Involvement of a proapoptotic gene (BBC3) in islet injury mediated by cold preservation and rewarming

Abstract

Long-term pancreatic cold ischemia contributes to decreased islet number and viability after isolation and culture, leading to poor islet transplantation outcome in patients with type 1 diabetes. In this study, we examined mechanisms of pancreatic cold preservation and rewarming-induced injury by interrogating the proapoptotic gene BBC3/Bbc3, also known as Puma (p53 upregulated modulator of apoptosis), using three experimental models: 1) bioluminescence imaging of isolated luciferase-transgenic (“Firefly”) Lewis rat islets, 2) cold preservation of en bloc-harvested pancreata from Bbc3-knockout (KO) mice, and 3) cold preservation and rewarming of human pancreata and isolated islets. Cold preservation-mediated islet injury occurred during rewarming in “Firefly” islets. Silencing Bbc3 by transfecting Bbc3 siRNA into islets in vitro prior to cold preservation improved postpreservation mitochondrial viability. Cold preservation resulted in decreased postisolation islet yield in both wild-type and Bbc3 KO pancreata. However, after culture, the islet viability was significantly higher in Bbc3-KO islets, suggesting that different mechanisms are involved in islet damage/loss during isolation and culture. Furthermore, Bbc3-KO islets from cold-preserved pancreata showed reduced HMGB1 (high-mobility group box 1 protein) expression and decreased levels of 4-hydroxynonenal (4-HNE) protein adducts, which was indicative of reduced oxidative stress. During human islet isolation, BBC3 protein was upregulated in digested tissue from cold-preserved pancreata. Hypoxia in cold preservation increased BBC3 mRNA and protein in isolated human islets after rewarming in culture and reduced islet viability. These results demonstrated the involvement of BBC3/Bbc3 in cold preservation/rewarming-mediated islet injury, possibly through modulating HMGB1- and oxidative stress-mediated injury to islets.

islet transplantation has been widely accepted as a safe and effective treatment for patients with type 1 diabetes (8). However, the success of islet transplantation is hindered by a shortage of donor islets. Prolonged pancreatic cold ischemia has been shown to negatively affect islet yield and viability after isolation (9, 20, 21, 36). Furthermore, the duration of pancreas cold ischemia has also been shown to be associated with the loss of islets during the initial 20 h of islet culture (17). Since postculture islet number and potency are key for transplantation success clinically, it is important to prevent cold ischemia/rewarming injury of islets. It has been widely accepted that cell injury and death mediated by donor organ cold preservation are caused mainly by hypoxia and hypothermia. This cell death occurs mostly during reoxygenation and rewarming through increased oxidative stress and activation of proinflammatory pathways (22, 34). The rewarming injury of islets is initiated by warm digestion during islet isolation and ensuing culture. Cold preservation-mediated cell death is generally considered to be due to necrosis, characterized by the loss of plasma membrane integrity, cell swelling, and organelle damage. Isolated human islets release HMGB1 (high-mobility group box 1 protein), a protein known to be released from necrotic cells (27). However, studies have also shown the involvement of apoptosis in cold ischemia-reperfusion injury, depending on the cell types and duration of cold preservation (4, 37).

The proapoptotic gene Bbc3, also known as PUMA (p53 upregulated modulator of apoptosis), a member of the Bcl-2 (B-cell lymphoma 2) family, plays an important role in β-cell death. We have shown previously that BBC3 expression is upregulated in human islets after exposure to recombinant human TNFα or in combination with IFNγ, leading to β-cell apoptosis through the mitochondrial pathway in vitro (31). High BBC3 mRNA levels in human islets are associated with poor graft function after transplantation in streptozotocin-induced diabetic NOD Scid mice. Studies have shown that Bbc3-mediated β-cell death is also associated with exposure to high glucose, endoplasmic reticulum (ER) stress, oxidative stress, and hypoxia through another Bcl-2 family protein, Bim (11, 24, 42, 45). Involvement of Bbc3 in ischemia-reperfusion-induced apoptosis was reported in intestinal cells, cardiomyocytes, and cerebral cells (2, 23, 43, 44). How Bbc3 contributes to β-cell death following pancreatic cold ischemia and islet isolation is not known.

Determination of the mechanism involved in cold ischemia/rewarming of the pancreas will facilitate the development of new strategies to prevent islet loss prior to transplantation and improve transplantation outcomes. In this study, we examined the involvement of Bbc3 in islet injury caused by pancreas/islet cold preservation and rewarming using three experimental models: 1) bioluminescence imaging of isolated firefly luciferase-transgenic Lewis rat (“Firefly” rat) islets transfected with Bbc3 siRNA, 2) cold preservation and islet isolation from en bloc-harvested Bbc3-knockout (KO) mouse pancreata, and 3) cold preservation/rewarming of human pancreata and isolated islets.

MATERIALS AND METHODS

Experimental animals.

Luciferase-transgenic “Firefly” Lewis rats (7) were obtained from the University of Missouri to establish a breeding colony at the City of Hope Animal Resource Center. Firefly or wild-type (WT) rats between 12 and 16 wk of age were used as pancreas/islet donors. A Bbc3-KO mouse colony was also established at City of Hope with C57BL/6-Bbc3^tm1Ast/J mice obtained from The Jackson Laboratory (Sacramento, CA). Age- (10–14 wk) and sex-matched Bbc3-KO or WT (C57BL/6) mice were used for this study. All of the animal procedures were approved by the Institutional Animal Care and Use Committee at the Beckman Research Institute of the City of Hope and conducted in accordance with federal, state, and local regulations.

Rat islet isolation and cold preservation.

Rat islets were isolated by the infusion of Hanks' balanced salt solution containing collagenase (CIzyme RI; VitaCyte, Indianapolis, IN) into the pancreatic duct, followed by warm digestion in a 37°C water bath, following the procedure suggested by the manufacturer. Immediately after isolation, islets were cold preserved for ≤24 h in 1 ml of organ cold preservation solution, University of Wisconsin (UW) solution, in a 1.8-ml microcentrifuge tube at 4°C. After cold preservation, the islets were cultured with Ham's F-12 medium (Irvine Scientific, Santa Ana, CA) containing 10 mM HEPES (Mediatech, Manassas, VA), 1 mM sodium pyruvate (Mediatech), 10 mM nicotinamide (Sigma-Aldrich, St. Louis, MO), and 10% FBS (Atlanta Biologicals, Lawrenceville, GA) in a 37°C incubator with 5% CO₂.

Small interfering RNA (siRNA) transfection to INS-1 cells and rat islets.

A rat insulinoma cell line, INS-1 (a gift from Dr. Ian Sweet, University of Washington), was maintained in RPMI 1640 medium (Mediatech, Manassas, VA) supplemented with 10% FBS, 10 mM HEPES, 1 mM sodium pyruvate, 50 μM β-mercaptoethanol (Sigma-Aldrich), and antibiotic-antimycotic (Gibco, Life Technologies, Carlsbad, CA). Three days prior to transfection, INS-1 cells were cultured in a 24-well plate (Corning Life Sciences, Tewksbury, MA) (70 × 10³ cells/well) with INS-1 culture media not containing antibiotic-antimycotic. INS-1 cells and islets isolated from the pancreata of Firefly rats were transfected with 50 nM of Bbc3 siRNA or 50 nM of a nontargeting scrambled control siRNA (SMARTpool siRNA, GE Dharmacon), using the DharmaFECT transfection reagent (GE Dharmacon) as described previously (31). The transfection rate was confirmed to be >80% in the INS-1 cells and >25% in the islets. The transfected cells were cultured for an additional 2 days prior to cold preservation.

MTT assay for cold-preserved INS-1 cells.

INS-1 cells transfected with Bbc3 siRNA or control siRNA were cold preserved for 16 h in 1 ml of UW solution in a 1.8 ml microcentrifuge tube and stored at 4°C. After cold preservation, the viable cell number was counted using trypan blue staining, and 20,000 cells/well were cultured in a 96-well plate with RPMI 1640 medium containing 10% FBS for 4 h in a 37°C tissue culture incubator. The cell viability was determined using the 3-(4,5-dimethylthiazolyl-2) 2,5-diphenyltetrazolium bromide (MTT) assay (Sigma-Aldrich) (26) after rewarming. The viability of Bbc3 siRNA-transfected and cold-preserved INS-1 cells was expressed as the MTT absorbance relative to the MTT absorbance of the control siRNA-transfected, cold-preserved INS-1 cells.

Bioluminescence imaging of Firefly rat islets.

The metabolic activity of freshly isolated or cold-preserved Firefly rat islets, with or without siRNA transfection, was determined in vitro by bioluminescence imaging. Medium containing 0.75 mg/ml d-luciferin bioluminescent substrate (PerkinElmer, Waltham, MA) was added to each well of a 24-well plate containing 60 hand-picked islets. The luminescence intensity was measured using the Xenogen imaging system (Caliper Life Sciences, Hopkinton, MA). The photon counts of Bbc3 siRNA-transfected islets after 16 h of cold preservation followed by 6 h of rewarming were compared with similarly treated control siRNA-transfected islets.

mRNA analysis by RT PCR.

Total RNA was extracted from islets or INS-1 cells using Tri Reagent (Molecular Research Center, Cincinnati, OH) and Direct-Zol RNA MicroPrep (Zymo Research). cDNA was synthesized from 1.0 μg of RNA from each sample using the Maxima First Strand cDNA Synthesis Kit (Thermo Fisher Scientific, Waltham, MA) and the GeneAmp PCR System 9700 (Life Technologies, Carlsbad, CA). The relative expression of BBC3/Bbc3 mRNA and β-actin mRNA was obtained using TaqMan Gene Expression Assays (Life Technologies) for rat Bbc3 (Rn00597992_m1), human BBC3 (Hs00248075_m1), rat ACTB (endogenous control, 4352340E), and human ACTB (4326315E) with 20× TaqMan Gene Expression Master Mix (Life Technologies) in a PCR 7500 System (Life Technologies). The data was analyzed using the 2^−ΔΔC_T method.

Mouse en bloc pancreas preparation, cold preservation, and islet isolation.

Under anesthesia, the pancreata of Bbc3-KO mice or WT mice (C57BL/6) were harvested en bloc with the spleen, aorta, vena cava, and a part of the intestine. Immediately after removal of the pancreas, 0.3 ml of cold UW solution was perfused through the aorta, and the tissues were cold-preserved on ice for 8 h in a petri dish containing UW solution. After cold preservation, collagenase was injected through the pancreatic duct and islets were isolated, as described above. Isolated islets were hand-picked for each assay and cultured in the wells of 24-well or six-well plates (Sarstedt, Newton, NC) for overnight culture/rewarming with Ham's F-12 medium containing 10 mM HEPES and 10% FBS in a 37°C incubator with 5% CO₂.

Islet morphology and viability assessment.

Forty hand-picked islets from each group were cultured overnight and stained with 15 μg/ml propidium iodide and 0.25 μM fluorescein diacetate (Sigma-Aldrich) in a 1.8-ml microcentrifuge tube. Five minutes after the incubation, the staining reagent was removed and the islets were washed with PBS containing Ca⁺⁺ and Mg⁺⁺ (Mediatech). The islet suspension was placed in a 96-well plate, and the viability was assessed using a fluorescent microscope (IX50; Olympus, Center Valley, PA). The images of the entire area of the well were merged, and the viability was obtained by calculating the percentage of the area stained with propidium iodide (red) divided by the area stained with fluorescein diacetate (green) using the CellSens 1.12 platform (Olympus) (18). After examining the islet viability, islets were stained with dithizone (diphenyl thiocarbazone) solution (Sigma-Aldrich) to confirm islet morphology and purity.

Western blot.

Digested tissues or cultured islets were washed twice with ice-cold PBS containing phosphate inhibitors (PhosSTOP phosphatase inhibitor cocktail tablets; Roche Diagnostics, Indianapolis, IN) and protease inhibitors (cOmplete protease inhibitor cocktail tablets; Roche Diagnostics). The total cell lysate was extracted, and a sample containing ∼10 μg of protein was used for Western blotting, as described previously (33). Antibodies directed against Bbc3 (no. 14570, rodent specific), BBC3 (no. 4976), cleaved caspase-3 (no. 9661), HMGB1 (no. 3935), β-actin (no. 4970), and anti-rabbit IgG HRP-linked antibody (no. 7074) were obtained from Cell Signaling Technology (Beverly, MA). For oxidative stress detection, the antibody against 4-hydroxynonenal (HNE; STA-035) was obtained from Cell Biolabs (San Diego, CA). The images were captured and quantitated using a GS-900 calibrated densitometer with Image Lab Software (Bio-Rad Laboratories, Hercules, CA).

HMGB1 ELISA.

Fifty isolated islets were cultured with 300 μl of Ham's F-12 culture media for 18 h. Medium supernatants from each well were collected for the HMGB1 ELISA (HMGB1 detection assay kit; Chondrex, Redmond, WA), which was conducted according to the manufacturer's instructions.

In vitro insulin release assay in a perifusion system.

One hundred islets were cultured overnight, and then the medium was changed to RPMI 1640 medium containing 3.3 mM glucose and 10% FBS and cultured for an additional 4 h prior to the assay. Insulin release in response to high glucose stimulation was examined using a perifusion system (32). In brief, the islets were placed between Cytodex beads (Cytodex 1; GE Healthcare Bio-Sciences, Uppsala, Sweden) in a small column and perifused with 37°C Krebs-Ringer buffer containing 3 mM glucose (basal buffer) for 15 min, followed by 17 mM glucose (stimulation buffer) for 15 min, and then returned to basal buffer. The effluent was collected in 1-min fractions (0.4 ml), and the insulin contents of each effluent were measured using an ELISA kit for mouse insulin (Mercodia, Winston Salem, NC). The area under the curve (AUC) was used to calculate the amount of insulin release: the area during the first 15 min as total insulin release to basal buffer and the next 15 min as total insulin release to stimulation buffer. The first peak insulin release in response to high-glucose stimulation was calculated by insulin release obtained by the AUC during the first 5 min of high-glucose buffer subtracted by the average basal insulin release in the same time period.

Samples collected during human pancreas preservation and islet isolation.

Samples were collected from donor human pancreata as well as at various time points during islet isolation (n = 3). All pancreata used had appropriate research consent. The use of human tissue was approved by the Institutional Review Board of the Beckman Research Institute at the City of Hope. Islets were isolated, following the standard operation procedures of the Southern California Islet Cell Resources Center (SC-ICRC). Pancreas biopsies were performed 1) before and 2) after cold preservation; samples of digested pancreatic tissues were collected 3) at the end of warm collagenase digestion (postdigestion) and 4) before islet purification (prepurification), and samples of isolated islets were obtained 5) following 2 days in culture. Collected samples were snap-frozen and stored at −80°C. The donor pancreata used in this study were stored in a specialty container designed to submerge the pancreas between the UW solution and a layer of oxygenated perfluorocarbon (2-layer method) on ice. An average cold ischemia of the pancreata used for this study was 380 ± 53 min, with a warm digestion time of 15 ± 1 min, followed by a cold digestion time of 41 ± 4 min. The median donor age was 44 ± 3 yr, with brain death duration of 1 ± 0 day and no warm ischemia. Protein was extracted from each sample, and BBC3 and β-actin were assessed by Western blot by the method described above.

Human islet cold preservation and culture.

Human islets were obtained through the City of Hope islet distribution program after 2–5 days in culture, following the Integrated Islet Distribution Program (IIDP) protocol. Islets were cold preserved in cryogenic vials (Thermo Scientific, Rochester, NY) in UW solution with or without exposure to hypoxic conditions (1% O₂ with 94% N₂ containing 5% CO₂) and maintained for ≤48 h on ice in a cold (4°C) room. After cold preservation, islets were transferred to a six-well plate and rewarmed in Ham's F-12 medium containing 10 mM HEPES and 10% FBS in a 37°C incubator containing 5% CO₂. In some experiments, the islets were placed in a petri dish with culture medium and stored in a chamber under hypoxic conditions at 37°C. Islet samples were collected for RNA for RT-PCR, protein extraction for Western blot, and islet viability tests as described above.

Statistical analysis.

The results are presented as means ± SE. Unpaired t-tests were used, unless specified as paired t-test, to compare the two variances. P < 0.05 was considered to be statistically significant.

RESULTS

Cold preservation/rewarming induces islet injury and the upregulation of Bbc3 mRNA expression.

The effect of cold preservation/rewarming on the metabolic activity of isolated islets was examined using Firefly rat islets and bioluminescence imaging. Since light emission of luciferase-positive cells requires ATP and oxygen, the luciferase-luciferin reaction can be used to determine the metabolic activity and ATP levels in the cells (13, 29). The photon intensity of cold-preserved islets has been shown to correlate with islet viability (29). In this study, the intensity of bioluminescence decreased during the 6 h of rewarming after 18 h of cold preservation, whereas no change was observed in islets without cold preservation (Fig. 1A). The bioluminescent intensity of each sample was compared with islets without cold preservation/culture (control, 100%) for further quantification. The bioluminescent intensity after 18-h cold preservation was similar to the prepreservation level (99 ± 16%), which was also similar to that measured in islets kept in culture without cold preservation. This result suggested that the ATP level was maintained during cold preservation (Fig. 1B). In the subsequent 6-h culture period, the bioluminescent intensity decreased to 35 ± 4% of the pre-cold preservation level, which indicated that a reduction in ATP and/or mitochondrial damage occurred in the islets during rewarming (Fig. 1B).

Fig. 1.

Bbc3-dependent decrease in mitochondrial activity after cold preservation/rewarming. A and B: bioluminescent imaging of “Firefly” rat islets after 18 h of cold preservation (top) or freshly isolated islets (bottom) at 0, 2, and 6 h of rewarming/culture at 37°C (A) and quantitation of luminescence at 0- and 6-h rewarming of islets after 18 h of cold preservation (black bars) or freshly isolated islets (open bars) (B). The photon counts of islets at each time point were normalized with those of the corresponding wells measured at the beginning of the experiment (pre-cold preservation, n = 3; *P < 0.05 and ††P < 0.01 by paired t-test). C: changes in Bbc3 mRNA expression after 24 h of cold preservation followed by an additional 24 h in culture (□), as measured by RT-PCR (n = 3; †P < 0.05 by paired t-test). Control islets were maintained in culture for the same period, ≤48 h (○). D and F: Bbc3 mRNA (%Actb) expression levels 2 days after transfection of Bbc3 siRNA or control siRNA to INS-1 cells (D) or isolated Firefly rat islets (F) measured by RT-PCR (n = 3; *P < 0.05). E: viability of INS-1 cells transfected with Bbc3 siRNA or control siRNA after 12 h of cold preservation, followed by 4 h of rewarming as assessed by 3-(4,5-dimethylthiazolyl-2) 2,5-diphenyltetrazolium bromide assay (n = 3; *P < 0.05). G and H: bioluminescent imaging of Firefly rat islets transfected with Bbc3 siRNA or control siRNA after 18 h of cold preservation, followed by 6 h of rewarming (n = 3; *P < 0.05).

To elucidate the relationship between cold preservation-induced islet injury and the proapoptotic gene Bbc3, we measured changes in Bbc3 mRNA expression levels in islets before and after cold preservation as well as after rewarming using RT-PCR. The Bbc3 mRNA expression did not change after 24 h of cold preservation, but it did show a twofold increase after 24 h of rewarming, whereas the Bbc3 mRNA levels did not change in the non-cold-preserved control islets during 48 h in culture (Fig. 1C).

Bbc3 siRNA transfection reduces cold preservation/rewarming injury in both INS-1 cells and isolated rat islets.

To further confirm the involvement of Bbc3 in cold preservation/rewarming injury, Bbc3 siRNA or a nontargeting control siRNA was transfected into INS-1 cells or isolated Firefly rat islets prior to cold preservation (Fig. 1, D and F). Bbc3 siRNA-transfected INS-1 cells showed better viability after cold preservation compared with nontargeting siRNA-transfected control cells, as measured by the MTT assay (Fig. 1E). In Bbc3 siRNA-transfected islets, the metabolic activity after 18-h cold preservation was significantly higher than that in control siRNA-transfected islets, as assessed by bioluminescence after 6 h of rewarming (Fig. 1, G and H). These results again confirmed the involvement of Bbc3 in cold preservation/rewarming-induced islet/β-cell injury.

Bbc3-KO mouse islets are less susceptible to cold preservation/rewarming injury than WT mouse islets.

The aim of this experiment was to further verify the involvement of Bbc3 in islet injury caused by pancreas cold preservation and rewarming/islet culture. To perform this study, an en bloc mouse pancreas harvest method was established to cold preserve the whole pancreas prior to islet isolation. Compared with islets isolated from freshly harvested non-cold-preserved WT mouse pancreata (fresh WT), the postisolation islet number significantly decreased after 8 h of cold preservation in both WT (8-h cold WT) and Bbc3-KO mice (8-h cold Bbc3-KO) (Fig. 2A). Furthermore, islets from the 8-h cold WT pancreata showed significantly lower viability than those from the fresh WT pancreata. In contrast, the loss of viability mediated by pancreas cold preservation was significantly lower in 8-h cold Bbc3-KO islets than 8-h cold WT islets, which maintained a level similar to that of fresh WT islets (Fig. 2, B and C).

Fig. 2.

Yield and viability of islets isolated from freshly harvested or cold-preserved en bloc pancreata. Islets were isolated from pancreata of freshly harvested wild-type (WT; C57BL/6J), 8-h cold-preserved WT, or Bbc3-knockout (KO) mice. A: postisolation islet number (n = 5). B: representative images of dithizone (diphenyl thiocarbazone) staining (top) and viability with fluorescent diacetate and propidium iodide (bottom) of islets after overnight culture/rewarming (bar, 200 μm). C: quantitation of islet viability after overnight culture by imaging analysis in (20 islets/sample in triplicate; n = 3). *P < 0.05; **P < 0.01.

Deletion of Bbc3 in collagenase-digested pancreatic tissue and isolated islets is associated with reduced HMGB1.

To examine the mechanism associated with cold preservation-induced islet death, cleaved caspase-3 and HMGB1 expression levels were examined in both digested pancreatic tissues and cultured islets by Western blot analysis. The pancreas cold preservation/rewarming injury first occured during warm enzymatic digestion; however, it is difficult to assess islet-specific cell death at this stage, as digested tissue contains a large amount of exocrine cells, and only 1–2% is islet tissue. Unexpectedly, caspase-3 activation was not detected in any of the samples (data not shown). In contrast, HMBG1, a known marker for necrosis, was increased by cold preservation of the pancreas (Fig. 3, A–D). The levels of HMGB1 in both digested pancreatic tissue and isolated islets from the 8-h cold Bbc3-KO group were significantly lower compared with the 8-h cold WT group (Fig. 3, A–D). The expression of HMGB1 in the digested pancreatic tissues and cultured islets showed the same trend as Bbc3 expression (Fig. 3, A–F). HMGB1 in the culture medium released from 8-h cold Bbc3-KO islets was significantly reduced compared with the both fresh WT and 8-h cold WT islets (Fig. 3G).

Fig. 3.

Intracellular HMGB1 (high-mobility group box 1 protein) expression and HMGB1 release from cultured islets into the medium. HMGB1, Bbc3, and β-actin expression in the total cell extract of the digested pancreatic tissues and overnight-cultured islets from fresh WT, 8-h cold WT, and 8-h cold Bbc3 pancreata was examined using Western blot analysis. Representative images of HMGB1, Bbc3, and actin. A and B: digested pancreatic tissue (A) and overnight-cultured islets (B). Quantitated HMGB1 and Bbc3 levels by densitmetry analysis: C and D: HMGB1 in digested pancreatic tissues (C) and in isolated and cultured islets (D). E and F: Bbc3 levels in digested pancreatic tissues (E) and in isolated and cultured islets (F). G: HMGB1 release during 18 h of islet culture (50 islets in 300 μl of medium) detected by the HMGB1 ELISA (n = 6). *P < 0.05.

Islets from cold-preserved Bbc3-KO pancreata show reduced oxidative stress.

Involvement of oxidative stress in islet injury mediated by pancreatic cold preservation was examined through the detection of 4-HNE protein adducts by the Western blot (Fig. 4). 4-HNE is produced by lipid peroxidation and is used as a marker for the presence of oxidative stress. The total levels of 4-HNE protein adducts were lower in 8-h cold Bbc3-KO islets than in 8-h cold WT islets (Fig. 4, B and D), whereas there was no difference in digested pancreatic tissues between the groups (Fig. 4, A and C).

Fig. 4.

Detection of oxidative stress by 4-hydroxynonenal (4-HNE) protein adducts. Representative images of 4-HNE protein adducts and β-actin in the total cell extract of digested pancreatic tissues (A) and overnight-cultured islets (B) from fresh WT, 8-h cold WT, and 8-h cold Bbc3-KO pancreata examined by Western blot analysis. C and D: quantification of 4-HNE protein adducts; all detected proteins by the antibody between 25 and 130 kDa in each lane were quantitated (n = 3). *P < 0.05.

Islet function is preserved in Bbc3-KO islets after cold preservation and rewarming.

The function of islets isolated from WT and Bbc3-KO mouse pancreata cold preserved for 8 h was measured by glucose-stimulated insulin release in a perifusion system. Both WT and Bbc3-KO islets responded to high glucose stimulation, but the first peak of insulin release was significantly lower in 8-h cold WT islets (Fig. 5B). Since the reduction of the first-phase insulin release to high glucose is considered as an indication of early islet damage (5), this may suggest increased dysfunction in cold-preserved WT islets. However, there was no significant difference in the total insulin release between 8-h cold WT islets and 8-h cold Bbc3-KO islets during the 15 min of low and high glucose perfusion (Fig. 5C).

Fig. 5.

Dynamic glucose-stimulated insulin release response was monitored using a perifusion system. Insulin released in response to low glucose (3 mM) and high glucose (17 mM) was measured by mouse insulin ELISA. A: average insulin release curve from 8-h cold WT islets (○; n = 3) and 8-h cold Bbc3-KO islets (■; n = 3). B: total amount of insulin released under the curve (AUC) of the first peak insulin release during the first 5 min after changing to high-glucose stimulation buffer (n = 3). C: AUC of the total insulin release in low-glucose buffer (15 min) and in high-glucose buffer (subsequent 15 min) from 8-h cold WT islets (open bars) and 8-h cold Bbc3-KO islets (black bars) (n = 3). *P < 0.05.

Upregulation of BBC3 mRNA and protein is associated with the reduced viability of human islets exposed to cold hypoxia and rewarming.

We further examined BBC3 involvement in human pancreas cold preservation, islet isolation, and culture/rewarming. The BBC3 protein level significantly increased in the tissue samples collected after warm digestion compared with that of the pancreatic tissue collected before cold preservation. The BBC3 level was far lower in isolated and cultured islets than in the digested tissues, indicating that cells other than islets, mostly exocrine acinar cells, express BBC3 in response to rewarming during the tissue digestion process (Fig. 6A). To examine the effect of cold preservation and rewarming on BBC3 expression, isolated human islets were cold preserved in either hypoxic or normoxic conditions, followed by rewarming at 37°C in culture. BBC3 mRNA did not increase in islets during the 24-h cold preservation under either normoxic or hypoxic conditions (Fig. 6B). Rewarming at 37°C after cold preservation increased BBC3 mRNA when islets were cold preserved under hypoxic conditions but showed no increase following the normoxic condition (Fig. 6B). BBC3 protein was also upregulated in isolated islets after 24-h cold preservation in hypoxic conditions, followed by culture at 37°C for 24 h, compared with isolated islets prior to preservation or those only cultured at 37°C for the same period. In contrast, BBC3 protein did not increase when islets were cold preserved under normoxic condition (Fig. 6C). Islet viability after cold preservation and rewarming decreased significantly compared with prepreservation, whereas hypoxia during cold preservation further augmented this cell death (Fig. 6D). As others have reported previously, hypoxia induces Bbc3 in the mouse pancreatic β-cell line MIN6 (45). We found an increase in both BBC3 mRNA and BBC3 protein in human islets cultured under hypoxia for 48 h (4.1 ± 1.3- and 4.4 ± 1.0-fold increase compared with preculture, respectively) and a great decrease in viability. However, a prolonged hypoxic cold preservation of ≤48 h did not increase either BBC3 mRNA or BBC3 protein expression (0.68 ± 0.1- and 1.25 ± 0.5-fold increase compared with preculture, respectively) but did increase cell death, suggesting that BBC3 is less likely to be involved in cell death under such conditions.

Fig. 6.

BBC3 upregulation after cold preservation/rewarming of human pancreas and isolated human islets. A: BBC3 protein expression in pancreatic tissue immediately after pancreas harvest prior to cold preservation (lane 1), after cold preservation (lane 2), in digested pancreatic tissue collected immediately after digestion (lane 3), before purification (lane 4) assessed by the Western blot, and in isolated islets after 2 days in culture (lane 5) assessed by the separate Western blot experiments (n = 3; *P < 0.05). B: BBC3 mRNA expression levels in isolated islets before and after 24-h cold preservation with or without hypoxic condition and 24-h rewarming in culture at 37°C (n = 3). *P < 0.05, statistical significance by unpaired Student's t-test; †P < 0.05, data analyzed by paired t-test; ††P < 0.01 by paired t-test. C: BBC3 protein expression in isolated islets before cold preservation (lane 1), after 48 h of culture at 37°C (lane 2), during 48 h of hypoxic culture at 37°C (lane 3), during 24 h of cold preservation followed by 24 h re-warming in culture at 37°C (lane 4), and during 24-h cold hypoxic preservation followed by 24 h of rewarming in culture at 37°C (n = 3, *P < 0.05 or †P < 0.05 by paired t-test). D: representative images of islets stained for viability with fluorescent diacetate and propidium iodide (bar, 1 mm) and viability quantitated by image analysis (n = 3; *P < 0.05 or †P < 0.05 by paired t-test).

DISCUSSION

We made three major observations in this study. First, Bbc3 contributes to cold preservation/rewarming injury in pancreatic islets. This is shown by the improved mitochondrial viability and reduction of postculture islet death by silencing Bbc3 in rat islets and in Bbc3-KO mouse islets. However, knocking out Bbc3 has only a minimal effect on the postisolation islet yield. Second, knocking out Bbc3 is associated with decreased cold preservation/rewarming-induced HMGB1 expression levels, which might explain the reduction in oxidative stress in the islets and the improvement in islet viability and function. Third, BBC3 protein increases after cold preservation/rewarming of human pancreata as well as isolated islets. Hypoxia during cold preservation elevates BBC3 mRNA and protein levels after rewarming and increases islet cell death.

Bbc3 is known to be induced by hypoxia and oxidative stress in cultured islets, contributing to β-cell apoptosis (11, 42, 45). Our study has shown that cold preservation of human islets under hypoxic conditions does not change BBC3 mRNA expression, but rather, the subsequent rewarming in culture increases BBC3 mRNA. This may indicate that the hypoxic event that occurs during cold preservation is triggered and/or amplified to upregulate BBC3 mRNA when the metabolic activity of the cell is restored at 37°C. Our results indicate that different mechanisms exist in the islet rewarming injury that takes place during islet isolation vs. islet culture. BBC3/Bbc3 plays a role in islet death during islet culture, but to a lesser extent during islet isolation. Cold storage-induced cell death is known to be primarily necrotic; however, rewarming injury is associated with apoptosis in the presence of ongoing necrosis in renal cell preservation (37). The ratio of apoptotic to necrotic cell death has been shown to depend on the duration of preservation and the cell type or experimental model (4, 37), which may explain the limited involvement of Bbc3 in improving the postisolation islet yield observed in our study.

HMGB1 released from the cell has been shown to be a marker for injury in human kidney, liver, and experimental islet transplantation (12, 15, 19, 25). We have shown that knocking out Bbc3 significantly decreases the intracellular HMGB1 protein levels in the digested pancreatic tissues and isolated islets after cold preservation as well as HMGB1 secretion into the culture media compared with both fresh and 8-h cold WT groups (Fig. 3). It has been shown that islets cultured under hypoxia for 12 h increased cytoplasmic HMGB1 staining as well as cellular HMGB1 mRNA and HMGB1 protein release in the media (15). The mechanism of HMGB1 reduction in the Bbc3-KO is not known. Although an interaction between HMGB1 and transcription factor p53 has been reported (1), and Bbc3 has been shown to be upregulated in a p53-dependent manner (16), additional studies are required to elucidate the interaction between HMGB1 and Bbc3.

HMGB1 is closely associated with oxidative stress. HMGB1 is actively released from monocytes and macrophages affected by oxidative stress or passively from cells undergoing necrosis. Released HMGB1 also activates inflammatory signaling pathways, thereby increasing cellular reactive oxygen species (40) and accelerating cell death. Several studies have indicated that islets contain lower levels of antioxidant enzymes; therefore, islets are extremely susceptible to oxidative stress, which can lead to islet death (41). We have shown reduced oxidative stress in cold preservation/rewarming-treated Bbc3-KO islets by examining 4-HNE protein adducts. Increased 4-HNE was shown previously to be associated with islet destruction (39). The reduction of 4-HNE protein adducts in cold preservation/rewarming-treated Bbc3-KO islets may be secondary to the reduced HMGB1 level and contribute to the improved postculture islet viability and function shown by the increased first-phase insulin release.

We have shown the increased mitochondrial viability by silencing Bbc3 siRNA in cold-preserved rat islets and INS-1 cells as well as the reduction of oxidative stress in islets isolated from cold-preserved Bbc3-KO pancreata. Bbc3 has been shown to be a dominant regulator of oxidative stress-induced mitochondrial damage and induces apoptosis in neuronal cells (38). In islets, several studies have shown that β-cell death is induced not by Bbc3 alone but together with Bim (6, 42). Depending on the cell type, stress model, and species, the mechanism of BBC3/Bbc3-mediated cell death may be different. Similar studies using Bbc3 and Bim double-KO mice may further improve islet viability after cold preservation/rewarming.

We have demonstrated the upregulation of BBC3 protein during islet isolation. However, the specific effect of BBC3 on human islet death mediated by pancreas cold preservation and islet isolation was not examined, tested, or determined due to the technical limitations of silencing BBC3 gene in donor pancreata prior to cold preservation. It is possible that BBC3 expression in donor pancreata is elevated due to the pathological changes and elevation of inflammatory cytokines occurring in a brain-dead donor following brain death (3, 35). In addition, multiple other donor factors, such as age, cause of death, hospitalization, brain death duration, and medical history, that may increase proinflammatory and oxidative reactions are also potential factors leading to BBC3-mediated cell death. These factors were not present in our Bbc3-KO mouse model. Therefore, it is possible that BBC3-mediated islet death in human islets during pancreas cold preservation and islet isolation may play more a significant role than what we have shown in the mouse model. Furthermore, the presence of increased ER stress in human islets during islet isolation has been reported (28), which may also influence BBC3-mediated islet death. It has been reported that several conditions, such as ischemic preconditioning of rat pancreata (10), ductal injection of a p38 mitogen-activated protein kinase inhibitor in canine pancreata (14), and c-Jun NH₂-terminal kinase inhibitor in porcine pancreata (30), improve islet yield and decrease apoptosis of isolated islets. How these treatments affect on Bbc3/BBC3-mediated islet death is not known.

In summary, we have demonstrated the involvement of BBC3/Bbc3 in cold preservation/rewarming-mediated islet injury in both human and rodent pancreata/islets. BBC3/Bbc3 is shown to influence the viability of islets after rewarming in culture but does not improve islet yield after isolation, indicating that different mechanisms are involved in the loss of cultured islets. This study also demonstrated a possible interaction between Bbc3, HMGB1, and oxidative stress. Further studies are necessary to elucidate the mechanism of cold ischemia/rewarming-induced islet death and contribute to the development of new strategies aimed at increasing the transplantable islet mass and function, thus improving transplant success.

GRANTS

This study was supported by Nora Eccles Treadwell Foundation and National Institute of Diabetes and Digestive and Kidney Diseases Grant DK-098446.

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the authors.

AUTHOR CONTRIBUTIONS

K.O., E.K., G.A., F.K., M.T., and Y.M. conception and design of research; K.O., E.K., H.K., J.R., G.A., M.P., A.R.O., L.V., and I.H.A.-A. performed experiments; K.O., H.K., and K.F. analyzed data; K.O., E.K., K.F., I.H.A.-A., F.K., M.T., and Y.M. interpreted results of experiments; K.O., M.P., and K.F. prepared figures; K.O. drafted manuscript; K.O., E.K., J.R., M.T., and Y.M. edited and revised manuscript; K.O., E.K., H.K., J.R., G.A., M.P., A.R.O., L.V., K.F., I.H.A.-A., F.K., M.T., and Y.M. approved final version of manuscript.