Islet transplantation is an attractive alternative to insulin administration for the therapy of type I diabetes mellitus. Unfortunately, pilot clinical trials of islet transplantation have not consistently yielded optimal results, with only a small percentage of grafts showing measurable function over time (1). Graft loss occurs within a few weeks after implantation in 60% of human islet allogeneic transplants, and there is substantial evidence to link this event to early death of a high number of transplanted cells. Although it remains unclear whether islet cell death occurs by necrosis or apoptosis, the latter seems to play a significant role in this particular setting. In vitro it is established that apoptosis participates in the death of freshly isolated islets cultured in standard conditions (2). Additionally, withdrawal of selected growth factors (3,4) and of native extracellular matrix proteins (5,6) can trigger islet apoptosis in experimental models. Finally, apoptosis has been shown in ischemia-reperfusion lesions occurring after organ transplantation (7,8), and it is conceivable that islet grafts might also undergo comparable or even more severe ischemia-reperfusion–like damage.
Apoptosis is a process of cell death in which the cell activates an intrinsic suicide mechanism. The best-defined apoptotic pathways proceed through engagement of Fas and TNF-receptor 1, in which binding of either receptor by a ligand or a cytokine induces a death signal. All apoptotic pathways so far described converge toward the activation of cytoplasmic cysteine proteases named caspases (9). To date, 10 different caspases have been defined (10). They share a specific enzymatic activity, cleaving their substrate after aspartic acid residues, and proenzymes are similarly processed into active form through cleavage at aspartic acid residues by other caspases or through autocatalysis. Targets of these proteins are not completely known, but their activation leads to the cleavage of DNA into nucleoside-sized fragments, a late event characterizing apoptotic cell death. Morphologically, cells undergoing apoptosis exhibit typical chromatin and cytoplasmic condensation followed by nuclear fragmentation and generation of one or more apoptotic bodies. Detection of DNA fragmentation by in situ labeling with the terminal deoxynucleotidyl transferase (TdT)-mediated dUTP nick end labeling (TUNEL) assay and agarose gel electrophoresis (showing a characteristic ladder-like pattern of DNA fragments), are commonly used to define apoptotic death. However, these tests only allow detection of apoptosis in its late stages. Measurement of caspase 3 activity and annexin V binding are newly available assays that allow the detection of early events of apoptosis. Caspase 3 is arguably the caspase that best correlates with apoptosis and is one of the more distal proteases in the apoptosis pathway (9). Annexin V binding detects specific early changes occurring in cells entering apoptosis, namely the expression of membrane-bound phosphatidylserine at the outer surface of the cell membrane (11). The events revealed by these tests have been shown to precede and predict the appearance of late apoptotic markers in a number of models (11–17).
The aim of this study was to validate the use of annexin V binding and caspase 3 activity measurement as reliable early markers of apoptosis. We analyzed apoptosis occurrence with annexin and caspase testing in islets of Langerhans isolated from rats under standard conditions and compared it with analysis of DNA fragmentation. Identification of such early markers seems to be a valuable means to increase our analytical power to gain information on islet preparation quality.
MATERIALS AND METHODS
Lewis rats were purchased from Harlan Sprague Dawley Inc. (Indianapolis, IN). All animals were kept at the University of Miami animal facilities and used in compliance with the U.S. Department of Agriculture and National Institutes of Health regulations. All animal manipulations were conducted and monitored under protocols reviewed and approved by the Institution Animal Care and Use Committee.
Islets were isolated as previously described (18) by intraductal collagenase digestion (collagenase type V, 1.5 mg/ml, Sigma Chemical Co., St. Louis, MO) and purified over a discontinuous Euroficoll gradient (1.110, 1.096, 1.069, 1.037). Islet purity, evaluated by dithizone staining, was >95% in all isolations. Islet number was expressed as islet equivalent number (IEQ). Before being used for in vitro assays, islets were cultured overnight at 37°C, in a 5% CO2 humidified atmosphere, in CMRL-1066 medium, supplemented with 20 mM l-glutamine, 100 U/ml penicillin, 100 μg/ml streptomycin, and 10% heat-inactivated bovine serum. Samples of 300 IEQ were plated in 2 ml of culture medium for the caspase 3 and annexin V assays, and samples of 1000 IEQ were plated in 6 ml of culture medium for TUNEL assay and agarose gel DNA electrophoresis and cultured for different time periods, as indicated.
Cytokine treatment of isolated islets.
Recombinant rat interleukin-1β (IL-1β; 50 U/ml), recombinant rat interferon-γ (IFN-γ; 103 U/ml), and recombinant rat tumor necrosis factor-α (TNF-α; 103 U/ml), all purchased from R&D Systems Inc. (Minneapolis, MN), were used in combination. Different incubation periods were used for each assay. For the caspase 3 and annexin V assays, islets were harvested before incubation and after 3, 6, and 12 hr. For TUNEL assay and DNA electrophoresis, islets were harvested before incubation and after 24 and 48 hr. Control islets remained untreated for the corresponding periods.
Caspase 3 activity measurement.
The fluorometric assay FluorAce Apopain Assay Kit (BIO-RAD, Hercules, CA) was used to measure caspase 3 activity. Islets were harvested, washed twice in phosphate-buffered saline (PBS) without calcium and magnesium (Gibco BRL, Grand Island, NY), and centrifuged at 750×g for 5 min. The supernatant was removed, and the pellet was stored at −80°C until the assay was performed. Cells were lysed by a 30-min incubation on ice in lysis buffer (10 mM HEPES, 2 mM EDTA, 0.1% saponin, 5 mM dithiothreitol, 1 mM phenylmethylsulfonyl fluoride, 10 μg/ml pepstatin, 10 μg/ml aprotinin, 20 μg/ml leupeptin), followed by four cycles of freezing and thawing in isopropanol-dry ice and 37°C water bath. Lysed cell extracts were centrifuged at 4°C for 15 min at 10,000×g. The supernatant was collected to assay caspase 3 activity. The fluorogenic peptide substrate acetyl-Asp-Glu-Val-Asp-7-amino-4-trifluoromethyl coumarin (Ac-DEVD-AFC) was used at a concentration of 55 μM. Concurrently, to distinguish caspase 3 activity from nonspecific protease activity, aliquots of the samples were preincubated for 30 min at room temperature before adding the substrate with the peptide fluoromethyl ketone Z-DEVD-FMK, which acts as a specific caspase 3 inhibitor. The inhibitor was treated with esterase (10%; Roche Boehringer Mannheim Corp., Indianapolis, IN) for 15 min and then used at a 15 μM concentration. After a 90-min incubation with the substrate, readings were performed on a VersaFluor fluorometer (BIO-RAD, Hercules, CA), with excitation and emission wavelengths of 390 nm and 520 nm, respectively. Results were expressed as units of apopain per 100 IEQ, according to the formula provided with the FluorAce Apopain Assay Kit.
Annexin V assay.
Translocation of the phospholipid phosphatidylserine from the inner side of the plasma membrane to its outer layer is an early event of apoptosis. Annexin V is a calcium-dependent phospholipid-binding protein with high affinity for phosphatidylserine. Propidium iodide (PI) is a standard cytometric viability probe that is excluded by cells with intact membrane. PI staining is performed simultaneously with the annexin V staining to differentiate apoptotic cells (single annexin V-positive) from necrotic cells (double annexin V-PI–positive), as necrotic cells also expose phosphatidylserine to annexin V because of the loss of membrane integrity (11).
Freshly collected islets were washed twice in PBS without calcium and magnesium, centrifuged at 750×g for 5 min, and dispersed into single cells by gentle continuous pipetting in trypsin-EDTA (0.05% trypsin, 0.53 mM EDTA; GibcoBRL), for 4 min. Enzymatic activity was stopped by the addition of heat-inactivated bovine serum, and cells were washed twice in PBS without calcium and magnesium. Approximately 5×105 cells were obtained per sample, and trypan blue exclusion test showed a cell viability >90%. After centrifugation at 750×g for 5 min, the supernatant was discarded, and the pellet was resuspended and incubated for 15 min at room temperature in HEPES buffer containing 2 μg/ml of fluorescent annexin V and 1 μg/ml of PI (Annexin-V-FLUOS Staining Kit, Roche Boehringer Mannheim Corp). Samples were resuspended in 300 μL of HEPES buffer and read within 2 hr on a FacScan flow cytometer (Becton Dickinson, San Jose, CA) using 488 nm excitation and a 530-nm bandpass filter for fluorescein detection and a 650-nm longpass filter for PI detection. Five thousand cells were analyzed for each sample. Results were expressed as percentage of annexin V-positive and annexin V-PI–double positive cells.
Detection of DNA fragmentation in situ was performed on 5-μm-thick paraffin sections of whole islets using the TUNEL assay (In Situ Cell Death Detection, POD, Roche Boehringer Mannheim Corp.). DNA strand breaks were detected by the enzymatic incorporation of fluorescein-labeled nucleotides to the free 3′-OH DNA ends. Addition of anti-fluorescein antibody Fab′ fragments conjugated with horseradish peroxidase, followed by substrate reaction with diamino-benzidine, allowed the histochemical analysis of the reaction site. Positive controls were treated with DNase (1 μg/μl, Roche Boehringer Mannheim Corp.). Negative controls were obtained by omitting the TdT in the reaction mixture. Analysis was performed under light microscopy. Approximately 700 cells from 10 islets were analyzed for each time point, and an apoptotic index (percentage of labeled cells) was determined.
DNA fragmentation analysis by gel electrophoresis.
Islets were incubated overnight at 50°C in lysis buffer (100 mM NaCl, 10 mM Tris-Cl, pH 8, 25 mM EDTA, pH 8, 0.5% sodium dodecyl sulfate, 0.1 mg/ml proteinase K). After DNA extraction, samples were treated at 37°C for 2 hr with 20 mM Tris-HCl containing 1 μg/ml RNase. A 100-bp DNA ladder (GibcoBRL) was used as a marker. A positive control was included, obtained from hepatocytes of a rat in which multiorgan failure was induced by intraperitoneal administration of lipopolysaccharide (10 mg/kg) and galactosamine (700 mg/kg). Electrophoresis was run at 100 V for 90 min on a 1.2% agarose gel. DNA was stained with ethidium bromide and visualized by ultraviolet light.
All statistics were performed using the Statistica software (Statsoft, Tulsa, OK). Results are expressed as mean±SEM. One-tailed Student’s t test was used for comparison of paired continuous variables. Values of P <0.05 were considered significant.
Analysis of caspase 3 activity.
Rat islets were first analyzed after isolation and overnight culture at 37°C. Measurable caspase 3 activity was detected at this first point in all samples analyzed. Without cytokine stimulation, the enzymatic activity remained at comparable levels throughout the subsequent 12 hr of culture. Cytokine stimulation, on the other hand, induced a rise in caspase 3 activity as early as 3 hr after treatment. Caspase activity levels at this time point were 0.8±0.3 U/100 IEQ in the control group and 1.4±0.45 U/100 IEQ in the cytokine-treated group (P <0.05). Differences in caspase activity were also observed between both groups at 6 hr (1.01±0.3 U/100 IEQ in the control group vs. 1.5±0.35 U/100 IEQ in the treated group, P <0.05) and 12 hr (0.95±0.35 U/100 IEQ in the control group vs. 1.5±0.4 U/100 IEQ in the treated group, P <0.01). Pretreatment of the samples with the caspase 3 inhibitor decreased the enzymatic activity by 80%, showing the specificity of the measured caspase activity. These data are summarized in Figure 1.
Analysis of annexin V binding.
Similarly, analysis of annexin V binding was performed after isolation and overnight culture at 37°C. Single cell suspensions were obtained by gentle trypsin-based dissociation. Preliminary analysis of the effects of trypsinization on the islets showed that annexin binding did not increase with the duration of trypsin treatment (percentages of annexin V-positive cells were 15.5% and 15.4% after 4 min and 10 min of trypsinization, respectively). This suggests that baseline levels of annexin positivity were not caused by trypsinization. Staining at this time point showed 21.0%±5.8% annexin V–single positive cells (apoptotic) and 9.1%±2.5% annexin V-PI–double positive cells (necrotic). Three hours after cytokine stimulation, a slightly higher percentage of apoptotic cells was recorded in the cytokine-treated group (27.1%±10.8%). At this time point, the percentage of annexin V-positive cells in the cytokine-treated group did not differ significantly from that recorded in the untreated samples (25.4%±9%). At 6 hr, the percentage of annexin V–single positive cells was higher in the treated group (27.5%±8.1% in the cytokine-treated group vs. 22.4%±7.2% in the control group). This difference reached statistical significance at P <0.01. Then, the percentage of positive cells increased concurrently in both groups until 12 hr (29.1%±7.4% in the control group and 32.9%±8.9% in the cytokine-treated group), no longer yielding statistically significant differences. These data are summarized in Figure 2 A.
Cytokine treatment resulted also in an increased percentage of necrotic cells at each time point. This difference reached statistical significance only at 3 hr, when double-positive cells were 10.6%±2.6% in the cytokine-treated group and 6.2%±1.7% in the control group (P <0.05). These data are presented in Figure 2 B.
We then assessed a current method of analysis of DNA fragmentation in the experimental conditions previously described. Islets were treated with the cytokine cocktail or untreated, and fixed in formalin at 0, 24, and 48 hr. These time points were selected based on described kinetics of DNA fragmentation after apoptotic signaling. Paraffin-embedded sections were analyzed with the TUNEL assay. The TUNEL assay was chosen because it allows detection of DNA fragmentation in situ and analysis of cell morphology. At time 0, apoptotic index was 1.5%±0.5%, and it increased to 17%±5% at 24 hr and 19%±5% at 48 hr in the control group (Fig. 3 A). In the treated group, the apoptotic index rose to 43%±5% at 24 hr and 69%±11% at 48 hr (Fig. 3 B). At these two time points, >80% of the treated islets also presented characteristic features of necrosis with loss of capsule integrity and uninhabited core.
DNA fragmentation analysis by gel electrophoresis.
The following apoptosis-detection assay performed on islet preparations was based on the analysis of ladder-like pattern of DNA fragmentation after electrophoresis on agarose gels. DNA was extracted at different time points from islet preparations either untreated or treated with the already described cytokine cocktail. DNA fragmentation was not detected on islet extracts at time 0, i.e., after isolation and overnight culture. In contrast, after 24 hr, gel electrophoresis showed a distinct DNA laddering in the cytokine-treated group, but not in the control group (Fig. 4). DNA fragmentation was no longer detectable after 48 hr in either group. A positive control was included and is shown in lane 4 of the gel, obtained from DNA extracted from the liver of a rat, in which apoptosis was induced by i.p. administration of lipopolysaccharide and galactosamine.
Although substantial progress in the field of islet transplantation has been recently witnessed, the results of clinical trials are still not optimal. A large percentage of islet transplants are lost early, within weeks from the implant. Several causes have been held responsible for this phenomenon, including rejection, recurrence of autoimmunity, and nonspecific transplant environment perturbation, including activation of coagulation and induction of an inflammatory response. Furthermore, the intrinsic quality of the islet preparation might play a fundamental role in determining the outcome of the graft. It would, therefore, be of great value to define reliable tests that allow us to predict islet cell function before transplantation. Islet preparation quality control is, in fact, a very undeveloped area, and the only assays available are usually time-consuming and often allow only retrospective analysis of the preparations, after they have been transplanted. For example, analysis of apoptosis based on DNA fragmentation, although accepted as a gold standard, only explores late events of programmed cell death. Identification of early markers of apoptosis would represent a tremendous advantage in terms of perspective analysis of islet quality control. Therefore, newly available assays for early apoptosis detection (annexin V and caspase 3 measurements) were analyzed in this study and compared with assays based on detection of DNA fragmentation.
A measurable level of apoptosis was detected after islet isolation and overnight culture only with caspase 3 and annexin V staining. Remarkably, neither TUNEL-positive nuclei nor DNA laddering were consistently observed at this time point.
Apoptosis was then induced by a cocktail of cytokines that has been described as a powerful apoptotic cell death stimulus in isolated islets (18–20). We reasoned that the delivery of this apoptotic signal would allow us to compare quantitative differences in the levels of apoptosis, using both early and late markers of programmed cell death, contributing to the validation of the assays studied.
An increase of caspase 3 activity was the earliest event detected after cytokine stimulation, reaching statistically significant differences as early as 3 hr after the delivery of the apoptotic stimulus. Annexin V-PI staining also showed significantly different percentages of positive cells in the treated versus untreated groups, although at a slightly later time (6 hr). The appearance of TUNEL-positive nuclei in the control group 24 and 48 hr later and the increase in the percentage in the cytokine-treated group clearly validated the signals detected by caspase 3 and annexin V assays. Cytokine treatment induced DNA fragmentation, as also assessed by the appearance of a DNA ladder pattern. Because of this late positivity and the long delay required for sample processing, these two assays (TUNEL and DNA laddering) appear less suited to assess islet preparation quality before transplantation.
The analysis of annexin V-PI double staining revealed that the baseline percentage of apoptotic cells was double that of necrotic cells, underlying the important part of apoptosis in isolation-induced cell death. The percentage of double-positive annexin V-PI cells increased significantly 3 hr after treatment with the cytokine cocktail, and continued to rise during the time of analysis. Recently, the transfection of human islet cells with the anti-apoptotic gene Bcl-2 has been shown to fully protect them from cytokine destruction (21), suggesting the quasi-exclusive apoptotic effect of this cytokine treatment. However, in our study, the morphologic changes observed in the TUNEL assay provide further evidence to sustain the results of the annexin V-PI double staining, suggesting that cytokine treatment induces necrosis as well as apoptosis. In fact, it has been suggested that the interchangeable use of the terms apoptosis and programmed cell death was inappropriate. A large morphologic diversity in developmental programmed cell death has already been reported. Its usual consequence is apoptosis, but in some instances, it exhibits several morphologic features of necrosis (22,23). Therefore, the necrosis we measured in the annexin V assay might partly represent the effect of programmed cell death.
Isolation-induced cell death might be in part related to anoikis (apoptosis owing to lack of interaction with surrounding matrix or cells) and lack of growth factors (3–5). Additionally, chemicals and contaminants present during isolation and purification might contribute to the observed apoptosis. For example, it is known that most collagenase preparations contain large amounts of contaminating endotoxin (24). Endotoxin can indeed mediate signal transduction and induce alteration in the antigenic profile of islets, and has been associated with apoptosis induction and primary nonfunction in islets (25).
A relatively high variability in the degree of postisolation apoptosis and necrosis was observed with the caspase 3 and annexin V assays in the islet preparations obtained from different isolations. This may seem quite surprising, inasmuch as the isolation procedure is standardized. This illustrates the presence of unknown and uncontrolled variables that determine the quality of the isolation. This is a well-known fact of islet isolation in the clinical setting, and renders these tests highly valuable for the assessment of islet quality after isolation.
Collectively, our data shows that the caspase 3 and annexin V assays are valuable and reliable tools for the detection of early events of apoptosis induced by the isolation procedure. Whereas caspase 3 measurement has the advantage of being detectable as early as 3 hr after the delivery of an apoptotic signal, annexin V-PI staining is more suitable to evaluate in a quantitative fashion viability of the preparations. These two assays are easy to perform and have a good reproducibility. If analysis of apoptosis or necrosis is to be a powerful assay for the prospective analysis of islet function, within time frames compatible with a transplantation procedure, the early assays seem to provide a concrete advantage, both in terms of baseline sensitivity and kinetics.
1. International Islet Transplant Registry. Newsletter no.8, August 1999.
2. Paraskevas S, Duguid WP, Maysinger D, Feldman L, Agapitos D, Rosenberg L. Apoptosis occurs in freshly isolated human islets under standard culture conditions. Transplant Proc 1997; 29: 750.
3. Petrik J, Arany E, McDonald TJ, Hill DJ. Apoptosis in the pancreatic islet cells of the neonatal rat is associated with a reduced expression of insulin-like growth factor II that may act as a survival factor. Endocrinology 1998; 139: 2994.
4. Ilieva A, Yuan S, Wang RN, Agapitos D, Hill DJ, Rosenberg L. Pancreatic islet cell survival following islet isolation: the role of cellular interactions in the pancreas. J Endocrinol 1999; 161: 357.
5. Thomas FT, Contreras JL, Bilbao G, Ricordi C, Curiel D, Thomas JM. Anoikis, extracellular matrix, and apoptosis factors in isolated cell transplantation. Surgery 1999; 126: 299.
6. Rosenberg L, Wang R, Paraskevas S, Maysinger D. Structural and functional changes resulting from islet isolation lead to islet cell death. Surgery 1999; 126: 393.
7. Borghi-Scoazec G, Scoazec JY, Durand F, et al. Apoptosis after ischemia-reperfusion in human liver allografts. Liver Transpl Surg 1997; 3: 407.
8. Kuo PC, Drachenberg CI, de la Torre A, et al. Apoptosis and hepatic allograft reperfusion injury. Clin Transplant 1998; 12: 219.
9. Thornberry NA, Lazebnik Y. Caspases: enemies within. Science 1998; 281: 1312.
10. Alnemri ES, Livingston DJ, Nicholson DW, et al. Human ICE/CED-3 protease nomenclature. Cell 1996; 87: 171.
11. Vermes I, Haanen C, Steffens-Nakken H, Reutelingsperger C. A novel assay for apoptosis: flow cytometric detection of phosphatidylserine expression on early apoptotic cells using fluorescein labelled annexin V. J Immunol Methods 1995; 184: 39.
12. Keane RW, Srinivasan A, Foster LM, et al. Activation of CPP32 during apoptosis of neurons and astrocytes. J Neurosci Res 1997; 48: 168.
13. Pepper C, Thomas A, Tucker H, Hoy T, Bentley P. Flow cytometric assessment of three different methods for the measurement of in vitro apoptosis. Leuk Res 1998; 22: 439.
14. Shounan Y, Feng X, O’Connell PJ. Apoptosis detection by annexin V binding: a novel method for the quantitation of cell-mediated cytotoxicity. J Immunol Methods 1998; 217: 61.
15. Namura S, Zhu J, Fink K, et al. Activation and cleavage of caspase-3 in apoptosis induced by experimental cerebral ischemia. J Neurosci 1998; 18: 3659.
16. Yue TL, Wang C, Romanic AM, et al. Staurosporine-induced apoptosis in cardiomyocytes: a potential role of caspase-3. J Mol Cell Cardiol 1998; 30: 495.
17. Yamada K, Ichikawa F, Ishiyama-Shigemoto S, Yuan X, Nonaka K. Essential role of caspase-3 in apoptosis of mouse beta-cells transfected with human Fas. Diabetes 1999; 48: 478.
18. Rabinovitch A, Suarez-Pinzon WL, Shi Y, Morgan AR, Bleackley RC. DNA fragmentation is an early event in cytokine-induced islet beta-cell destruction. Diabetologia 1994; 37: 733.
19. Delaney CA, Pavlovic D, Hoorens A, Pipeleers DG, Eizirik DL. Cytokines induce deoxyribonucleic acid strand breaks and apoptosis in human pancreatic islet cells. Endocrinology 1997; 138: 2610.
20. Ling Z, In’t Veld PA, Pipeleers DG. Interaction of interleukin-1 with islet beta-cells: distinction between indirect, aspecific cytotoxicity and direct, specific functional suppression. Diabetes 1993; 42: 56.
21. Rabinovitch A, Suarez-Pinzon W, Strynadka K, et al. Transfection of human pancreatic islets with an anti-apoptotic gene (bcl-2) protects beta-cells from cytokine-induced destruction. Diabetes 1999; 48: 1223.
22. Clarke PGH. Developmental cell death: morphological diversity and multiple mechanisms. Anat Embryol 1990; 18: 195.
23. Schwartz LM, Smith SW, Jones ME, Osborne BA. Do all programmed cell deaths occur via apoptosis? Proc Natl Acad Sci USA 1993; 90: 980.
24. Vargas F, Vives-Pi M, Somoza N, et al. Endotoxin contamination may be responsible for the unexplained failure of human pancreatic islet transplantation. Transplantation 1998; 65: 722.
25. Berney T, Molano RD, Cattan P, et al. Endotoxin-mediated delayed islet graft function is associated with increased intra-islet cytokine production and islet cell apoptosis. Transplantation 2001; 71: 125.