bind any P-selectin upregulated onthe renal endothelium after brain death and during cold ischemia.This strategy was chosen as it had been shown previously thatselectins are upregulated rapidly and that associated migrationof PMN into the injured tissues begins while the organ is stillin the host, well ahead of comparable cellular activity in LDgrafts (3,4). As noted again in the present experiments, cellsurface molecules and inflammatory mediators, already upregulatedin the tissues 6 h after brain death, increase progressivelywith reperfusion and remain elevated for several days thereafter.Thus, in group 6 animals, an additional dose of rPSGL-Ig wasgiven to recipients with revascularization of the transplantto extend the blockade of selectins during the initial daysafter reperfusion. As pharmacokinetic studies have determinedthat the half-life of rPSGL-Ig in normal rats is 100 h, thisprotocol also provided an adequate concentration of the blockingligand during the first days after engraftment, during whichtime much activity of host leukocytes and their products isongoing (19). Subsequent administration of the material didnot seem appropriate as most of the selectin activity has ceased.
The chronic process was manifest primarily in BD donor allograftsby progressive fibrosis, glomerulosclerosis, and tissue remodelingassociated with the presence of macrophages and their products(31). As also noted previously, this late activity was presentbut significantly less intense in chronically rejecting livingdonor (LD) controls (13). The extent of proteinuria, developingafter c. 8 wk if the donor was brain dead (BD) and after c.12 wk if an LD was used, was commensurate with morphologic changes,although it should also be noted that the hypertension thatdeveloped in recipients of BD donor kidneys may have contributedto the increased protein loss in this group. In contrast, itseems clear that initial selectin blockade during the eventssurrounding transplantation inhibited or prevented the subsequentprogression of CR as neither leukocyte infiltration nor activationoccurred after treatment with rPSGL-Ig.
As a continuum between early nonspecific, donor-associated injuriesand alloantigen-specific events within an allograft may limitboth the numbers of donors accepted and diminish the later resultsof transplantation, the importance of using a variety of noveltherapeutic strategies to normalize potentially inflamed kidneysis obvious. Administration of steroids has been shown to increasesurvival of grafts from BD donors after prolonged ischemia (32).Using the same rat model as in the present studies, treatmentof the BD donor with prednisolone reduced cellular infiltrationand the transcription of various leukocyte and endothelial cell-derivedcytokines and other mediators. Within 24 h, for instance, upregulationof Fas ligand and perforin was decreased (32). The inductionof protective genes including heme-oxygenase-1 have gained recentinterest, although mechanisms of inflammation and protectionof the endothelium remain elusive (33). Heme-oxygenase-1 inductionprolonged the survival of heart grafts in mice and was associatedwith the absence of chronic graft rejection and long-term graftacceptance (34). CD28/B7 blockade of T cell-associated eventswith CTLA4Ig is effective in reducing the effects of cold ischemia;it may also influence the sequelae of brain death (35). In addition,new immunosuppressive drugs such as rapamycin may reduce earlyinflammatory changes in transplanted organs. As more is learnedabout the complex sequelae of brain death injury, which seemmore complex than those of ischemia alone, selective treatmentof the organ, the donor, and/or the recipient may be given.Inhibition of one of the earliest steps, the upregulation ofselectins, as described in these experiments, may be one suchstrategy.
Acknowledgments
This work was supported by United States Public Health ServiceGrants 5RO1 1 DK 46190-28 and 1PO1 AI40152-05.
Cecka JM, Shoskes DA, Gjertson DW: Clinical impact of delayed graft function for kidney transplantation. Transplant Rev 15: 57–67, 2001[CrossRef]
Terasaki PI, Cecka JM, Gjertson DW, Takemoto S: High survival rates of kidney transplants from spousal and living unrelated donors. N Engl J Med 333: 333–336, 1995[Abstract/Free Full Text]
Takada M, Nadeau KC, Hancock WW, Mackenzie HS, Shaw GD, Waaga AM, Chandraker A, Sayegh MH, Tilney NL: Effects of explosive brain death on cytokine activation of peripheral organs in the rat. Transplantation 65: 1533–1542, 1998[CrossRef][Medline]
Pratschke J, Wilhelm MJ, Kusaka M, Beato F, Milford EL, Hancock WW, Tilney NL: Accelerated rejection of renal allografts from brain-dead donors. Ann Surg 232: 263–271, 2000[CrossRef][Medline]
Shivalkar B, Van Loon J, Wieland W, Tjandra-Maga TB, Borgers M, Plets C, Flameng W: Variable effects of explosive or gradual increase of intracranial pressure on myocardial structure and function. Circulation 87: 230–239, 1993[Abstract/Free Full Text]
Amado JA, Lopez-Espadas F, Vazquez-Barquero A, Salas E, Riancho JA, Lopez-Cordovilla JJ, Garcia-Unzueta MT: Blood levels of cytokines in brain-dead patients: Relationship with circulating hormones and acute-phase reactants. Metabolism 44: 812–816, 1995[CrossRef][Medline]
Harms J, Isemer FE, Kolenda H: Hormonal alteration and pituitary function during course of brain-stem death in potential organ donors. Transplant Proc 23: 2614–2616, 1991[Medline]
Okamoto S, Corso CO, Nolte D, Rascher W, Thiery J, Yamaoka Y, Messmer K: Impact of brain death on hormonal homeostasis and hepatic microcirculation of transplant organ donors. Transpl Int 11 (Suppl 1): S404–S407, 1998
Herijgers P, Leunens V, Tjandra-Maga TB, Mubagwa K, Flameng W: Changes in organ perfusion after brain death in the rat and its relation to circulating catecholamines. Transplantation 62: 330–335, 1996[CrossRef][Medline]
Vestweber D, Blanks JE: Mechanisms that regulate the function of the selectins and their ligands. Physiol Rev 79: 181–213, 1999[Abstract/Free Full Text]
Springer TA: Traffic signals for lymphocyte recirculation and leukocyte emigration: The multistep paradigm. Cell 76: 301–314, 1994[CrossRef][Medline]
Takada M, Nadeau KC, Shaw GD, Marquette KA, Tilney NL: The cytokine-adhesion molecule cascade in ischemia/reperfusion injury of the rat kidney. Inhibition by a soluble P-selectin ligand. J Clin Invest 99: 2682–2690, 1997[Medline]
Hancock WH, Whitley WD, Tullius SG, Heemann UW, Wasowska B, Baldwin WM3rd, Tilney NL: Cytokines, adhesion molecules, and the pathogenesis of chronic rejection of rat renal allografts. Transplantation 56: 643–650, 1993[Medline]
Somers WS, Tang J, Shaw GD, Camphausen RT: Insights into the molecular basis of leukocyte tethering and rolling revealed by structures of P- and E-selectin bound to SLe(X) and PSGL-1. Cell 103: 467–479, 2000[CrossRef][Medline]
McEver RP, Cummings RD: Perspectives series: Cell adhesion in vascular biology. Role of PSGL-1 binding to selectins in leukocyte recruitment. J Clin Invest 100: 485–491, 1997[Medline]
Yang J, Furie BC, Furie B: The biology of P-selectin glycoprotein ligand-1: Its role as a selectin counterreceptor in leukocyte-endothelial and leukocyte-platelet interaction. Thromb Haemost 81: 1–7, 1999[Medline]
Kumar A, Villani MP, Patel UK, Keith JC Jr, Schaub RG: Recombinant soluble form of PSGL-1 accelerates thrombolysis and prevents reocclusion in a porcine model. Circulation 99: 1363–1369, 1999[Abstract/Free Full Text]
Morgan A, Jones ND, Nesbitt AM, Chaplin L, Bodmer MW, Emtage JS: The N-terminal end of the CH2 domain of chimeric human IgG1 anti-HLA-DR is necessary for C1q, Fc gamma RI and Fc gamma RIII binding. Immunology 86: 319–324, 1995[Medline]
Khor SP, McCarthy K, DuPont M, Murray K, Timony G: Pharmacokinetics, pharmacodynamics, allometry, and dose selection of rPSGL-Ig for phase I trial. J Pharmacol Exp Ther 293: 618–624, 2000[Abstract/Free Full Text]
Kato S, Luyckx VA, Ots M, Lee KW, Ziai F, Troy JL, Brenner BM, MacKenzie HS: Renin-angiotensin blockade lowers MCP-1 expression in diabetic rats. Kidney Int 56: 1037–1048, 1999[CrossRef][Medline]
Tullius SG, Hancock WW, Heemann U, Azuma H, Tilney NL: Reversibility of chronic renal allograft rejection. Critical effect of time after transplantation suggests both host immune dependent and independent phases of progressive injury. Transplantation 58: 93–99, 1994[Medline]
Walker LM, York JL, Imam SZ, Ali SF, Muldrew KL, Mayeux PR: Oxidative stress and reactive nitrogen species generation during renal ischemia. Toxicol Sci 63: 143–148, 2001[Abstract/Free Full Text]
Amersi F, Dulkanchainun T, Nelson SK, Farmer DG, Kato H, Zaky J, Melinek J, Shaw GD, Kupiec-Weglinski JW, Horwitz LD, et al.: A novel iron chelator in combination with a P-selectin antagonist prevents ischemia/reperfusion injury in a rat liver model. Transplantation 71: 112–118, 2001[Medline]
Yoshimura T, Kawashima M, Nakamura T, Isowa N, Bando T, Hasegawa S, Kondo H, Toyokuni S, Wada H: A novel selectin blocker alleviates oxidative stress of lung reperfusion injury. J Surg Res 101: 91–98, 2001[CrossRef][Medline]
Arita K, Uozumi T, Oki S, Kurisu K, Ohtani M, Mikami T: The function of the hypothalamo-pituitary axis in brain dead patients. Acta Neurochir 123 (1–2): 64–75, 1993[CrossRef][Medline]
Sugimoto T, Sakano T, Kinoshita Y, Masui M, Yoshioka T: Morphological and functional alterations of the hypothalamic-pituitary system in brain death with long-term bodily living. Acta Neurochir 115 (1–2): 31–36, 1992[CrossRef][Medline]
Fassbender K, Rossol S, Kammer T, Daffertshofer M, Wirth S, Dollman M, Hennerici M: Proinflammatory cytokines in serum of patients with acute cerebral ischemia: Kinetics of secretion and relation to the extent of brain damage and outcome of disease. J Neurol Sci 122: 135–139, 1994[CrossRef][Medline]
Zhai QH, Futrell N, Chen FJ: Gene expression of IL-10 in relationship to TNF-alpha, IL-1beta and IL- 2 in the rat brain following middle cerebral artery occlusion. J Neurol Sci 152: 119–124, 1997[CrossRef][Medline]
Diacovo TG, Puri KD, Warnock RA, Springer TA, von Andrian UH: Platelet-mediated lymphocyte delivery to high endothelial venules. Science 273: 252–255, 1996[Abstract]
Frenette PS, Denis CV, Weiss L, Jurk K, Subbarao S, Kehrel B, Hartwig JH, Vestweber D, Wagner DD: P-Selectin glycoprotein ligand 1 (PSGL-1) is expressed on platelets and can mediate platelet-endothelial interactions in vivo. J Exp Med 191: 1413–1422, 2000[Abstract/Free Full Text]
Pratschke J, Wilhelm MJ, Laskowski I, Kusaka M, Beato F, Tullius SG, Neuhaus P, Hancock WW, Tilney NL: Influence of donor brain death on chronic rejection of renal transplants in rats. J Am Soc Nephrol 12: 2474–2481, 2001[Abstract/Free Full Text]
Pratschke J, Kofla G, Wilhelm MJ, Vergopoulos A, Laskowski I, Shaw GD, Tullius SG, Volk HD, Neuhaus P, Tilney NL: Improvements in early behavior of rat kidney allografts after treatment of the brain-dead donor. Ann Surg 234: 732–740, 2001[CrossRef][Medline]
Platt JL, Nath KA: Heme oxygenase: Protective gene or Trojan horse. Nat Med 4: 1364–1365, 1998[CrossRef][Medline]
Hancock WW, Buelow R, Sayegh MH, Turka LA: Antibody-induced transplant arteriosclerosis is prevented by graft expression of anti-oxidant and anti-apoptotic genes. Nat Med 4: 1392–1396, 1998[CrossRef][Medline]
Chandraker A, Takada M, Nadeau KC, Peach R, Tilney NL, Sayegh MH: CD28-b7 blockade in organ dysfunction secondary to cold ischemia/reperfusion injury. Kidney Int 52: 1678–1684, 1997[Medline]
Received for publication October 2, 2001.
Accepted for publication April 3, 2002.
,
Top
Abstract
Introduction
LEW rat model of chronic rejection was used to examine (1) whether the initial inflammatory events that develop within kidney allografts from brain-dead donors could be normalized using a recombinant soluble form of P-selectin glycoprotein ligand and (2) whether amelioration of these early changes would alter the inexorable progression of chronic allograft rejection. Untreated living donor controls experienced unrelenting chronic rejection over time. This complex process was accelerated in brain-dead donor kidneys. Treatment with P-selectin glycoprotein ligand prevented the early inflammatory changes in the transplanted organs and their subsequent (200 d) functional and morphologic manifestations, particularly when the soluble ligand was administered both to the donor before organ removal and to the recipient after engraftment. This strategy of using a naturally occurring selectin ligand to prevent donor-associated chronic graft dysfunction may be of special clinical interest in cadaver donor transplantation. 
/
, R73), CD4 (W3/25) and CD8 (OX-8), B cells (CD45RA, OX-33), natural killer cells (CD161, 10/78), mononuclear phagocytes (CD68, ED1), and neutrophils (PMN, RP3, F. Sendo, Yamagata, Japan). Cell activation was assessed using mAb to rat MHC class II antigens (RT1B, OX-3), P-selectin (CD62P; BD PharMingen, San Diego, CA), and E-Selectin (CD62E, BBA-1; British Biotechnology, Cowley, UK). Cytokine expression was determined using mAb to rat interleukin-1
(IFN-
----, syngeneic control, group 1; 
, allogeneic living donor (LD) control, group 2; -X- allogeneic brain-dead (BD) donor control, group 3; —
—, allogeneic BD donor + P-selectin glycoprotein ligand (rPSGL-Ig) perfused into the graft in situ 6 h after induction of brain death, group 4; -*- rPSGL-Ig injected intravenously into the allogeneic donor 3 h after induction of brain death, group 5; 
, rPSGL-Ig injected intravenously into both BD donor and allograft recipient, group 6; n = 8 animals/group, *groups 2 and 3 versus groups 4 to 6, P < 0.0001.
