| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
ARTILCE |
B Translocation Mediates Double-Stranded Ribonucleic Acid-Induced NIT-1 ß-Cell Apoptosis and Up-Regulates Caspase-12 and Tumor Necrosis Factor Receptor-Associated Ligand (TRAIL)
Department of Pathology (M.A.R., J.K.C.), University of British Columbia, and British Columbia Research Institute for Children’s and Women’s Health (M.A.R., L.M., E.P., J.K.C.), Vancouver, British Columbia, Canada V5Z 4H4
Address all correspondence and requests for reprints to: Janet K. Chantler, Ph.D., British Columbia Research Institute for Children’s and Women’s Health, Room 318, 950 West 28th Avenue, Vancouver, British Columbia, Canada V5Z 4H4. E-mail: chantler{at}interchange.ubc.ca.
Abstract
The mechanism of induction of apoptosis by double-stranded RNA (dsRNA) is not fully characterized. The dsRNA is normally present in extremely low quantities in cells, but following infection with RNA viruses, large quantities of the dsRNA viral replicative intermediate may be produced triggering the antiviral response as well as cell death. In this report, transfection of polyinosinic-polycytidylic acid [poly(I:C)] into NIT 1 cells has been used as a model of intracellular dsRNA-induced ß-cell apoptosis. At 18 h post transfection, 45% of the cells were apoptotic as indicated by terminal deoxynucleotidyl transferase-mediated deoxyuridine triphosphate nick end-labeling (TUNEL) staining, and this was accompanied by an increase in nuclear factor 
B (NF-B) p50/p65 nuclear translocation and cleavage of caspases 3 and 8. The NF-B inhibitor peptide, SN50, significantly reduced caspase-3 activity and the percentage of TUNEL-positive cells, substantiating a role for NF-B in inducing intracellular dsRNA-mediated apoptosis. Concomitantly, RNA-dependent protein kinase activity was observed at 3 h post transfection along with phosphorylation and degradation of inhibitory B-
. Expression of TRAIL (TNF-related apoptosis-inducing ligand), Fas, IL-15, and caspase-12 mRNAs was up-regulated in the presence of poly(I:C) but not when SN50 was also added. In contrast, there was no change detected in Fas, Fas-associated death domain, Bcl-2, Bcl-xl, Bax, p53, or XIAP(X-linked inhibitor of apoptosis protein) expression up to 12 h after poly(I:C) transfection. In addition, caspase-12 was cleaved, and phosphorylation of eukaryotic initiation factor 2 occurred, suggesting that an endoplasmic reticulum stress pathway was involved in addition to NF-B induction of an extrinsic pathway, possibly mediated by TNF-related apoptosis-inducing ligand.
PANCREATIC ß-CELL death is the fundamental cause of type 1 diabetes, and various models have been used to study the mechanisms by which it occurs. Under natural conditions, autoimmune killing of ß-cells by self-reactive T cells is thought to be the most important mechanism, and both viruses (1, 2, 3) and cytokines (4, 5) are known to be capable of causing direct damage to ß-cells both in vivo and in vitro.
The model for ß-cell death reported here involves transfection of a murine ß-cell line (NIT-1) with the synthetic double-stranded RNA (dsRNA) molecule, polyinosinic: polycytidylic acid [poly(I:C)], an efficient trigger of cellular antiviral responses and cell death (6, 7). dsRNA is a biologically active molecule that is highly resistant to degradation by cellular ribonucleases (8). It is normally present in cells in extremely small amounts (estimated at 0.5% of total cell RNA), located in the nucleus in HnRNA and mitochondria (6, 8). However, during infection with RNA viruses, 50-fold increases in the cytoplasmic content of dsRNA have been detected, mainly representing the replicative intermediate formed during viral RNA replication (6). This presence of foreign dsRNA in cells, a signal that can be mimicked by transfecting cells with poly(I:C), triggers a rapid response associated with interferon induction and cellular apoptosis.
There has been considerable research on the intermediates involved in dsRNA-triggered cell death, and several of the key components of the cellular response have been identified. These include RNA-dependent protein kinase (PKR), a dsRNA-dependent protein kinase, and nuclear factor B (NF-B), a family of transcription factors activated by PKR. The importance of PKR in the antiviral response is highlighted by the fact that many viruses have been shown to encode proteins that inhibit its activity (7, 9). In addition, PKR null cells are resistant to dsRNA-induced apoptosis (10). Among its many activities, PKR phosphorylates eukaryotic initiation factor 2 (eIF2), resulting in inhibition of protein synthesis, likely a contributing factor to the induction of cell death (11). PKR also activates members of the NF-B family of transcription factors, by enhancing phosphorylation of their inhibitors, the inhibitory Bs (IBs) (12). Additionally, it has recently been shown that dsRNA-induced apoptosis of rat islets is PKR dependent (13).
The NF-B transcription factors (homo- and heterodimers of different family members) regulate a wide range of genes that modulate the antiviral response (interferons- and -ß) as well as the induction of inflammatory cytokines like IL-1ß and TNF, chemokines, and also major histocompatibility complex molecules. In some systems, NF-Bs have been associated with cell survival and proliferation (14), and in others, including following PKR overexpression and poly(I:C) treatment, they are associated with cell death (12, 15). How this wide range of functions is controlled is not fully understood but is linked to the formation of different homodimer and heterodimer pairs of NF-Bs and to other factors with which these associate to form the transcription complex (or enhanceasome) for different genes.
Previous studies using ß-cell lines as well as dissociated and intact murine and human islets have shown that addition of high levels of poly(I:C) (100–400 µg/ml) and interferon 
(IFN) to the medium, induces nitric oxide (NO) production and cell death, whereas poly(I:C) alone is ineffective (13, 16, 17). Aside from the extensive cell death via NO production after treatment with extracellular dsRNA and IFN, low levels of apoptosis have been seen in these systems. The effects of extracellular dsRNA are thought to be mediated through Toll-like receptor 3 (TLR3) leading to NF-B activation (18), but intracellular dsRNA interacts with a variety of cell proteins including PKR, ribonuclease L, and PKR-activating protein (PACT) (19, 20).
The present study was carried out to investigate the effects of introducing foreign dsRNA [poly(I:C)] by transfection into NIT-1 ß cells in the absence of exogenous cytokines. The results show that low concentrations of intracellular poly(I:C) induce NIT-1 ß cell death by this method and that NF-B translocation is a key factor in triggering apoptosis in a process associated with caspase-3, -8, and -12 cleavage as well as transcriptional induction of caspase-12, Fas, IL-15, and the TNF receptor-associated ligand (TRAIL).
Materials and Methods
Materials and cells
NIT-1 cells (American Type Culture Collection, Manassas, VA) were cultured in F12/K medium supplemented with 10% (vol/vol) heat-inactivated fetal bovine serum and 1% penicillin, streptomycin, neomycin antibiotic. J774 cells were cultured in RPMI 1640 medium supplemented with 10% heat-inactivated fetal bovine serum, 1% penicillin, streptomycin, neomycin antibiotic, 20 mM L-glutamine, and 1 mM sodium pyruvate. Cells were either trypsinized (J774) or treated with cell dissociation buffer (NIT-1, Gibco, Gaithersburg, MD), and diluted in complete media without antibiotics and seeded at 2 x 105 cells/ml 24 h before addition of poly(I:C) (Sigma, St. Louis, MO). All media, supplements and Lipofectamine 2000 (LF 2000) were obtained from Invitrogen (Carlsbad, CA).
Antibodies used were donkey antirabbit IgG [heavy and light chain (H+L)] peroxidase conjugated (Amersham Corp.); NF-B p50 and p65, anti-Bcl-2 and anti-PKR(B10) (Santa Cruz Biotechnology, Inc., Santa Cruz, CA); anti-poly(ADP-ribose) polymerase (PARP) (R&D Systems, Minneapolis, MN); anti-inducible nitric oxide synthase (iNOS) (Transduction Laboratories, Lexington, KY); anti-caspase-3 (p17), anti-phospho-eIF2 and eIF2, and anti-phospho-IB and IB (Cell Signaling, New England Biolabs, Beverly, MA); anti-Fas-associated death domain (FADD) (clone 1F7, Upstate Biotechnology, Lake Placid, NY); rabbit antihuman X-linked inhibitor of apoptosis protein (XIAP) (Aegera); anti-Fas (clone 13, Transduction Laboratories); anti-p53 (clone G59–12, BD PharMingen); anti-Bax (clone 6A7, BD PharMingen, San Diego, CA); anti-Bcl-x (clone 44, Transduction Laboratories); anti-caspase-12 (Chemicon, Temecula, CA); anti-dsRNA (J2; Biocenter, Inc., Szeged, Hungary) and immunoPure rabbit antigoat IgG, (H+L) and rabbit antimouse IgG (H+L), both peroxidase conjugated (Pierce, Rockford, IL).
Cell viability assays
Eighteen hours after addition of 1 µg/ml poly(I:C):LF 2000 to NIT-1 ß-cells, cells were harvested and an aliquot was examined for viability by trypan blue exclusion. Data represent the mean of three samples within a representative experiment.
A second aliquot of cells was analyzed by terminal deoxynucleotidyl transferase-mediated deoxyuridine triphosphate nick end-labeling (TUNEL) staining. Cells fixed in 4% (vol/vol) paraformaldehyde in PBS were deposited on a slide using a cytospin. The cells were then treated as described in the manufacturer’s instructions for the in situ cell death detection kit, fluorescein (Roche). A counterstain for cell nuclei [4',6'-diamino-2-phenylindole (DAPI)] was applied in mounting medium (Vector Laboratories, Burlingame, CA). Cells were visualized under UV using an Axioplan 2 Imaging Universal microscope (Carl Zeiss, Gottingen, Germany) with appropriate blue or green fluorescence filters. Percent apoptosis was calculated by dividing the number of green, TUNEL-staining cells by the total number of blue DAPI staining nuclei. At least five fields per sample containing a minimum of 100 cells were counted.
Immunohistochemistry for dsRNA
Poly(I:C) (1 µg/ml) was added to NIT-1 cells (seeded at 2 x 105 cells/chamber) in the presence of LF2000. Five hours later, cells were fixed in 4% paraformaldehyde in PBS and then washed in PBS. Cells were permeabilized in 50% acetone/50% methanol and blocked with 3% BSA in PBS. A 1:50 dilution of J2 anti-dsRNA monoclonal antibody in 3% BSA in PBS was added and the slides incubated 1 h at 37 C. After washing in PBS, antimouse peroxidase conjugate (1:500 in 3% BSA) was added, and the slides were incubated for 30 min at 37 C. After washing in PBS, diaminobenzidine tetrahydrochloride substrate (Sigma) was added for a 10-min incubation. Cells were visualized using a Axioplan 2 Imaging Universal microscope (Zeiss) under bright field. At least five fields per sample containing a minimum of 100 cells were counted.
Western blot analysis
NIT-1 cells were solubilized in 2x suspension buffer [200 mM NaCl, 20 mM Tris-HCl (pH 7.6), 2 mM EDTA (pH 8.0), 2% (vol/vol) Triton X-100] containing complete mini-protease inhibitor (Roche Molecular Biochemicals, Indianapolis, IN). Samples were centrifuged to remove cell debris and the protein concentrations of the supernatants calculated by protein assay reagent (Bio-Rad Laboratories, Hercules, CA). Proteins were separated on 12% sodium dodecyl sulfate (SDS)-polyacrylamide gels using standard conditions (21) and transferred for 1 h at 100 V to polyvinyl difluoride (PVDF) membrane (Bio-Rad Laboratories). Membranes were blocked in 5% (wt/vol) skim milk in Tris-buffered saline-Tween 20 [12.5 mM Tris-HCl (pH 7.5), 68.5 mM NaCl, 0.1% (vol/vol) Tween 20] for 1 h and incubated with primary antibody in 5% skim milk in Tris-buffered saline-Tween 20 4 C overnight. Detection of bound antibody was carried out using peroxidase-labeled secondary antibodies and chemiluminescent substrate (ECL, Amersham). Conditions for PARP antibody were according to the manufacturer’s instructions (R&D Systems).
Gel shift analysis
NIT-1 cells (3 x 105 cells/ml) were incubated for indicated times with different concentrations of poly(I:C) in complex with LF2000. Cells were harvested and stored at -70 C until processing. Nuclear extracts were prepared as described (22). Protein concentrations were calculated from the supernatants using the protein assay reagent (Bio-Rad) and stored -70 C until gel shift analysis.
The probe consisted of a double-stranded oligonucleotide containing the consensus binding sequence for NF-B (5'-AGT TGA GGG GAC TTT CCC AGG C-3') end-labeled with [-32P]ATP using T4 polynucleotide kinase (Promega, Madison, WI). Gel shift assays on nuclear protein were performed in the core gel shift assay kit (Promega) using 2 µg nuclear protein and 50,000 cpm of probe per sample.
Kinase assay
NIT-1 cells (4 x 106) were lysed in 1% Triton buffer [20 mM Tris (pH 7.5), 150 mM NaCl, 1% Triton X-100, 50 mM NaF, 5 mM EDTA, 1 mM Na3VO4, 10 µg/ml each of aprotinin, leupeptin, and pepstatin, 1 mM phenylmethylsulfonyl fluoride]. Cellular lysates (150 µg total protein) were immunoprecipitated for 2 h with 2 µg anti-PKR (B10) monoclonal antibody. Immune complexes were recovered with protein A+B agarose (Oncogene Research Products). After three washes, the first using lysis buffer, followed by lysis buffer, no detergent, and then kinase buffer, beads were eluted with SDS sample buffer and immunoprecipitates were resolved on 7.5% SDS-PAGE, transferred onto membrane and probed with anti-PKR monoclonal antibody (B10) for total PKR expression. For immune complex kinase assay, immunoprecipitated PKR was washed as above, resuspended in kinase buffer [10 mM piperazine-N,N'-bis(2-ethane sulfonic acid) (pH 7.0), 5 mM MgCl2, 0.5 mM dithiothreitol, 0.2 mM Na3VO4, 5 mM NaF] and incubated with 1 µCi 32P-ATP for 10 min at 30 C. The reactions were stopped by adding 2x SDS sample buffer, boiled for 5 min and resolved on 7.5% SDS-PAGE, dried, and bands detected by autoradiography.
Nitrite determination
Culture fluid (50 µl) was combined with 50 µl of each of the Greiss reagents [1% (wt/vol) sulfanilamide and 0.1% (wt/vol) naphthylethylenediamine dihydrochloride in 2.5% phosphoric acid] and incubated at room temperature for 10 min. The iNOS inhibitor aminoguanidine was added at 1 mM to appropriate samples 1 h before addition of poly(I:C) or lipopolysaccharide. Absorbance was measured at 540 nm, and nitrite concentrations were calculated from a sodium nitrite standard curve.
Caspase activity assays
ApoAlert caspase-3 activity assay (CLONTECH Laboratories, Palo Alto, CA) was performed according to the manufacturer’s instructions using DEVD-AFC as substrate (asp-glu-val-asp linked to the fluorochrome 7 amino 4 trifluoromethyl coumarin). Emissions were read using a fluorometer at 390 nM. Emissions of samples were compared with uninduced control cells to determine the fold increase in caspase-3 activity.
Real-time PCR and RT-PCR
Total RNA was extracted using Trizol (Invitrogen) from NIT-1 ß-cells treated with 1 µg/ml poly(I:C) for 12 h with or without addition of 100 µg/ml SN-50 peptide. A reverse transcription reaction was performed on equivalent concentrations of total RNA for each sample using SuperScript II (Invitrogen) and random hexamers according to the manufacturer’s instructions. This was followed by real-time PCR using 100-µM primers giving 60-nucleotide products specific to ß-actin (forward: 5'-CGATGCCCTGAGGCTCTTT; reverse: 5'-TGGATGCCACAGGATTCCAT) and caspase-12 (forward: 5'-TGCTGACAGCTCCTCATGGA; reverse: 5'-AACTGATCACGTGGACAAAGCTT) or murine TRAIL (forward: 5'-GCCACAGACACTTTCGGTGTT; reverse: 5'-TGATCTCATTTTGCGGAAAGAA) and Sybr Green PCR mastermix (Applied Biosystems, Foster City, CA) using the ABI Prism 7000. Samples were analyzed in triplicate with instrument settings of 60 cycles and 25 ml sample volume. For RT-PCR, reactions were incubated for 5 min at 94 C followed by addition of Taq DNA polymerase (Invitrogen) and a further 32 cycles performed at 94 C for 45 sec, 58 C for 45 sec, and 72 C for 80 sec. PCRs were performed according to the manufacturer’s instructions (Invitrogen). Primers (1 µM final) used were glyceraldehyde-3-phosphate dehydrogenase (forward: 5'-TGTTCCTACCCCCAATGTGT; reverse: 5'-TGTGAGGGAGATGCTCAGTG), IL-15 (forward: 5'-GCCATAGCCAGCTCATCTTC; reverse: 5'-GCTGTTTGCAAGGTAGAGCA), and Fas (forward: 5'-ATGCACACTCTGCGATGAAG; reverse: 5'-TTCAGGGTCATCCTGTCTCC).
Results
Poly(I:C) in the presence of LF2000 enters NIT-1 cells
Immunohistochemistry was performed to determine the proportion of ß-cells that contained poly(I:C) after transfection with LF 2000. Five hours after addition of either 1 µg/ml poly(I:C) alone or in combination with LF2000, the cells were washed, fixed, and stained with an antibody to dsRNA. At least 80% of the NIT-1 cells treated with poly(I:C) + LF2000 contained dsRNA (Fig. 1
, C and D), but cells treated with poly(I:C) alone (either 1 or 100 µg/ml) contained little or no detectable dsRNA (Fig. 1, A, B, and D). These results indicate that transfected poly(I:C) is internalized and is exerting its effects from within the cytosol in the same way as it would during viral infection.
|
) as measured by trypan blue exclusion. Higher amounts of dsRNA did not further increase this figure; therefore, 1 µg/ml was used for subsequent experiments. Cell viability was not reduced with addition of 1 or 10 µg/ml poly(I:C) in the absence of transfection reagent (data not shown). The proportion of TUNEL-positive NIT-1 cells treated with LF2000 alone was approximately 10% at 24 h (Fig. 2B), but in the presence of poly(I:C) plus LF2000, apoptosis levels were 45% (Fig. 2B), indicating that the majority of NIT-1 cell apoptosis is due to addition of poly(I:C).
|
Caspase inhibition eliminates cell death and apoptosis in NIT-1 cells transfected with poly(I:C)
The extent of cell death in NIT-1 cells treated with poly(I:C) mediated by caspase-3 was examined by addition of 100 µM of the general caspase-inhibitor zVAD to cells 1 h before the addition of 1 µg/ml poly(I:C) plus LF2000. At 24 h following poly(I:C) administration, caspase-3 activity was eliminated in the cells pretreated with zVAD (Fig. 2D
), with decreases in cell death (Fig. 2A) and reductions in apoptosis (Fig. 2B) to untreated control levels. The addition of inhibitors for caspase-8 (zIETD), and caspase-9 (zLEHD) also significantly reduced caspase-3 activity (Fig. 2D).
Caspase-3 and -8 and PARP are cleaved in NIT-1 ß-cells treated with poly(I:C)
To confirm that caspase-3 was activated in the presence of intracellular dsRNA, total protein extracts of cells treated with 0, 0.25, or 1 µg/ml poly(I:C) at 0, 3, 6, 12, and 24 h were analyzed on Western blots using an antibody specific to the 17-kDa cleaved form of caspase-3. In Fig. 3B, cleaved caspase-3 can be seen on Western blots 6 h after addition of poly(I:C). Fluorometric assays for caspase-3 activity showed that caspase-3 activity increased with time and concentration of poly(I:C) addition, confirming that the cleaved caspase-3 is active (Fig. 3A). The timing and extent of PARP cleavage closely followed that of caspase-3, with cleaved PARP apparent at 6 h post poly(I:C) treatment, and the majority of the 116-kDa PARP cleaved to a 25-kDa product by 24 h, the time of extensive reduction in cell viability (Fig. 3C). Cleaved caspase-8 was visible by 24 h.
|
B-, and eIF-2 are phosphorylated in NIT-1 cells treated with poly(I:C), PKR is phosphorylated as early as 3 h after addition of poly(I:C). Active PKR has been shown to cause the phosphorylation of the translation factor eIF2, resulting in inhibition of protein synthesis. To determine whether this occurred following poly(I:C) transfection, eIF2 phosphorylation was examined using an antibody specific for the phosphorylated form of eIF2 on Western blots of total cellular extracts from NIT-1 cells treated with poly(I:C). As early as 12 h after addition of poly(I:C) to NIT-1 cells and before the appearance of cytopathology, an increase in the phosphorylated form of eIF2 occurs (Fig. 4B). In addition, activation of PKR induces a pathway of phosphorylation events leading to the phosphorylation of IB, causing this protein to be ubiquitinated and degraded, thus freeing NF-B to translocate to the nucleus. Figure 4C shows degradation of IB as early as 3 h after poly(I:C) treatment with an accompanying increase in phosphorylated IB of 2-fold at 3 h and 4.5-fold at 12 h, as determined using antibodies specific for the phosphorylated form of this protein and densitometry (Fig. 4, C and D).
|
B is translocated to the nucleus of NIT-1 cells after addition of poly(I:C)B has been shown to be a downstream mediator of the effects of dsRNA in various cell systems, and it has been shown that PKR activation triggers NF-B nuclear translocation. We therefore examined ß cells treated with poly(I:C) for the nuclear translocation of this transcription factor. Gel shift assays for NF-B nuclear translocation were performed on nuclear extracts from NIT-1 cells treated with 0, 0.25, or 1 µg/ml poly(I:C) for 0, 3, 6, 12, or 24 h (Fig. 5A). The results indicate that NF-B was translocated to the nucleus 3 h after addition of poly(I:C), before measurable loss of cell viability occurred (Fig. 5A, lanes 5 and 6). The level of NF-B in the nucleus appeared maximal at 12 h (Fig. 5A, lanes 11 and 12), preceding the time of maximal loss of cell viability at 24 h when little intranuclear NF-B was detected (Fig. 5A, lanes 14 and 15). An excess of cold NF-B oligonucleotide eliminated the NF-B band, and addition of p50 and p65 antibodies caused a block and a supershift of the NF-B complex (Fig. 5A, lanes 17, 18, and 19, respectively) indicating that the band observed in these gel shift experiments is the p50/p65 complex of NF-B.
|
B nuclear translocation is responsible for caspase-induced apoptosis in NIT-1 cells treated with poly(I:C)B nuclear translocation is responsible for the observed cell death in ß-cells treated with dsRNA, an inhibitor of nuclear translocation of NF-B p50 (SN-50) was added to NIT-1 cells 1 h before the addition of poly(I:C). As can be seen in Fig. 5, B and C, apoptosis in the presence of 100 µM SN-50 decreased to levels of control cells. No NF-B was visible in nuclear extracts of NIT-1 cells treated with 100 µM SN-50 plus 1 µg/ml poly(I:C) 6 h post addition of dsRNA (Fig. 5D). To determine whether the nuclear translocation of NF-B is required for the activation of caspase-3 in poly(I:C)-treated NIT-1 cells, total cellular extracts of NIT-1 cells treated with SN-50 before addition of poly(I:C) were assayed by fluorescence assay using the DEVD-AFC substrate of caspase-3. As can be seen in Fig. 5E, addition of 100 µM SN-50 substantially reduced caspase-3 activity, compared with cells treated with poly(I:C) alone.
Up-regulation of proapoptotic proteins Fas, FADD, p53, or Bax is not responsible for the observed poly(I:C)-induced NIT-1 apoptosis
To identify proteins that could be involved in induction or prevention of apoptosis following addition of poly(I:C) to NIT-1 cells, Western blots to detect a number of candidate proteins were carried out in cells treated with 1 or 10 µg/ml poly(I:C) for varying times. As can be seen in Fig. 6A, no change in protein levels of FADD, p53, or Bax proteins was observed over control levels up to 12 h following poly(I:C) treatment, and Fas was undetectable. In addition, levels of the antiapoptotic proteins Bcl-2, Bcl-x, and XIAP did not increase in response to dsRNA. By 24 h after poly(I:C) transfection, when extensive apoptosis was detectable, protein translation was significantly reduced and levels of the majority of proteins were decreased. Interestingly, although the antiapoptotic protein XIAP was also reduced at this time, levels of Bcl-2 and Bcl-x remained constant.
|
B inhibitor SN-50, and analyzing by real-time PCR. Upon treatment of NIT-1 ß-cells with poly(I:C), caspase-12 mRNA levels increased 40-fold over those of LF2000-treated cells and dropped to only 8-fold over background in the presence of SN-50 (Fig. 7A). Correspondingly, levels of the 60-kDa procaspase-12 protein increased in response to addition of 1 µg/ml poly(I:C) by 12 and at 24 h after poly(I:C) transfection, the levels of procaspase-12 were reduced and a 40-kDa cleavage product, indicating active caspase-12, was visible (Fig. 7B).
|
nor Fas ligand are responsible for PKR-induced apoptosis through FADD (23), so we looked at TRAIL, another member of this family. TRAIL has been shown to be an efficient killer of human ß-cells (24) and is also involved in human T-cell leukemia virus type I T-cell death (25). To determine whether TRAIL was up-regulated in the presence of intracellular dsRNA, real-time PCR was performed on total RNA extracted from NIT-1 cells treated with 1 µg/ml poly(I:C) for 12 h. TRAIL mRNA levels increased 33-fold over LF2000-treated control cells (Fig. 8A). This increase in TRAIL was significantly reduced to only 6-fold over controls in the presence of the NF-B inhibitor peptide, SN-50 (Fig. 8).
|
, both Fas and IL-15 mRNAs are up-regulated in the presence of intracellular poly(I:C), and this increase in mRNA is considerably reduced in the presence of the SN-50 peptide. Discussion
Viruses have a well-established association with type 1 diabetes and are known to be capable of inducing ß-cell death both in vitro and in vivo (26, 27). The viruses implicated are largely RNA viruses that replicate through a double-stranded intermediate that is the major inducer of the antiviral response and cell death (6, 28). In model systems, poly(I:C) has been used to examine the mechanism of virus-induced cell death by mimicking the dsRNA replicative intermediate. In a separate study on the role of coxsackieviruses in causing diabetes in SJL mice (29), we have shown that intracellular viral dsRNA displays a similar punctate staining pattern to that shown in Fig. 1B for poly(I:C). In addition, long-term persistence of viral dsRNA in pancreas was found in this study, indicating that this stable form of the viral genome may play a role in an ongoing inflammatory process long after infectious virus is no longer detectable. This provides a strong rationale for investigating ß-cell death induced by dsRNA or its synthetic counterpart poly(I:C) to understand the pathways involved and identify strategies to limit the damage caused in vivo.
The model that we have used comprises ß-cells transfected with poly (I:C) whose intracellular concentration can be controlled more easily than the amount of viral replicative intermediate following infection. Although there have been several studies on the effects of extracellular RNA plus interferon on ß-cells (13, 17), little is currently known about the effects of intracellular dsRNA in these cells. Additionally, NIT-1 ß-cells have been shown not to express the enzyme iNOS, allowing for examination of cell death pathways induced by dsRNA in the absence of NO-induced necrosis.
In this article, we show that, in the absence of supplementary cytokines, low concentrations of transfected poly (I:C) (1 µg/ml) caused significant caspase activation resulting in apoptosis of NIT-1 ß cells (Fig. 2B) as early as 12 h later. In comparison, without transfection, high concentrations (400 µg/ml) of poly (I:C) alone (without cytokines) did not induce apoptosis of transformed ß-cell lines (30), suggesting that uptake of dsRNA in the absence of transfection reagent is inefficient. In fact, although we were readily able to detect dsRNA by immunostaining of poly (I:C)-transfected cells (Fig. 1C), only a few cells with light diffuse staining were seen when 100 µg of poly (I:C) was added to the medium (Fig. 1B). Moreover, extracellular dsRNA has recently been shown to bind to Toll-like receptor 3, one of a family of receptors that are highly conserved through evolution and thought to play a critical role in innate immune defenses (18). Although TLR3 has been shown to activate NF-B species and induce interferon synthesis, these effects are mediated through an adaptor molecule called MyD88, and the pathways may differ significantly from those induced by intracellular dsRNA (18). This, together with the requirement for IFN to induce cell death with extracellular dsRNA, most likely explains differences seen between the effects of dsRNA added to the medium and the present study in which transfected dsRNA was used. Nevertheless, it appears that both extracellular and intracellular pathways induced by dsRNA are linked by activation of PKR. Our results show that PKR is phosphorylated 3 h following transfection of poly(I:C), coinciding temporally with the observed IB degradation/phosphorylation and NF-B nuclear translocation observed in our system.
The use of the SN50 inhibitor of NF-B p50 subunit nuclear translocation has enabled the contribution of this transcription factor to the induction of apoptosis to be determined. In Fig. 5A, intranuclear levels of NF-B (p50/p65) increased considerably at 3 h following poly (I:C) transfection and remained high for at least the next 9 h. In contrast, following addition of 100 µg/ml of SN50, only background levels of the complex were detected in the nucleus. (Fig. 5D). This was accompanied by a significant reduction of both apoptosis and the level of caspase activation (Fig. 5, C and E), confirming the critical role of NF-B (p50/p65) translocation in inducing cell death. Because phosphorylation of the translation factor eIF2 was also detected, this experiment enabled us to determine the relative contribution of NF-B-dependent transcription and inhibition of translation in transducing the effects of poly(I:C) at this time (12 h). Moreover, because any up-regulated genes must be translated to be effective, the fact that NF-B translocation is required suggests that any effects of protein synthesis inhibition must come into play later.
To determine which genes were up-regulated in response to NF-B translocation, three different approaches were used. First, Western blots to detect increased expression of death-promoting genes or down-regulation of survival factors like bcl2 and bcl-x were performed. In addition, both RNase protection assays and real-time PCR were carried out on genes that were thought to be potential candidates for induction of apoptosis. Western blot analysis showed that the expression of a range of genes known to mediate cell death in other systems (including Fas, FADD, p53, and Bax) was not altered at 12 h following poly (I:C) transfection (Fig. 6). In addition, no change in the amounts of survival factors (XIAP, Bcl2, or Bcl-x) was detected, a similar result to that found in cytokine-mediated ß-cell death, in which no changes in Bcl-2 or Bcl-x gene expression was observed by microarray analysis (5). Similarly, RNase protection assay analysis did not show up-regulation of Fas, FADD, or TNF, although IL-6, a regulatory cytokine, was induced (not shown). As a more sensitive measure of Fas mRNA production, RT-PCR was performed for both Fas and IL-15, genes shown to be up-regulated via extracellular poly(I:C) in other systems. The results show that both Fas and IL-15 mRNA were produced in the presence of intracellular poly(I:C), and this up-regulation was considerably reduced in the presence of the SN-50 peptide (Fig. 8B). Because no Fas protein was detected on Western blots, it remains to be seen whether the increase in Fas mRNA contributes to the observed NIT-1 ß-cell apoptosis.
In view of the fact that translation inhibition is associated with endoplasmic reticulum (ER) stress, the up-regulation and/or cleavage of caspase-12 was also examined. ß-Cells are known to be particularly susceptible to ER stress because of their highly developed ER system associated with insulin secretion (31). Procaspase-12 is localized on the cytosolic side of the ER membrane and is activated by calpain and/or caspase-7 in response to calcium release from the ER (31). Caspase-12 in turn activates caspase-9 independently of the mitochondrial pathway associated with cytochrome c and Apaf 1, resulting in caspase-3 and caspase-6 cleavage (32). In poly(I:C)-treated ß-cells, transcriptional up-regulation of caspase-12, which was dependent on NF-B translocation (i.e. inhibited by SN50) was observed by real-time PCR (Fig. 7A). In addition, both increased levels of caspase-12 and its cleavage product were detected on Western blots (Fig. 7B), thus implicating this ER stress pathway in the apoptosis seen. Interestingly, the caspase-12-mediated pathway has recently been shown to be active during replication of a cytopathic strain of bovine diarrhea virus, an RNA virus that replicates in the cytoplasm through a dsRNA intermediate and is known to inhibit host translation as well as causing cell death (33).
The result that several key mediators of the extrinsic cell death pathway (including FasL and TNF) were not up-regulated agrees with studies that were unable to block apoptosis caused by dsRNA by inhibiting either Fas/FasL or TNF (23, 34). The fact that NF-B translocation played a key role in causing the apoptosis seen led us to look at TRAIL, a member of the TNF family of proteins, which is known to be NF-B dependent and has recently been shown to be much more effective in inducing human ß-cell death than FasL, TNF, or lymphotoxin (24). Using real-time PCR, NIT-1 cells transfected with poly (I:C) showed a more than 30-fold increase in TRAIL mRNA, which was reduced to only 6-fold in the presence of SN50. This potential importance of TRAIL in mediating ß-cell death is currently under further investigation.
In conclusion, our results show that dsRNA transfected into ß-cells induces apoptosis by a process that is largely dependent on NF-B nuclear translocation. This is followed by the induction of at least two apoptotic pathways, an intrinsic one, which involves ER stress and translation inhibition and an extrinsic one in which TRAIL may play a role. Our results suggest that inhibitors of NF-B activation or translocation may be useful in preventing ß-cell apoptosis triggered by dsRNA, a molecule that may be critical in mediating virus-induced cell death.
Acknowledgments
We thank Dr. Bruce Verchere for advice and critical reading of this manuscript and the members of the Diabetes Group and beta-cell apoptosis network (ßCAN) at the British Columbia Research Institute for Children’s and Women’s Health for advice and encouragement.
Footnotes
This work was supported by Juvenile Diabetes Research Foundation Grant 20R60596 and a Canadian Diabetes Association Post-Doctoral Fellowship (to M.A.R.).
Abbreviations: AFC, 7 Amino 4 trifluoromethyl coumarin; DAPI, 4',6'-diamino-2-phenylindole; DEVD, peptide asp-glu-val-asp; dsRNA, double-stranded RNA; eIF2, eukaryotic initiation factor; ER, endoplasmic reticulum; FADD, Fas-associated death domain; H+L, heavy and light; IFN, interferon ; IB, inhibitory B; iNOS, inducible NO synthase; NF-B, nuclear factor B; NO, nitric oxide; PARP, poly(ADP-ribose) polymerase; PKR, RNA-dependent protein kinase; poly I:C, polyinosinic-polycytidylic acid; PVDF, polyvinyl difluoride; SDS, sodium dodecyl sulfate; TLR3, Toll-like receptor 3; TRAIL, TNF receptor-associated ligand; TUNEL, terminal deoxynucleotidyl transferase-mediated deoxyuridine triphosphate nick end-labeling; XIAP, X-linked inhibitor of apoptosis protein.
Received February 28, 2003.
Accepted for publication June 16, 2003.
References
B by the dsRNA-dependent protein kinase, PKR involves the IB kinase complex. Oncogene 19:1369–1378[CrossRef][Medline]
subunit of eukaryotic translation initiation factor 2 and NF-B. Mol Cell Biol 19:4653–4663: dsRNA-dependent protein kinase apoptosis and nitric oxide-dependent necrosis. Am J Pathol 159:273–283B and Rel proteins: evolutionarily conserved mediators of immune responses. Annu Rev Immunol 16:225–260[CrossRef][Medline]
B by double-stranded RNA (dsRNA) in the absence of protein kinase R and RNase L demonstrates the existence of two separate dsRNA-triggered antiviral programs. Mol Cell Biol 21:61–72 and IL-1 ß to induce both chemokine expression and nuclear factor-B-dependent apoptosis in pancreatic ß-cells: potential mechanisms for viral-induced insulitis and ß-cell death in type 1 diabetes mellitus. Endocrinology 143:1225–1234B by Toll-like receptor 3. Nature 413:732–738[CrossRef][Medline]
B in the absence of virus. Sci STKE 2001:RE1
receptors. Oncogene 19:3665–3674[CrossRef][Medline]
B signaling pathway governs TRAIL gene expression and human T-cell leukemia virus-I Tax-induced T-cell death. J Biol Chem 276:40385–40388This article has been cited by other articles:
 |
|
 |
|
  Z. Dogusan, M. Garcia, D. Flamez, L. Alexopoulou, M. Goldman, C. Gysemans, C. Mathieu, C. Libert, D. L. Eizirik, and J. Rasschaert Double-Stranded RNA Induces Pancreatic {beta}-Cell Apoptosis by Activation of the Toll-Like Receptor 3 and Interferon Regulatory Factor 3 Pathways Diabetes, May 1, 2008; 57(5): 1236 - 1245. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 F. Ortis, A. K. Cardozo, D. Crispim, J. Storling, T. Mandrup-Poulsen, and D. L. Eizirik Cytokine-Induced Proapoptotic Gene Expression in Insulin-Producing Cells Is Related to Rapid, Sustained, and Nonoscillatory Nuclear Factor-{kappa}B Activation Mol. Endocrinol., August 1, 2006; 20(8): 1867 - 1879. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. Salaun, I. Coste, M.-C. Rissoan, S. J. Lebecque, and T. Renno TLR3 Can Directly Trigger Apoptosis in Human Cancer Cells. J. Immunol., April 15, 2006; 176(8): 4894 - 4901. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. E. Jung, M. B. Gatch, and J. W. Simpkins Estrogen Neuroprotection Against the Neurotoxic Effects of Ethanol Withdrawal: Potential Mechanisms Experimental Biology and Medicine, January 1, 2005; 230(1): 8 - 22. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology |
