INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Ischemia-reperfusion (IR) injury occurs in the context of lung transplantation when blood supply is reintroduced to the ischemic graft at the completion of the implantation procedure. IR injury results in life-threatening graft dysfunction in 20% of lung transplant patients, and contributes significantly to the first-year postoperative mortality rate of 15% in this population (1, 2). Clinically, IR injury appears as pulmonary edema and acute respiratory distress syndrome (3). Consequently, patients are subjected to prolonged intubation, ventilator-associated infections, longer stays in the intensive care unit (ICU), and early postoperative mortality (6, 7).

The pathophysiology of IR injury in the lung after transplantation involves alveolar-capillary barrier leakage, neutrophil migration, interstitial and alveolar edema, tissue inflammation, and cell injury and death (8, 9). Alteration of nitric oxide synthase (NOS) activity, oxygen free-radical production, lipid peroxidation, and proinflammatory cytokine release (tumor necrosis factor- [TNF]-, interferon- [IFN]-, and interleukin- [IL]-1) probably mediates IR injury (10).

The focus of the present study was on the mode of cell death incurred by the transplanted lung in the setting of IR injury. Clinical and experimental transplantation studies have examined the types of cell death that occur after IR injury in the liver, kidney, and intestine, and have demonstrated various degrees of cell death (14). All of these studies showed a significant correlation between graft reperfusion injury and increasing numbers of apoptotic or necrotic cells in grafts. In a study of 20 transplanted human lung grafts, we studied the induction of apoptosis before, during, and after transplantation (17). We found that apoptotic cells were absent from lungs after various periods of cold preservation (1 to 5 h) and warm ischemia. However, at 60 min after reperfusion, 21% of cells were apoptotic, and at 120 min after reperfusion, 34% were apoptotic. The human study did not allow us to examine the effect of a number of variables that are probably critical in this setting. In the present study we therefore used a rat single-lung transplant model to examine apoptotic and necrotic cell death after IR injury.

With this model we were able to subject the donor lung to varying ischemic times, and could accurately assess graft function in a controlled setting without the usual confounding variables present in human studies. We hypothesized that an increased ischemic period would lead to an increase in the number of apoptotic and necrotic cells. We further hypothesized that the degree and type of cell death would correlate with lung function. Our objectives were: (1) to develop a technique to quantify the cells undergoing apoptotic versus necrotic cell death in the same specimen; (2) to determine the effect of ischemic time on the mode of cell death after cold preservation and after reperfusion; and (3) to determine the relationship between apoptotic and necrotic cell death and posttransplant lung function. Our data provide insight into the relationships among ischemia, reperfusion, and cell death, and suggest novel approaches to treating IR injury in the context of lung transplantation.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Animals

Experiments were performed with male inbred (250 to 350 g) Lewis rats (Charles River Inc., Montreal, PQ, Canada). All animals received care in compliance with the Principles of Laboratory Animal Care formulated by the National Society for Medical Research, the Guide for the Care and Use of Laboratory Animals (18), and the Guide to the Care and Use of Experimental Animals formulated by the Canadian Council on Animal Care. The experimental protocol was approved by the Animal Care Committee of the Toronto General Hospital Research Institute.

Lung Transplantation Procedure

Harvest and storage. We used a rat single-lung transplant model. Donor rats were anesthetized with an intraperitoneal injection of 1 ml of sodium pentobarbital (Somnotol; MTC Pharmaceuticals, Cambridge, ON, Canada) and intubated through a tracheostomy with a 14-gauge intravenous catheter. The tracheostomy tube was then connected to a volume-controlled ventilator (Rodent Ventilator Model 683; Harvard Instruments, South Natick, MA) and the animals were ventilated at a rate of 70 breaths/min, a tidal volume (VT) of 10 ml/kg, an inspired oxygen fraction (FI_O2 of 1.0, and a positive end-expiratory pressure (PEEP) of 2 cm H₂O. Following this, a median laparosternotomy was performed and 300 U of heparin [USP] (Hepalean; Organon Teknika, Toronto, ON) was injected into the inferior vena cava (IVC). After a period of 5 min, 0.5 ml of arterial blood was taken from the abdominal aorta for blood gas analysis. For retrieval of the heart-lung block, the IVC was incised, the left atrial appendage was excised, and a 14-gauge intravenous catheter was placed into the main pulmonary artery (PA) through an anterior incision in the right ventricular outflow tract. The lungs were then flushed through this catheter with 20 ml of low-potassium dextran glucose (LPDG) preservation solution (Perfadex; Biophausia, Uppsala, Sweden) containing 500 µg/L of prostaglandin (PG) E₁ (Prostin VR; Upjohn, Don Mills, ON, Canada) infused from a height of 30 cm. This was the same protocol as used in our clinical and experimental lung transplant program (19, 20). Immediately after flushing of the lungs, the tracheostomy tube was clamped following inspiration, to preserve the lungs in the inflated state. The heart-lung block was then removed and placed in iced LPDG at 4° C. The left lung was prepared for transplantation with the placement of three 14-gauge cuffs into the left PA, left pulmonary vein (PV), and left main bronchus (MB), respectively. Both the right and left lungs were placed into 40 ml of LPDG at 4° C for each of the specific storage periods.

Transplantation. Recipient animals were anesthetized and a tracheostomy was performed as described for the donor animals. The recipient animals were ventilated with a gas mixture of 75% O₂ and 25% room air at a rate of 70 breaths/min and a VT of 10 ml/kg. A left thoracotomy was performed through the fifth intercostal space. The left lung was mobilized by dividing the pulmonary ligament. The hilar structures were then dissected free. The left PA, PV, and MB were identified and clamped with microsurgical aneurysm clamps. A ventral incision was made in each of these structures. The cuffs on the donor lung structures were placed into the corresponding recipient structures through each incision. The anastomoses were secured with No. 7.0 polypropylene ties. The implantation time or warm ischemia time was standardized at 20 min; this was the period when the deflated donor lung was in the chest at 37° C and the anastomotic procedure was underway. After the 20-min warm ischemia time, the transplanted lung was reinflated after removal of the MB clamp, and blood was reintroduced by unclamping first the PV and then the PA. The animal was ventilated with an FI_O2 of 1.0 at a rate of 70 breaths/min, VT of 10 ml/kg, and PEEP of 2 cm H₂O during the 2-h reperfusion period. At 1 min into the reperfusion period, each animal received 1 ml of 0.9% normal saline intraperitoneally for volume replacement. Heparin was not given to the recipient animals, since it is not routinely administered to lung transplant patients prior to graft reperfusion. Each animal was covered during the reperfusion period to prevent hypothermia. Two hours after graft reperfusion, 0.5 ml of blood was taken from the left PV of the transplanted lung, at a point distal to the cuff anastomosis, for blood gas analysis.

Tissue Treatment

All study lungs were flushed for 5 min with 20 ml of a 500 µM trypan blue (Sigma Chemical Co., St. Louis, MO) solution through the main PA, and then with 20 ml of 0.9% normal saline and 10 ml of 4% paraformaldehyde. Trypan blue was dissolved in Krebs-Henseleit buffer (pH 7.4) (Sigma Chemical Co.). The lungs were then fixed in 10% formalin. The middle thirds of the left and right lungs were used for histologic examination, since these sectors were representative of peripheral and central parenchymal areas.

Histologic Evaluation and Viability Assessment: Triple Staining Technique

The formalin-fixed lung tissues were embedded in paraffin and cut into 4-µm-thick tissue slices. These were mounted on saline-treated glass slides for histologic asessement.

Detection of apoptosis through in situ terminal deoxynucleotidyl transferase (TdT)-mediated deoxyuridine triphosphate nick end-labeling (TUNEL) was undertaken, using the ApopTag Kit (Oncor, Gaithersburg, MD) according to the manufacturer's instructions. The TUNEL method is based on the enzymatic ability of TdT to catalyze a template-independent addition of deoxyribonucleotide triphosphate to the 3'-OH ends of double- or single-stranded DNA. Sections were deparaffinized and rehydrated. Protein digestion was done by applying proteinase K (20 µg/ml) to the slides for 15 min at room temperature (RT), which was followed by four washes in distilled water for 2 min each. Equilibration buffer was applied to the sections, which were then incubated in a humidified chamber for 3 min. Following this the sections were incubated with TdT in a humidified chamber at 37° C for 1 h. Fluorescein-labeled antidigoxigenin antibody was applied to the sections, and they were then incubated in a humidified chamber for 30 min at RT. The sections were then washed with phosphate-buffered saline, and an antifade preparation containing propidium iodide (PI) (Oncor) was applied for nuclear staining.

TUNEL-stained tissue sections were examined with fluorescent microscopy. First, the PI staining (red) was examined through a 520-nm filter at a magnification of ×100. PI stains all nucleated cells (alive, necrotic, and apoptotic) in the same manner. The magnification was then increased to ×400 and a color photomicrograph was taken. The same area as was viewed for the PI staining was then similarly examined for apoptotic staining (bright green), using a 590 nm filter at a magnification of ×400. After a color slide photograph was taken through this filter system, the filters and the UV light were shut down and a third picture of the same area was taken with standard light microscopy, to detect the trypan blue stain (dead cells). Two randomly chosen areas from each slide were examined with this triple staining method, and are shown in Figures 1 and 3.

View larger version (115K): [in this window] [in a new window]   Figure 1.   Cell death in lungs after cold ischemic preservation. Triple staining combining PI, TUNEL, and trypan blue staining. Representative images of tissue specimens following 20 min and 6, 12, 18, and 24 h of cold ischemic preservation. With the triple staining technique, no apoptotic (TUNEL-positive) cells, which would have appeared as bright green cells, were seen at the end of cold ischemic preservation in any group. The 18-h and 24-h groups, however, showed significant cellular necrosis (original magnification: ×400). PI = propidium iodide, TUNEL = terminal deoxyuridine triphosphate nick end-labeling technique, CIT = cold ischemia time.

View larger version (101K): [in this window] [in a new window]   Figure 3.   Cell death in lungs after transplantation and reperfusion. Triple staining combining PI, TUNEL, and trypan blue staining. With the triple staining technique, approximately 30% of cells were apoptotic after 6 h and 12 h of preservation followed by 2 h of reperfusion, and almost no cells were necrotic. However, after 18 h and 24 h of cold ischemic preservation and 2 h of graft reperfusion, less than 2% of cells were apoptotic and about 30% were necrotic. Apoptotic cells stained positively with both the TUNEL (white arrows) and the trypan blue (black arrows) techniques, whereas necrotic cells stained positively only with the trypan blue (black arrowheads) and not with the TUNEL (white arrowheads) technique (original magnification: ×400). PI = propidium iodide, TUNEL = terminal deoxyuridine triphosphate nick end-labeling technique, CIT = cold ischemia time.

Cell counts were done by two researchers in a blinded fashion. To count cells, the slides were projected onto a grid comprising 12 fields. Six randomly chosen fields were used. These six fields were used for all study slide counts. PI-stained cells were counted first, followed by TUNEL-positive cells and finally by trypan blue-stained cells. Only cells that could clearly be identified as individual cells were counted as cells. If they did not fulfil this criterion they were considered as background staining.

The PI-stained cell count represents the total number of cells (total cells: alive + necrotic + apoptotic cell.) The TUNEL-positive cell count represents the number of apoptotic cells. The trypan blue-stained cell count identifies the number of dead cells (necrotic + apoptotic cells). Therefore, the trypan blue-stained cell count (necrotic + apoptotic cells) minus the TUNEL-positive cell count (apoptotic cells) equals the number of necrotic cells. The numbers of necrotic and apoptotic cells are given as percentages of the total number of cells (PI-stained cell count).

Experimental Protocol

Lungs from seven groups of animals (n = 5 for each group) were studied.

Controls. Two groups of controls were studied, as follows:

1. Trypan blue controls (tCTRL). These lungs were flushed with the preservation solution followed by the trypan blue solution to quantify cell death. They were then assessed for the number of apoptotic cells on the basis of TUNEL staining. This group provided a measure of the total number of dead cells and the number of apoptotic cells in nonischemic lungs (baseline). The number of necrotic cells was derived by subtraction of the number of apoptotic cells (TUNEL-positive; see details as subsequently described) from the total number of dead cells (Trypan blue-positive).

2. Control (cCTRL). These lungs were flushed with the preservation solution and were assessed for the number of apoptotic cells on the basis of TUNEL staining. This group served as a control for the combined assessment regimen of flushing with trypan blue followed by TUNEL staining, to determine whether the initial flush with trypan blue solution affected the outcome of TUNEL staining. It also provided a measure of apoptosis in nonischemic lungs.

Transplant groups. Lungs from five groups of animals underwent different periods of cold storage followed by transplantation and by 2 h of graft reperfusion. The following time periods were used for cold storage: 20 min (minimal possible ischemic time), 6 h, 12 h, 18 h, and 24 h. After storage, the right lungs were flushed through the right PA with the trypan blue regimen. The left lungs were transplanted into recipient Lewis rats and then underwent 2 h of reperfusion. The left lungs were then flushed with the trypan blue solution and the tissues were harvested and fixed as described earlier.

Statistical Analysis

All data are expressed as mean ± SD. One-way analysis of variance (ANOVA) was used to determine statistical significance. A value of p < 0.05 was considered statistically significant. When statistical significance was reached, it was followed by a post hoc analysis using the Student-Newman-Keuls test. The Pearson's product moment test was used to correlate Pa_O2 values with the number of apoptotic and necrotic cells in tissue samples. The SigmaStat software package version 1.0 (Jandel Scientific, San Rafael, CA) was used for all statistical analyses.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Cell Necrosis Induced by Prolonged Hypothermic Preservation

After cold ischemic storage alone (right lungs), minimal cell death was seen in the 6-h and 12-h groups. Significant cell death was detected only in lungs after 18 h and 24 h of cold storage as determined through trypan blue-positive staining (Figure 1). Almost no TUNEL-positive cells (bright green staining) were seen in these groups, and the detected dead cells were therefore identified as necrotic.

Figure 2 shows the results of the histologic analyses performed on tissue specimens from the two control groups and the cold-preserved, nontransplanted right donor lungs for five preservation periods (20 min and 6, 12, 18, and 24 h). The control, 20-min (minimal ischemia), 6-h, and 12-h groups each had a percentage of dead cells (Trypan blue-positive) that was less than 3% of total cells. However, the percentage of dead cells increased to 10.5 ± 1.5% of total nucleated cells (1.1 ± 1.2% apoptotic cells and 9.3 ± 1.6% necrotic cells) after 18 h of cold preservation (p < 0.05). After 24 h of cold preservation, no apoptotic cells were detected, whereas 27.3 ± 8.4% of the cells were necrotic. This was significantly different from all other groups (p < 0.05).

View larger version (16K): [in this window] [in a new window]   Figure 2.   Cell death in lungs after 20 min and 6, 12, 18, and 24 h of cold storage. Cell death was assessed as described in Figure 1 and was quantified as the percentage of the total number of cells in tissue sections. Lungs stored for 18 h and 24 h showed significantly increased cell necrosis after cold storage. This was not seen in lungs stored for 20 min, 6 h, or 12 h. None of the groups demonstrated a significant degree of apoptosis at the end of cold ischemia. #p < 0.05 (necrosis and total nonviable cells) versus all other groups; *p < 0.05 (necrosis and total nonviable cells) versus cCTRL, tCTRL, 20-min, 6-h and 12-h groups. cCTRL: normal control lungs without trypan blue flush, tCTRL = normal control lungs with trypan blue flush.

Apoptosis Induced by Reperfusion of Transplanted Lungs after Short Periods of Ischemic Hypothermic Preservation but Not by Reperfusion after Prolonged Ischemia

Representative staining results for the transplanted left lungs after transplantation and 2 h of reperfusion are shown in Figure 3. In comparison with cell death in lungs after cold ischemic storage alone (Figure 1), only small amounts of dead cells were found in the control group, whereas after reperfusion, TUNEL-positive cells became a prominent feature in both the 6-h and the 12-h groups, and necrosis remained predominant in the 18-h and 24-h groups.

Findings in the histologic analyses performed on tissue specimens from the control groups and on the transplanted and reperfused left lungs from the five cold preservation periods (20 min and 6, 12, 18, and 24 h) are shown in Figure 4. An increase in the total percentage of dead cells was seen in all transplant groups as compared with the control groups (p < 0.05), except for the 20-min transplant group. There was no significant difference from the controls in the percentages of total dead cells in any of the four groups with 6-h, 12-h, 18-h, and 24-h storage. However, there were dramatic differences in the types of dead cells among groups, with apoptotic cells predominating after reperfusion in the 6-h and 12-h lungs, and necrotic cells predominating after reperfusion in the 18-h (21%) and 24-h (29%) lungs. The percentages of apoptotic cells in the 6-h (28.5 ± 4%) and 12-h (29.8 ± 3%) groups were significantly different from those in all other groups, as were the percentages of necrotic cells, which were < 2% in both groups (p < 0.05). The ratios of apoptotic to necrotic cells were 26:1 in the 6-h group and 34:1 in the 12-h group, whereas in the 18-h and 24-h groups, the ratios changed to 1:9 and 1:44, respectively.

View larger version (24K): [in this window] [in a new window]   Figure 4.   Cell death in lungs after 20 min and 6, 12, 18, and 24 h of cold storage followed by transplantation and reperfusion. Cell death was quantified as the percentage of the total number of cells in tissue sections. Whereas almost no cell death was seen in reperfused lungs that were stored for 20 min, apoptosis was seen as the prominent mode of cell death in lungs after 6 h and 12 h of cold storage followed by 2 h of reperfusion. Lungs kept in cold storage for 18 h and 24 h followed by 2 h of reperfusion showed similar numbers of dead cells as in lungs in the 6-h and 12-h groups, but the dead cells were predominantly necrotic. *p < 0.05 versus control groups and 20-min, 18-h, and 24-h groups; #p < 0.05 versus control groups and 20-min group; p < 0.05 versus control groups and 20-min, 6-h, and 12-h groups. p < 0.05 versus all other groups.

Necrosis but Not Apoptosis Was Associated with Deteriorating Organ Function

All donor organs were comparable in terms of preretrieval blood gas values, with an average Pa_O2 of 545 ± 59 mm Hg and Pa_CO2 of 33 ± 6 mm Hg. There were no significant differences among the groups.

After 2 h of graft reperfusion, the mean Pa_O2 values in the 20-min group were comparable with preretrieval values. However, the mean Pa_O2 values decreased continuously (p < 0.05) as ischemic time increased from 20 min to 24 h (Figure 5). The mean Pa_CO2 values measured 2 h after reperfusion were stable in the 20-min, 6-h, and 12-h groups, but increased significantly in the 18-h and 24-h groups (p < 0.05) (Figure 5). The pH values were not significantly different either within or among the groups.

View larger version (15K): [in this window] [in a new window]   Figure 5.   Mean PaO2 and PaCO2 values in transplanted lungs after 2 h of graft reperfusion. Posttransplant PaO2 levels decreased significantly with the length of cold organ storage. PaCO2 levels were stable in blood taken from lungs after 20 min, 6 h, and 12 h of cold storage and 2 h of graft reperfusion, but were significantly increased after 18 h and 24 h of cold storage and reperfusion. ¶p < 0.05 for PaO2 versus all other groups; *p < 0.05 for PaO2 versus 12-h, 18-h, and 24-h groups; #p < 0.05 for PaO2 versus 18-h and 24-h groups; p < 0.05 for PaO2 versus 24-h groups; p < 0.05 for PaCO2 versus 20-min, 6-h, and 12-h groups; §p < 0.05 for PaCO2 versus all other groups.

The observed switch from apoptotic to necrotic cell death as the predominant mode of cell death in transplanted and reperfused lungs is illustrated in Figure 6. The numbers of apoptotic and necrotic cells in the 20-min group were less than 3% in both. In the 6-h and 12-h groups, the number of necrotic cells after reperfusion was significantly lower than was the number of apoptotic cells. This phenomenon changed in the 18-h and 24-h groups, in which the numbers of necrotic cells became significantly higher than the numbers of apoptotic cells. A significant negative statistical correlation was seen (p < 0.0001) between the percentage of necrotic cells in the reperfused lungs and lung function as assessed by Pa_O2 (Figure 7A), but there was no significant correlation between apoptotic cell death and graft function, as shown in Figure 7B.

View larger version (11K): [in this window] [in a new window]   Figure 6.   Change from apoptosis to necrosis as the major type of cell death in lungs after 20 min and 6, 12, 18, and 24 h of cold preservation followed by transplantation and reperfusion. In reperfused lungs that were preserved only for 20 min, no significant apoptosis or necrosis was seen. After longer periods of ischemia followed by reperfusion, significant cell death was observed. After 6 h and 12 h, apoptosis was the dominant mode of cell death, and after 18 h and 24 h this changed in that necrosis became the dominant mode. Interestingly, the two types of cell death did not coexist in the groups with greater degrees of cell death, and the percentages of dead cells in the 6-, 12-, 18-, and 24-h groups were very similar. *p < 0.05 versus apoptosis in the 6-, 12-, 18-, and 24-h groups.

View larger version (16K): [in this window] [in a new window]   Figure 7.   Statistical correlation between lung function (PaO2) and cell death. (A) A significant statistical correlation between necrosis and graft function was seen over the entire study period, whereas (B) apoptosis did not significantly correlate with lung function. r = correlation coefficient.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

In this study we examined the mode of cell death following IR injury in a rat single-lung transplant model. To our knowledge, this is the first description of a correlation between the mode and quantity of cell death and posttransplant lung function. First, we have shown that the length of the cold ischemic preservation period affects the degree of cell death in the preserved, nontransplanted lung. After both 18 h and 24 h of preservation, the degree of cell death increased significantly. Although reliable and reproducible long-term organ preservation in the setting of clinical lung transplantation has not been described, the currently accepted reasonable storage time for human lungs is between 6-8 h. Our findings of increased cellular necrosis in lungs preserved for a long period, and of a statistical correlation with deteriorating graft function following transplantation, may explain at the histopathologic level the limitations in safe ischemic time in lung transplantation. Second, we examined the proportions of apoptotic and necrotic cell death after preservation, transplantation, and reperfusion. We describe a switch from a predominantly apoptotic mode of cell death in transplanted lungs after 6 h and 12 h of preservation and reperfusion to a predominantly necrotic mode of cell death after 18 h and 24 h of preservation and reperfusion. Third, we demonstrated that the percentage of necrotic cells inversely correlates with oxygenation, the major endpoint reflecting posttransplant lung function.

The percentage of apoptotic cells has been examined in other organs after IR injury. After human liver transplantation and reperfusion, 18% of cells stained positively for apoptosis (14). The percentage of apoptotic cells measured in our rat study is comparable with that in our human study after 1 to 5 h of preservation and 2 h of reperfusion, in which we demonstrated that 34% of the cells were apoptotic (17). In the animal model used in the present study, we found 30% of the cells to be apoptotic after 6 h of preservation and transplantation with 2 h of reperfusion. The finding of similar rates of apoptosis in both the rat model and human lung transplant recipients suggests that the cell injury causing such death is probably a result of ischemia and reperfusion, rather than of immunologic factors related to graft rejection, since the rat model is isogeneic and the human transplants were allogeneic.

Other groups have examined the extent of cell death following lung transplantation. D'Armini and coworkers showed a cellular loss of 22 to 26% with a trypan blue flush technique in lungs from non-heart-beating donor animals after 4 h of circulatory arrest. The type of injurious insult in that study was quite different from that in our study. In contrast to common clinical conditions, the lungs in D'Armini and coworkers' study were subjected to 4 h of warm ischemia without cold preservation before the warm ischemic period. After the 4-h warm-ischemic period, the lungs were either flushed with Euro Collins solution or with University of Wisconsin solution and stored for 4 h at 4° C. The trypan blue flush was performed after this cold storage period. Therefore, the trypan blue-stained cells probably represented cells that had been fatally injured by the extended period of warm ischemia (21). The trypan blue flush for detecting dead cells has also been used in other organ systems in transplantation-related studies, such as the liver (22). Schemmer and colleagues have shown increased levels of damaged cells after liver transplantation when the liver was aggressively manipulated during retrieval. This demonstrated that surgical manipulation during organ retrieval may contribute to cell injury and adversely affect posttransplant graft function (22).

The studies just described examined either apoptosis or nonspecific cell death, but not both in the same tissue samples. We have tried to achieve this by adopting a triple staining technique previously described in cell culture models for the examination of different modes of cell death (23). For the detection of dead cells in lungs, the transvascular trypan blue flush has been used and validated (26, 27). This technique has also been used in other organ and tissue studies, and it has gained increasing popularity because of its simplicity, with the easily identifiable endpoint of visual identification of blue-stained nuclei of dead cells, produced according to the principles of dye exclusion (28). In the present study we used trypan blue staining to identify all dead cells (necrotic and apoptotic). For detecting apoptosis in situ in tissue samples, the TUNEL technique has been proven to be the most sensitive technique now available (29), and has therefore been used in studies of tissue sections (14, 33). Ansari and coworkers have suggested the TUNEL technique as a helpful tool for quantifying apoptosis in tissue sections (30).

Although it has been shown to be more accurate and sensitive than morphologic studies (34), limitations of the specificity of the TUNEL and other techniques for detecting apoptosis have been reported (35). Darzynkiewicz and associates refer to three systematic errors that have been made in the detection of apoptosis with cell culture models. These are: (1) misclassification of nuclear fragments and individual apoptotic bodies as single apoptotic cells; (2) assumption that the apoptotic index represents the rate of cell death; and (3) failure to confirm by microscopy that cells classified by flow cytometry as apoptotic or necrotic do indeed show morphology consistent with these classifications (35). We attempted to obviate potential sources of error by counting only cells that could clearly be identified as cells and which did not appear to represent background or fragment staining. In our previous examination of this process in transplanted human lungs, we confirmed through electron microscopy that TUNEL-positive cells were in fact undergoing apoptosis (17). Since both trypan blue staining and the TUNEL assay are microscopy-based techniques, we combined them in the present study to investigate the effects of ischemia and reperfusion on cell death. Thus, these two modes of cell death can be examined on the same tissue microscopy slide. Although this combination has not been used in tissue specimens, it has been used to examine cell viability in various cell culture studies (23).

An important issue is to confirm the specificity of this new method, which still requires validation. Interestingly, although most of the TUNEL-positive staining in our study was clearly nuclear, some of the apoptotic cells in reperfused lungs in the 6-h and 12-h groups appeared to exhibit a cytoplasmic distribution of TUNEL staining. This type of staining was not seen in our previous studies of transplanted human lungs or in our pilot studies of transplanted rat lungs, in which only TUNEL was used for detecting apoptosis. The observed cytoplasmic staining may therefore be an artifact resulting from interference between the immunofluorescent TUNEL staining and trypan blue. However, the total number of TUNEL-positive cells found in these two settings was the same (approximately 30%), and this artifact therefore did not appreciably interfere with our results. We used the combined TUNEL and trypan blue staining method to analyze all of the experimental groups in our study. TUNEL-positive staining was seen only in the 6-h and 12-h groups after reperfusion, and not in the 18-h or 24-h groups, either after cold ischemic preservation or after reperfusion, even though substantial necrosis was seen in the latter two groups. This is further evidence that the TUNEL method specifically stains apoptotic cells in the presence of trypan blue. Accordingly, this technique can easily distinguish apoptosis from necrosis. Alternatively, the specific injury seen with ischemia and reperfusion of the lung may lead to a unique diffuse cytoplasmic distribution of nuclear fragments. It was beyond the scope of our study to explain this finding, but further work is clearly required to determine if this is the case.

Taking these limitations into account, we have shown that the combination of the TUNEL and trypan blue techniques is valid in tissue work when the trypan blue is given as a transvascular in situ flush to the whole organ and the TUNEL is done secondarily on paraffin-embedded sections of the blue stained tissue samples. Consequently, this technique provides a useful approach to studying cell death and injury related to organ preservation and reperfusion, and prospectively to potential pharmacologic or gene therapeutic strategies to prevent cell death in experimental lung transplantation.

Perry and coworkers emphasize that cells at early stages of apoptosis can actively exclude trypan blue and thus do not stain blue, whereas cells in later stages of apoptosis take up trypan blue and therefore stain blue (25). In our study we found that 97% of TUNEL-positive cells also stained with trypan blue, suggesting that at 2 h after graft reperfusion, cells in the lungs we studied were mainly at a later stage of cell death, and that the trigger for injury probably occurs in the early phase after reperfusion.

That necrosis rather than apoptosis was the major type of cell death found in lungs after 18 h and 24 h of preservation, transplantation, and reperfusion may explain the severity of lung dysfunction seen after longer preservation periods. After 24 h of preservation of the nontransplanted lungs in our study, 27% of the cells were dead (trypan blue-positive) and did not stain with the TUNEL technique. There was no significant increase in the percentage of dead cells after transplantation and 2 h of reperfusion. We did not detect any apoptotic cells after this latter procedure. It is likely that the lung cells in the grafts in these groups were already severely injured by the long preservation period, and were thus only minimally further affected by reperfusion. Apoptosis is an energy-dependent process of programmed cell death that is part of what are referred to as genetic "suicide" programs (35, 36). The absence of apoptotic cell death at 24 h may have been due to such severe damage to the cells that they no longer had the energy reserves to undergo the energy-requiring but less tissue damaging process of programmed cell death.

Because apoptotic cells in an organ system are phagocytosed by macrophages before their membranes break down and intracellular enzymes are released, apoptosis does not lead to significant tissue inflammation, which is characteristically seen with necrosis (36). Ischemia has been described as a potential inducing factor for apoptosis (37). TNF-, which is a mediator of IR injury in the early phase after organ reperfusion (11), is a potent activator of apoptosis (38). An interaction between oxidants and TNF- in the activation of nuclear factor B, which is also known to be involved in signaling for apoptosis, has also been described (39).

Apoptosis is a relatively "quiescent" form of cell death as compared with necrosis, which induces inflammation and cytokine release. Ideally, it is important to minimize all cell death related to the injury of transplantation. However, if an inevitable degree of cell death must occur, it would seem to protect the whole organism for cells to undergo apoptosis rather than necrosis.

In summary, we have shown that with shorter periods of ischemia, the mode of cell death after reperfusion is primarily apoptotic, and that with longer periods of ischemia, cell death after reperfusion is primarily necrotic. Increasing necrosis was associated with a significant deterioration in transplanted lung function. Clearly, cell death is an important component of transplantation-related injury. An improved understanding of the mechanisms of cell death may be important for the development of novel strategies to prevent or to modify the mode of cell death in organ transplantation, and consequently to decrease the severity of organ dysfunction after transplantation. These strategies might involve a variety of approaches to modifying the balance between apoptosis and necrosis, including gene therapy (e.g., transfection of antiapoptotic genes such as bcl-2) and inhibition with caspase or angiotensin converting enzyme, as well as improved organ preservation strategies. The model and techniques used in the study reported here will be useful for testing these new strategies, which it is hoped will help to resolve one of the major problems in human organ transplantation: the injury due to preservation, ischemia, and reperfusion.