Role for Activating Transcription Factor 3 in Stress-Induced β-Cell Apoptosis (original) (raw)

Mol Cell Biol. 2004 Jul; 24(13): 5721–5732.

Matthew G. Hartman,1,2 Dan Lu,1,2 Mi-Lyang Kim,1,2 Gary J. Kociba,3 Tala Shukri,4 Jean Buteau,5,‡ Xiaozhong Wang,6,§ Wendy L. Frankel,7 Denis Guttridge,8 Marc Prentki,5 Shane T. Grey,9 David Ron,6 and Tsonwin Hai1,2,*

Matthew G. Hartman

Department of Molecular and Cellular Biochemistry,1 Center for Molecular Neurobiology,2 Department of Veterinary Biosciences,3 Department of Molecular Virology, Immunology, and Medical Genetics,8 Department of Pathology, Ohio State University, Columbus, Ohio,7 Immunology Research Center, Beth Israel Deaconess Hospital, Harvard Medical School, Boston, Massachusetts,4 Department of Nutrition, University of Montreal, Montreal, Quebec, Canada,5 Skirball Institute, NYU School of Medicine, New York, New York,6 Arthritis and Inflammation Research Program, Garvan Institute of Medical Research, Darlinghurst, Australia9

Dan Lu

Mi-Lyang Kim

Gary J. Kociba

Tala Shukri

Jean Buteau

Xiaozhong Wang

Wendy L. Frankel

Denis Guttridge

Marc Prentki

Shane T. Grey

David Ron

Tsonwin Hai

*Corresponding author. Mailing address: 1060 Carmack Rd., Columbus, OH 43210. Phone: (614) 292-2910. Fax: (614) 292-5379. E-mail: ude.uso@2.iah.

‡Present address: College of Physicians and Surgeons, Columbia University, New York, NY 10032.

§Present address: Department of Molecular Genetics, Baylor College of Medicine, Houston, TX 77030.

Received 2004 Mar 30; Accepted 2004 Apr 1.

Supplementary Materials

[Supplemental material]

GUID: 54476992-2C94-4CA2-9E12-E725491A83DC

GUID: 8F50C47B-66A4-4003-8EFC-43D9F587601F

Abstract

Activating transcription factor 3 (ATF3) is a stress-inducible gene and encodes a member of the ATF/CREB family of transcription factors. However, the physiological significance of ATF3 induction by stress signals is not clear. In this report, we describe several lines of evidence supporting a role of ATF3 in stress-induced β-cell apoptosis. First, ATF3 is induced in β cells by signals relevant to β-cell destruction: proinflammatory cytokines, nitric oxide, and high concentrations of glucose and palmitate. Second, induction of ATF3 is mediated in part by the NF-κB and Jun N-terminal kinase/stress-activated protein kinase signaling pathways, two stress-induced pathways implicated in both type 1 and type 2 diabetes. Third, transgenic mice expressing ATF3 in β cells develop abnormal islets and defects secondary to β-cell deficiency. Fourth, ATF3 knockout islets are partially protected from cytokine- or nitric oxide-induced apoptosis. Fifth, ATF3 is expressed in the islets of patients with type 1 or type 2 diabetes, and in the islets of nonobese diabetic mice that have developed insulitis or diabetes. Taken together, our results suggest ATF3 to be a novel regulator of stress-induced β-cell apoptosis.

It is widely accepted that autoimmunity is the main cause of type 1 but not type 2 diabetes. Despite this difference, β-cell death plays an important role in the pathophysiological progression of both diseases (15, 43, 45). On one hand, proinflammatory cytokines (interleukin-1β [IL-1β], tumor necrosis factor alpha [TNF-α], and gamma interferon [IFN-γ]) destroy β cells in the islets of Langerhans, leading to the pathogenesis of type 1 diabetes (11, 14, 15, 42); on the other hand, elevated glucose and free fatty acids (FFAs)—common metabolic abnormalities in type 2 diabetes—induce β-cell death, contributing to the progression of the disease (13, 32, 35, 53, 62). Emerging evidence indicates that activation of the NF-κB and Jun N-terminal kinase/stress-activated protein kinase (JNK/SAPK) signaling pathways is a key event leading to cell death, when β cells are exposed to these signals: proinflammatory cytokines, elevated glucose, and elevated FFAs (12, 15, 16, 43, 49). Furthermore, activation of these pathways has been demonstrated to impair insulin signaling (1, 17, 18, 36) and play a role in type 2 diabetes (63, 71). Therefore, these stress-activated signaling pathways constitute a common molecular mechanism in the pathophysiological progression of type 1 and type 2 diabetes.

Thus far, inducible nitric oxide (NO) synthase (iNOS), whose expression leads to NO production, is one of the best known target genes for these pathways (14-16, 42, 54). Several lines of evidence indicate that iNOS plays an important role in the pathogenesis of diabetes. (i) iNOS is induced in the islets by cytokines (16) and is expressed in the islets of diabetes prone BB rats (33) and nonobese diabetic (NOD) mice (55, 58). (ii) Transgenic mice expressing iNOS in β cells develop β-cell destruction and diabetes (60). (iii) β cells lacking functional iNOS are partially protected from stress-induced cell death (19, 41, 59).

In this report, we demonstrate that activating transcription factor 3 (ATF3), another stress-inducible gene, may be a downstream target of the NF-κB and JNK/SAPK signaling pathways and may play a role in β-cell apoptosis. ATF3 encodes a member of the ATF/CREB family of transcription factors (24, 26), and its expression is induced in a variety of tissues by different stress signals (for reviews, see references 25 and 27), including in the brain by seizure; in the liver by carbon tetrachloride; and in the pancreas, heart, and kidney by ischemia coupled with reperfusion (ischemia-reperfusion). Importantly, the induction of ATF3 correlates with cellular damage: all signals that induce ATF3 also induce cellular damage, and signals that do not induce ATF3 do not induce damage (25, 27). However, the functional significance of ATF3 is not clear. To date, both protective and detrimental effects of ATF3 expression have been reported. In cardiac myocytes, ectopic expression of ATF3 by adenoviral vector inhibited adriamycin-induced apoptosis (47), indicating a protective role of ATF3. Consistently, adenovirus-mediated expression of ATF3 protected superior cervical ganglion neurons from nerve growth factor withdrawal-induced apoptosis (46). However, in HeLa cells ectopic expression of ATF3 enhanced the ability of etoposide or camptothecin to induce apoptosis (44), suggesting a proapoptotic role of ATF3. Consistent with the proapoptotic role of ATF3, transgenic mice expressing ATF3 have functional defects in the corresponding tissues: mice expressing ATF3 in the heart have conduction abnormalities and contractile dysfunction (48); mice expressing ATF3 in the liver and pancreatic ductal epithelium have liver dysfunction and defects in endocrine pancreas development (2, 3). Therefore, the physiological function of ATF3 has been suggested to be either protective or detrimental. To address this controversy, we took both loss-of-function and gain-of-function approaches. In this report, we describe the generation of knockout mice deficient in ATF3 and transgenic mice expressing ATF3 specifically in β cells. Our results suggested ATF3 to be a stress-inducible, proapoptotic gene in β cells.

MATERIALS AND METHODS

Cell culture and treatments.

INS832/13 cells were grown in RPMI 1640 medium as described previously (30) and were incubated with reduced glucose (5 mM) for 24 h before treatments and (i) IL-1β (2,000 U/ml), TNF-α, or IFN-γ (1,000 U/ml each), or in combination (R&D Systems); (ii) bovine serum albumin (BSA) alone (0.5%) or BSA coupled with palmitate (0.4 mM) as detailed elsewhere (56); and (iii) 0.05 to 0.5 mM _S_-nitroso glutathione (GSNO) (Calbiochem). For inhibitors, cells were incubated with 20 μM Bay11-7082 or Bay11-7085 (Biomol), NF-κB SN50 or NF-κB SN50 M (4 μg/ml; Biomol), or 5 μM (unless otherwise indicated) JNKI-1 (Cleveland Clinic Foundation) for 30 min prior to IL-1β treatment.

Enzyme-linked immunosorbent assay (ELISA) analysis of DNA fragmentation.

DNA fragments were quantified at 24 h after glucose or palmitate treatment using the Cell Death Detection ELISA Plus Kit (Roche), which allows the determination of mono- and oligonucleosomes generated by apoptotic cleavage of DNA. The amount of nucleosomes (quantified photometrically by peroxidase activities) was determined on a FLUOstar Optima microplate reader (BMG) at 405 nm.

RNA isolation, real-time PCR, immunoblot, and immunohistochemistry.

Total RNA was isolated at 1 h after treatment unless otherwise indicated, and ATF3 or glyceraldehyde-3-phosphate dehydrogenase mRNA analyzed by real-time PCR as previously (2). Nuclear extracts were isolated at 1 h after treatment unless otherwise indicated, and equal amount of extract was analyzed by immunoblot using anti-ATF3 (Santa Cruz), antiactin (Sigma), or anti-extracellular signal-regulated kinase (anti-ERK) (Cell Signaling Technology) antibodies. Immunohistochemistry was determined as previously described (2). Human pancreata were retrieved from the archival files at Ohio State University. Only cases with minimal autolysis were used. The diagnoses of diabetes (type 1 or 2) were based on the medical records.

Analyses of JNK/SAPK, p38, and NF-κB activity.

JNK/SAPK was immunoprecipitated from INS832/13 cell extract at 30 min after IL-1-β treatment, and assayed using glutathione _S_-transferase-Jun(1 to 79 amino acids) as substrate in the absence or presence of indicated inhibitors. p38 activation was assayed by the phosphorylation of its substrate ATF2 using phospho-specific antibody (Cell Signaling). NF-κB activity was determined by electrophoretic mobility shift assay (EMSA) as previously described (23).

Generation of PDX-ATF3 transgenic mice.

A 1.0-kb enhancer region (kb −2.7 to −1.7) of the PDX promoter (C. Wright at Vanderbilt University) was inserted upstream of the E1B TATA box to drive the expression of ATF3 as shown (see Fig. 4). Transgenic mice were generated in FVB/N mice and identified by PCR.

PDX-ATF3 transgenic mice have small and abnormal islets. (a) Schematic of the transgenic construct. (b and c) Pancreatic sections of nontransgenic (Non-Txg) or transgenic (Txg) mice were stained with H&E (b) and analyzed for size distribution of their islets (c) (x axis, 1 = 1,000 pixel units). For H&E stain, multiple sections from more than five mice in each group were analyzed and representative figures are shown. The green lines delineate one islet from each group. For size distribution, 200 islets from three non-Txg and 50 islets from three Txg mice were analyzed. (d) Pancreatic sections from nontransgenic (upper panel) or transgenic (lower panel) mice were analyzed by immunofluorescence using insulin (green) and glucagon (red) antibodies. A representative figure from each group is shown.

Serum or blood analysis, staining of tissue sections and measurement of islet area.

Blood glucose was measured by Glucometer Elite (Bayer) and other parameters were measured as previously (3). Immunohistochemistry was carried out as previously (3).

Generation and confirmation of ATF3 knockout mice.

ATF3 genomic clone was isolated from a 129SVJ library and knockout mice were generated in the 129SVJ background. Knockout allele was distinguished from the wild-type allele by Southern blot or PCR. For Southern blot, genomic DNA was digested by EcoRV and analyzed by a 5′ probe (EcoRI-XhoI fragment) (see Fig. 3 and Fig. S1 in the supplemental material), or digested by EcoRI and analyzed by a 3′ probe (BglII fragment) (see Fig. 3a and Fig. S1 in the supplemental material). For PCR, three primers were used: 5′-AGAGCTTCAGCAATGGTTTGC-3′ (primer 1), 5′-TGAAGAAGGTAAACACACCGTG-3′ (primer 2), and 5′-ATCAGCAGCCTCTGTTCCAC-3′ (primer 3). Congenic knockout mice in the background of C57BL/6 were generated by 10 backcrosses. Wild-type or knockout mice were injected with lipopolysaccharide (LPS) at 1.5 mg/kg of body weight (i.p.) or phosphate-buffered saline as a control. At 4 h after treatment, mice were sacrificed and nuclear extract from liver was analyzed by immunoblotting.

Islets deficient in ATF3 are partially protected from cytokine- or NO-induced apoptosis. (a) Schematic of the knockout construct (for more details, see Fig. S1 in the supplemental material). (b) Genomic DNA from wild-type (+/+), heterozygous (+/−) or homozygous (−/−) ATF3 knockout mice were analyzed by Southern blot. (c) Wild-type (+/+) and knockout (−/−) mice were treated with phosphate-buffered saline as control (−) or LPS (+) for 4 h to induce ATF3 in the liver, and liver extracts were assayed by immunoblot for ATF3 or ERK (control). (d) Primary islets from wild-type (WT) or ATF3 knockout (KO) mice were treated with IL-1β-IFN-γ at the concentrations detailed in the Materials and Methods for the indicated periods and assayed by immunoblot for ATF3 or actin. (e) (Right panel) Primary islets from wild-type (WT) or ATF3 knockout (KO) mice were treated with medium (as a control), IL-1β-IFN-γ (2 Cyto.), IL-1-β-IFN-γ-TNF-α (3 Cyto.), or IL-1-β-IFN-γ-Jo 2 (2 Cyto.+Fas), and assayed for apoptosis at 24 h after treatment. Mean ± SE of percentage increase in apoptosis from five experiments are shown (right panel). *, P < 0.05 versus wild type. (Left panel) Representative flow cytometry data for medium control and 2 Cyto. treatment are shown. A°: % of cells with subdiploid DNA content (excluding cell debris). (f) INS832/13 cells were treated with increasing amounts of GSNO (0.05 mM, lane 4; 0.1 mM, lane 5; 0.5 mM, lane 6) for 2 h or were left untreated (medium control, lane 3) and then were assayed by immunoblot for ATF3 or ERK. Lanes 1 and 2 are extracts from liver treated (+) or untreated (−) with LPS as in panel c. (g) Primary islets from wild-type mice were treated with 0.625 mM GSNO for the indicated periods and assayed by immunoblot for ATF3 or actin. (h) Primary islets from wild-type (WT) or ATF3 knockout (KO) mice were treated with 0.625 mM GSNO, or treated with IL-1β-IFN-γ (2 Cyto.) in the absence (−) or presence (+) of the iNOS inhibitor l-NIO, and assayed for apoptosis at 24 h after treatment. Means ± standard errors of percentage increase in apoptosis from 5 experiments are shown. * P < 0.01 versus wild type; # P < 0.05 versus (−) l-NIO.

Primary islets and flow cytometric analysis of apoptosis.

Mouse islets were isolated with Liberase (Roche, Indianapolis, Ind.) as previously described (22) and were treated as indicated in the figure legend with the following: medium (control), IL-1β (20 U/ml), IFN-γ (200 U/ml), TNF-α (200 U/ml), anti-Fas monoclonal antibody Jo-2 (2 μg/ml; Pharmingen), or 0.625 mM GSNO. When indicated, l-_N_5-(1-iminoethyl) ornithine dihydrochloride (l-NIO) was included at 500 μM. For GSNO treatment, NO content in the medium was confirmed by Griess reagent. At 24 h after treatment, islets were dispersed, stained with propidium iodide, and analyzed by flow cytometry for apoptosis as described previously (22). Apoptotic cells were scored as cells with a hypodiploid DNA content (<2N). Cell debris and apoptotic cell-free fragments were excluded by discounting the events with an FL-2 area profile below that of chicken erythrocyte nuclei. Percentage increase in apoptosis (induced by a given treatment) was calculated according to an established method (22, 41): (% apoptotic cells after treatment − % apoptotic cells in medium control) ÷ (100% − % apoptotic cells in medium control).

Institutional reviews.

Animal experiments and experiments using patient samples were approved by the appropriate Institutional Review Board at Ohio State University.

RESULTS

Induction of ATF3 in β cells by signals relevant to type 1 or type 2 diabetes.

Although ATF3 has been known to be a stress-inducible gene, it is not clear whether it is induced in β cells by signals relevant to type 1 or type 2 diabetes. To address this question, we examined ATF3 expression in INS832/13 insulin-positive β cells (30). As shown in Fig. 1a, IL-1β alone transiently increased ATF3 mRNA level as indicated by real-time PCR, but TNF-α or IFN-γ alone did not (not shown). Interestingly, TNF-α plus IFN-γ enhanced and prolonged the induction of ATF3 by IL-1β (Fig. 1a). Immunoblot analysis confirmed the production of ATF3 protein (Fig. 1b). This TNF-α-IFN-γ-mediated potentiation parallels the well-documented potentiation of IL-1β-induced apoptosis by these two cytokines (42), supporting the notion that ATF3 expression contributes to apoptosis (see below). Combinations of two cytokines indicated that IL-1β-TNF-α or IL-1β-IFN-γ induced ATF3 at about 70 to 90% efficiency compared to all three cytokines, and TNF-α-IFN-γ induced ATF3 at around 10% efficiency (Fig. 1c). Elevated glucose (25 mM), a condition that induced apoptosis in INS832/13 cells (Fig. 1e, bar 3), induced ATF3 around 30-fold (Fig. 1d, bar 3). The fatty acid palmitate (_n_-hexadecanoate, 16:0), which induced apoptosis modestly (Fig. 1e, bar 2), induced ATF3 by twofold (Fig. 1d, bar 2). Interestingly, glucose plus palmitate (glucolipotoxicity [53]) induced ATF3 and apoptosis with a higher efficiency than either treatment alone (Fig. 1d and e). The induction of ATF3 by palmitate is consistent with the DNA microarray result that ATF3 is induced in MIN-6 β cells by this fatty acid (7).

ATF3 is induced by cytokines and elevated glucose or palmitate. (a to c) INS832/13 cells were treated with the indicated cytokines as detailed in Materials and Methods and assayed at the indicated time points by real-time PCR for ATF3 mRNA (a and c) or by immunoblot for ATF3 protein (b). For real-time PCR, glyceraldehyde-3-phosphate dehydrogenase mRNA was used as an internal control, and the level from untreated cells was defined as 1. (c) The level of induction by all three cytokines was arbitrarily defined as 100%. (d) INS832/13 cells were treated with glucose or palmitate (0.4 mM coupled with 0.5% BSA) as indicated, and ATF3 mRNA was assayed by real-time PCR at 1 h after induction. Note the difference in the scales of the y axes. (e) INS832/13 cells were treated with glucose or palmitate, and apoptosis was determined by ELISA analysis of DNA fragments at 24 h after treatment. For real-time PCR, the means ± standard errors from three to five experiments are shown; for ELISA analysis, means ± standard errors from three experiments are shown.

NF-κB and JNK/SAPK pathways in the induction of ATF3 by IL-1β.

As described in the introduction, the NF-κB and JNK/SAPK pathways play an important role in β-cell apoptosis and diabetes. Interestingly, the human ATF3 promoter contains NF-κB, AP-1, and ATF/CRE sites (37), binding sites recognized by the transcription factors activated by the NF-κB and JNK/SAPK pathways (for a review, see reference 68). These observations, combined with the induction of ATF3 by stress signals in β cells (above), prompted us to examine whether the activation of these pathways is necessary for the induction of ATF3 by IL-1β using an inhibitory approach. To inhibit the NF-κB pathway, we used two types of reagents: (a) small compound inhibitors of IKK—Bay11-7082 and Bay11-7085 (52), and (b) a cell-permeable peptide NF-κB SN50 which blocks the nuclear translocation of NF-κB p50 (40). These inhibitors reduced the induction of ATF3 by IL-1β in INS832/13 cells as indicated by real-time PCR, but a control peptide NF-κB SN50 M which contains a mutation in the peptide did not affect ATF3 induction (Fig. 2a). Immunoblot analysis indicated that the inhibitors also reduced the ATF3 protein level (Fig. 2b). EMSA confirmed that Bay11-7082 and Bay11-7085 inhibited the NF-κB DNA binding activity in the nuclear extracts (Fig. 2c). However, the inhibitors did not inhibit JNK/SAPK (Fig. 2d) or p38 kinase (Fig. 2e) which is also induced by IL-1β. To inhibit the JNK/SAPK pathway, we used the JNKI-1 peptide, a cell-permeable peptide that inhibits the activation of this pathway (6). JNKI-1 reduced the induction of ATF3 by IL-1β at both the mRNA (Fig. 2f) and protein (Fig. 2g) levels. The inhibition at the protein level appears to be more efficient than that at the mRNA level. Further investigation is required to explain this apparent difference. The specificity of the peptide is indicated by its ability to inhibit JNK/SAPK (Fig. 2h), but not NF-κB (Fig. 2i) or p38 (Fig. 2j) pathway. Interestingly, combination of both inhibitors (JNKI-1 plus Bay11-7082) did not result in a complete inhibition of ATF3 induction (Fig. 2f), suggesting that other pathways are also involved.

Inhibition of the NF-κB or JNK/SAPK pathway reduces the induction of ATF3 by IL-1β. (a to c) INS832/13 cells were treated with IL-1β in the absence or presence of the indicated inhibitors or a control peptide inhibitor NF-κB SN50 M. ATF3 mRNA was assayed at 1 h after IL-1β treatment by real-time PCR (a); ATF3 protein was assayed at 1 h after treatment by immunoblot (b); NF-κB DNA binding activity in the nuclear extract was assayed by EMSA at 15 min after treatment or at the indicated time points (c). (d) INS832/13 cells were treated with IL-1β for 30 min, and JNK/SAPK was assayed by immunoprecipitation coupled with kinase (IP-kinase) assay using glutathione _S_-transferase-Jun as the substrate in the absence or presence of the indicated inhibitors. (e) INS832/13 cells were untreated (lane 1) or treated with IL-1β (lanes 2 to 4) in the absence (lane 2) or presence of Bay 11-7082 (lane 3) or NF-κB SN50 (lane 4) for 30 min. p38 activation was examined by the phosphorylation of its substrate ATF2. (f and g) Same as panel a and b except JNKI-1 was used as the inhibitor. (h) JNK/SAPK was assayed by IP-kinase assay in the absence or presence of JNKI-1. (i) NF-κB DNA binding activity in the nuclear extract was assayed by EMSA as described in panel c except JNKI-1 was used as the inhibitor. (j) p38 activation was assayed as in panel e except JNKI-1 was used as the inhibitor. For real-time PCR, means ± standard errors from at least three experiments are shown.

Partial protection from cytokine-induced apoptosis in primary islets deficient in ATF3.

To investigate the functional significance of ATF3, we generated mice deficient in ATF3. The mutant allele lacks exon B, which contains the AUG initiation codon (Fig. 3a; see Fig. S1 in the supplemental material), and can be distinguished from the wild-type allele by Southern blot (Fig. 3b) or PCR (see Fig. S1 in the supplemental material). Homozygous knockout mice (ATF3−/−) did not produce any ATF3 protein in the liver after intraperitoneal injection of LPS, a treatment that significantly induced ATF3 in the wild-type liver (Fig. 3c). Therefore, it confirms that the knockout mice lack functional ATF3 genes. The ATF3−/− mice have no lethality or obvious phenotypes, consistent with the notion that ATF3 is a stress-inducible gene and is not required under normal conditions. To determine whether ATF3 plays a role in stress-induced β-cell apoptosis, we compared primary islets isolated from ATF3+/+ and ATF3−/− mice for cytokine-induced apoptosis. We first confirmed that IL-1β-IFN-γ induced ATF3 in ATF3+/+ islets (Fig. 3d). As expected, ATF3 protein was absent in the ATF3−/− islets after cytokine treatment, further confirming the functional knockout of ATF3 gene. We then treated the islets with cytokines and assayed apoptosis at 24 h after treatment by propidium iodide stain followed by flow cytometry. Results from five experiments indicated that ATF3−/− islets were partially protected from two-cytokine (IL-1β-IFN-γ)-induced apoptosis (P < 0.05) but were only marginally protected (not statistically significant) from three-cytokine (IL-1β-IFN-γ-TNF-α)- or two-cytokine-plus-Fas-induced apoptosis (Fig. 3e, right panel). A representative flow cytometry result is shown (left panel).

As shown above, the NF-κB and JNK/SAPK pathways play an important role in the induction of ATF3 by IL-1β. Overwhelming evidence in the literature indicates that these pathways mediate cytokine-induced expression of iNOS (15, 22, 43). Induction of iNOS leads to NO production, and NO donor GSNO is widely used to mimic the action of iNOS. We found that GSNO induced ATF3 in both INS832/13 β cells (Fig. 3f) and primary islets (Fig. 3g). Interestingly, ATF3−/− islets were partially protected (P < 0.01) from GSNO-induced apoptosis (Fig. 3h). That is, NO in the absence of ATF3 (ATF3−/− background) fails to elicit efficient killing, suggesting that the induction of ATF3 by NO plays a role in NO-induced β-cell apoptosis. Previously, iNOS−/− islets (ATF3+/+) were shown to be partially protected from cytokine-induced apoptosis (41), indicating that induction of ATF3 (by cytokines) in the absence of iNOS (iNOS−/− background) fails to elicit efficient killing. Taken together, these observations suggest that induction of both ATF3 and iNOS is necessary for stress signals to elicit efficient β-cell killing. Consistent with this interpretation, iNOS inhibitor l-NIO further decreased two-cytokine-induced apoptosis in ATF3−/− islets (Fig. 3h).

Dysfunction in transgenic mice expressing ATF3 in islets.

To further address the functional significance of ATF3 expression in islets, we took a gain-of-function approach and generated transgenic mice expressing ATF3 under the control of the fragment from kb −2.7 to −1.7 of the PDX-1 promoter (Fig. 4a). This fragment was demonstrated to target transgenes selectively in the developing islets and in β cells after birth (21, 70). Previously, we reported that transgenic mice expressing ATF3 under the control of the transthyretin promoter have defects in glucose homeostasis (2). However, the transthyretin promoter fragment directed the expression of ATF3 in both the liver and pancreatic ductal epithelium (2). Therefore, it was not possible to ascertain the impact of ATF3 expression in islets. Using the PDX-ATF3 construct, we obtained five transgenic founders which did not express the transgene (presumably due to mosaicism or silencing), but could pass it on to the progeny. F1 mice from all five founders expressed the transgene and died before mating. Therefore, no transgenic lines were established and all results were derived from the analyses of F1 mice. Three founders gave rise to litters of small size and low transgenic transmission rate (much lower than the expected 50%). The remaining two founders gave rise to the expected transgenic transmission rate (approximately 50%) and their progeny was further analyzed. F1 mice (PDX-ATF3) from founder 1 died within several days after birth and displayed islets with reduced size (Fig. 4b) and number (5 ± 0.05 versus 20 ± 0.1 per 107 pixel area in nontransgenic mice, P < 0.01). Analyses of islet population indicated a shift in size distribution toward small islets in the transgenic mice (Fig. 4c). Immunofluorescence analysis indicated abnormal distribution of hormone-positive cells (Fig. 4d). Transgenic pancreata (lower panel) had fewer insulin-positive cells (green) than nontransgenic pancreata (upper panel). Furthermore, glucagon-positive cells (red) in transgenic pancreata formed clusters with insulin-positive cells but failed to form the proper mantle/core arrangement of the α/β cells as in the nontransgenic pancreata (glucagon-positive α cells at the periphery of the islets and insulin-positive β cells at the core). These transgenic mice had low body weight, and defects consistent with β-cell deficiency (transgenic versus nontransgenic, P < 0.001): high glucose, low insulin, high β-hydroxybutyrate, and high triglyceride (Table 1). Due to the small size of the mice, sera from three to four mice were combined as one sample for analyses, and at least four samples were used to generate the above data. Analysis of glucagon did not show statistically significant difference between transgenic and nontransgenic mice (Table 1). F1 mice from founder 2 displayed less severe phenotypes. Although the precise reasons are not clear, variation in the severity of phenotypes among different transgenic founder lines is a common phenomenon. Islets from progeny of founder 2 were not greatly reduced in size or number, but displayed abnormal morphology with rough surface (not shown). Many of these mice died before 1 week and none survived to adulthood.

TABLE 1.

PDX-ATF3 mice have low body weight and defects secondary to β-cell deficiency_a_

Mouse group	Mean body wt ± SE (g)	Mean blood glucose level ± SE (mg/dl)	Mean level in serum ± SE
Insulin (ng/dl)	Glucagon (pg/ml)	TG (mg/dl)	β-OH-buty (mg/dl)
Non-Txg	1.54 ± 0.10	73 ± 3	4.3 ± 1.1	180 ± 21	125 ± 11	9.6 ± 0.8
Txg	1.26 ± 0.08	347 ± 35_b_	<0.2_b_	231 ± 61	194 ± 8_b_	16.6 ± 1.9_b_

Expression of ATF3 in the islets of diabetic mice and patients.

Taken together, the above results (ATF3 induction, loss-of-function and gain-of-function studies) support the notion that stress-induced ATF3 expression in β cells plays a role in β-cell apoptosis. To address whether ATF3 is expressed in the pathophysiological context of diabetic progression, we examined ATF3 expression in the islets of NOD mice which develop diabetes spontaneously (28). We examined three to four mice at various age groups (1, 2, 3, 8, 12, 17, and 33 weeks old), and representative figures for three age groups are shown (Fig. 5a to f): 1 week (Fig. 5a and d), 12 week with insulitis (Fig. 5b and e, blood glucose 120 mg/dl), and 33 week with diabetes (Fig. 5c and f, blood glucose >500 mg/dl). As shown by immunohistochemistry, ATF3 is expressed in the islets at the diabetic stage (Fig. 5f), and prediabetic but insulitis stage (Fig. 5e). Hematoxylin and eosin (H&E) staining of the adjacent sections revealed that, at the insulitis stage, ATF3 is expressed in cells adjacent to the infiltrating lymphocytes (compare Fig. 5b and 5e). Figure 5c shows two islets from a mouse with overt diabetes (>500 mg/dl): one islet (upper right corner) has infiltrating lymphocytes and ATF3 was expressed next to the lymphocytes; the other islet (lower left corner) has deformed shape and ATF3 was expressed throughout the islet. The specificity of the signals is demonstrated by the lack of signals in cells away from the lymphocytes (Fig. 5e), and the lack of signals in islets before the NOD mice develop insulitis (Fig. 5d). In the NOD model, some mice do not develop diabetes even by 30 weeks of age (28). To determine whether those NOD mice that do not develop diabetes express ATF3, we examined several nondiabetic NOD mice (>33 weeks old). Their islets had normal appearance without obvious deformation or infiltrating lymphocytes, and did not express ATF3 to any significant degree (see Fig. S2 in the supplemental material). All these results demonstrate a correlation between diabetic progress and ATF3 expression. Significantly, in addition to the diabetic mouse model, ATF3 is expressed in the diabetic human islets: it is expressed in the islets of both type 1 and type 2 diabetic patients but not in the islets of nondiabetic patients (Fig. 5g to i). Multiple sections from two type 1 and three type 2 patients were examined, and representative results are shown.

ATF3 is expressed in the diabetic pancreata. (a to f) Pancreata from NOD mice at 1 week (a and d), 12 weeks (b and e), or 33 weeks (c and f) of age were stained by H&E (a to c) or analyzed by immunohistochemistry for ATF3 (d to f). Multiple sections from three to four mice at various age groups were analyzed, and representative figures are shown. In panel b infiltrating lymphocytes in the islets are delineated by blue dots, and in panel e islets are delineated by black dots. (g to i) Archived paraffin sections from nondiabetic (g) or type 2 (h) or type 1 (i) diabetic patients were analyzed by immunohistochemistry for ATF3. Multiple sections from two type 1 and three type 2 patients were analyzed, and representative figures are shown. Bar = 100 μm.

To determine whether ATF3 is expressed in insulin-positive (insulin+) cells, we carried out immunohistochemistry assay using antibodies against ATF3, insulin or glucagon on available paraffin sections (which are close but not immediately adjacent). Figure 6 shows the representative pictures from this series of experiments and indicates that ATF3-positive (ATF3+) cells corresponded to insulin+, but not glucagon-positive (glucagon+) cells. Panels a to c are derived from an NOD mouse that had developed insulitis. The border of one islet was delineated by dotted line. Insulin+ cells occupied less than half of the islets (the rest of the islet was filled with infiltrating lymphocytes), and ATF3+ cells were in the area where insulin+ cells reside. Panels d to f are derived from a type 1 patient. Again, the ATF3+ cells corresponded to insulin+ cells. As expected, most of the insulin+ cells had been destroyed and only a few clusters of insulin+ cells were found in type 1 samples. A low-magnification (×20) picture (see Fig. S3 in the supplemental material) revealed a large field with one cluster of cells enlarged (insets). The type 1 sample also contained many atrophic islets with glucagon+ but not insulin+ cells. In these islets, no ATF3+ cells were found. An enlarged view of three atrophic islets are shown (see Fig. S3 in the supplemental material [dotted box]). Our observations that type 1 sample contained a few islets still with insulin+ cells and many atrophic islets with glucagon+ but not insulin+ cells are consistent with previous analyses of type 1 pancreata (20; S. Bonner-Weir personal communication). Figure 5g to i are derived from a type 2 patient. We note that human and rodent islets have a different appearance at first glance: instead of the mantle-core arrangement of the α/β cells in rodent islets, human islets have composites of several rodent-like islet subunits (5; S. Bonner-Weir, personal communication). Taken together, these immunohistochemistry data (from NOD mice, type 1 and type 2 patients) support the notion that ATF3 is expressed in insulin+ cells. Recently, ATF3 was reported to be expressed in wild-type glucagon+ α cells (66). We do not know the reasons for this apparent discrepancy, except that it may be due to the differences in strains or antibodies.

ATF3-positive cells correspond to insulin- but not glucagon-positive cells. Immunochemistry was carried out for insulin, ATF3 or glucagon using available paraffin sections (which are close but not immediately adjacent). (a to c) Pancreata from NOD mice that had developed insulitis (at 17 weeks of age) were analyzed for insulin (a), ATF3 (b), or glucagon (c). Bar = 100 μm (d to f) Archived paraffin sections from a type 1 patient were analyzed for insulin (d), ATF3 (e), or glucagon (f) content. Bar = 200 μm. (g to i) Same as panels d to f except sections from type 2 patients were used. Bar = 200 μm.

DISCUSSION

In this report, we describe several observations that implicated ATF3 in β-cell apoptosis. (i) ATF3 is induced by signals and signaling pathways that have been demonstrated to induce β-cell apoptosis and implicated in the development of type 1 or type 2 diabetes. (ii) Expression of ATF3 in β cells leads to islet dysfunction and defects secondary to β-cell deficiency. (iii) Islets deficient in ATF3 are partially protected from cytokine- and NO-induced apoptosis. (iv) ATF3 is expressed in the islets of both type 1 and type 2 diabetic patients and in the islets of NOD mice that have developed insulitis or diabetes. Taken together, our results support a proapoptotic role of ATF3 in β cells. This conclusion is consistent with the observation that mouse embryonic fibroblasts derived from ATF3−/− mice are partially protected from stress-induced apoptosis (C. Yan, D. Lu, T. Hai, and D. D. Boyd, submitted for publication). As described in the introduction, ATF3 has been implicated to be either pro- or antiapoptotic. This discrepancy may be due to the context dependency of ATF3: it is antiapoptotic in some cells under certain conditions, but proapoptotic in others.

Since β-cell apoptosis plays an important role in the pathophysiological progression of diabetes, our results support the speculation that stress-induced expression of ATF3 plays a role in diabetes. However, much more work is required to substantiate this speculation. In an attempt to test the potential role of ATF3 in diabetes, we analyzed wild-type and ATF3 knockout mice using the multiple low-dose streptozotocin (MLD-STZ) model (38, 39), where insulitis and hyperglycemia are induced in male mice in an accelerated manner (over a 2-week period) following five daily injections of low does STZ (40 mg/kg of body weight). Importantly, previous studies have demonstrated an autoimmune component in this diabetes model (see reference 29 and references therein). Preliminary results from analysis of 20 mice in each group showed no significant difference in the blood glucose levels of ATF3 knockout mice from that of wild-type mice (data not shown). We suggest two potential explanations for this result. First, knockout of one stress-inducible gene, ATF3, is not sufficient to protect islets from the level of insults induced by STZ. As described above, ATF3 knockout does not provide significant protection from apoptosis when TNF-α or Fas pathway was activated. Since apoptosis is regulated by complex cellular processes that involve many cross-interacting pathways and genes, it is reasonable that knockout of ATF3 alone is not sufficient for full protection. Second, ATF3 knockout mice may be partially protected from STZ-induced insults but the protection is not discernible by measuring the blood glucose levels. Thus, more work including detailed histological analysis is necessary for definitive conclusion. In this context, we note that knockout of IL-1 receptor was recently reported to delay, but not prevent, diabetes in NOD mice (61).

On the basis of our observations and reports on iNOS in the literature, we propose the following model. Expression of ATF3 and iNOS genes is induced by cytokines, at least in part, through the NF-κB and JNK/SAPK signaling pathways. The induction of iNOS leads to NO production, which in turn further induces ATF3 gene expression. This supposition is supported by a recent observation that iNOS inhibitor reduced cytokine-induced ATF3 expression in INS-1 β cells at late time points (>8 h) but not at the early time points (34). Since elimination of either ATF3 or iNOS reduced cytokine-induced apoptosis (see reference 41 for iNOS knockout and this report for ATF3 knockout), it appears that induction of both genes is necessary for efficient β-cell death.

As shown in our results, ATF3 knockout islets were protected from 2 cytokine (IL-1β-IFN-γ)-induced apoptosis, but not three-cytokine- or two-cytokine-plus-Fas-induced apoptosis. One element common to Fas and TNF-α (the third cytokine) is the activation of the death receptor pathway—the FADD-caspase 8 pathway (4, 10, 64). Therefore, our results suggest that ATF3 plays a role for (IL-1β-IFN-γ)-induced apoptosis but may not play a significant role in the death-receptor pathway. We note that NO was reported to mediate cell death in β cells via a pathway distinct from Fas-mediated pathway (72).

Although ATF3 is induced by a variety of seemingly diverse stress signals such as ischemia-reperfusion, hyperglycemia, hyperlipidemia, cytokines, and UV light, many of these signals elicit oxidative stress, that is, an imbalance between the reactive species and antioxidant molecules (57). Therefore, ATF3 can be viewed as an oxidative-stress induced gene. Consistent with this notion, ATF3 is induced by H2O2 and this induction is repressed by the antioxidant _N_-acetyl-l-cysteine (2). Accumulating evidence indicates that oxidative stress induces a variety of responses relevant to diabetes, including activation of the NF-κB and JNK/SAPK pathways, insulin-resistance, β-cell dysfunction, and β-cell destruction (17, 18, 43, 57). Therefore, our finding that ATF3, an oxidative-stress-inducible gene, plays an important role in β-cell destruction (and perhaps diabetes) is consistent with the current literature. The significance of our finding is that it is the first to implicate ATF3 in β-cell apoptosis, and it supports a potential link between ATF3 and iNOS in cytokine-induced gene networks (34).

Recently, gadd153/CHOP knockout islets were demonstrated to be partially protected from NO-induced cell death (51), a result similar to that from the ATF3 knockout islets. Previously, we demonstrated that gadd153/CHOP is an ATF3-interacting protein (9) and its corresponding gene may be a downstream target of ATF3 (69). Therefore, ATF3 may affect the ability of NO to modulate the cell death machinery either directly or indirectly—through the interaction with other proteins or the regulation of downstream genes (such as gadd153/Chop10). We note that both ATF3 and gadd153/CHOP are induced by endoplasmic reticulum (ER) stress (8, 31, 67). Significantly, deletion of gadd153/CHOP gene has been demonstrated to delay the onset of diabetes in Akita mice (50), a model where diabetes is thought to be induced by ER stress (65). Therefore, it would be interesting to see whether the deletion of ATF3 also protects β cell from ER stress-induced apoptosis and delay the onset of diabetes in Akita mice. Finally, ATF3 is induced in β cells by cytokines and elevated glucose or FFA and is expressed in the islets of type 1 and 2 patients, supporting the emerging notion that the mechanisms leading to β-cell death in both forms of diabetes may share more similarity than previously thought.

Supplementary Material

Acknowledgments

We thank Susan Bonner-Weir at Joslin Diabetes Center for expert comments and advice on islet data, C. Wright at Vanderbilt University for the PDX promoter fragment, and the KECK Genetic Research Facility at the Ohio State University for generating the PDX-ATF3 mice.

This work was supported by grants NIH ES08690 and DK59605, ADA research grant 868688 (T.H.), grants NIH DK47119 and ES08681 (D.R.), the Juvenile Diabetes Research Foundation (M.P.), and grant JDRF 2000-719 (S.T.G.). M.P. is a Canadian Institute of Health Research (CIHR) Scientist, and J.B. is supported by a CIHR Ph.D. studentship.

Footnotes

REFERENCES

1. Aguirre, V., T. Uchida, L. Yenush, R. Davis, and M. F. White. 2000. The c-Jun NH2-terminal kinase promotes insulin resistance during association with insulin receptor substrate-1 and phosphorylation of Ser307. J. Biol. Chem. 275**:**9047-9054. [PubMed] [Google Scholar]

2. Allen-Jennings, A. E., M. G. Hartman, G. J. Kociba, and T. Hai. 2001. The roles of ATF3 in glucose homeostasis: A transgenic mouse model with liver dysfunction and defects in endocrine pancreas. J. Biol. Chem. 276**:**29507-29514. [PubMed] [Google Scholar]

3. Allen-Jennings, A. E., M. G. Hartman, G. J. Kociba, and T. Hai. 2002. The roles of ATF3 in liver dysfunction and the regulation of phosphoenolpyruvate carboxykinase gene expression. J. Biol. Chem. 277**:**20020-20025. [PubMed] [Google Scholar]

4. Baud, V., and M. Karin. 2001. Signal transduction by tumor necrosis factor and its relatives. Trends Cell Biol. 11**:**372-377. [PubMed] [Google Scholar]

5. Bonner-Weir, S., and F. E. Smith. 1994. Islets of Langerhans: morphology and its implications, p. 15-28. In C. R. Kahn and G. C. Weir (ed.), Joslin's diabetes mellitus, 13th ed. Lea and Febiger, Philadelphia, Pa.

6. Bonny, C., A. Oberson, S. Negri, C. Sauser, and D. F. Schorderet. 2001. Cell-permeable peptide inhibitors of JNK. Diabetes 50**:**77-82. [PubMed] [Google Scholar]

7. Busch, A. K., D. Cordery, G. S. Denyer, and T. J. Biden. 2002. Expression profiling of palmitate- and oleate-regulated genes provides novel insights into the effects of chronic lipid exposure on pancreatic beta-cell function. Diabetes 51**:**977-987. [PubMed] [Google Scholar]

8. Cai, Y., C. Zhang, T. Nawa, T. Aso, M. Tanaka, S. Oshiro, H. Ichijo, and S. Kitajima. 2000. Homocysteine-responsive ATF3 gene expression in human vascular endothelial cells: activation of c-Jun NH2terminal kinase and promoter response element. Blood 96**:**2140-2148. [PubMed] [Google Scholar]

9. Chen, B. P. C., C. D. Wolfgang, and T. Hai. 1996. Analysis of ATF3: a transcription factor induced by physiological stresses and modulated by gadd153/Chop10. Mol. Cell. Biol. 16**:**1157-1168. [PMC free article] [PubMed] [Google Scholar]

10. Chen, G., and D. V. Goeddel. 2002. TNF-R1 signaling: a beautiful pathway. Science 296**:**1634-1635. [PubMed] [Google Scholar]

11. Corbett, J. A., and M. L. McDaniel. 1992. Does nitric oxide mediate autoimmune destruction of β-cells? Possible therapeutic interventions in IDDM. Diabetes 41**:**897-903. [PubMed] [Google Scholar]

12. Dinarello, C. A. 1998. Interleukin-1, p. 35-72. In A. W. Thomson (ed.), The cytokine handbook, 3rd ed. Academic Press, New York, N.Y.

13. Donath, M. Y., D. J. Gross, E. Cerasi, and N. Kaiser. 1999. Hyperglycemia-induced β-cell apoptosis in pancreatic islets of Psammomys obesus during development of diabetes. Diabetes 48**:**738-744. [PubMed] [Google Scholar]

14. Eizirik, D. L., M. Flodström, A. E. Karlsen, and N. Welsh. 1996. The harmony of the spheres: inducible nitric oxide synthase and related genes in pancreatic beta cells. Diabetologia 39**:**875-890. [PubMed] [Google Scholar]

15. Eizirik, D. L., and T. Mandrup-Poulsen. 2001. A choice of death—the signal-transduction of immune-mediated beta-cell apoptosis. Diabetologia 44**:**2115-2133. [PubMed] [Google Scholar]

16. Eizirik, D. L., and D. Pavlovick. 1997. Is there a role for nitric oxide in β-cell dysfunction and damage in IDDM? Diabet. Metab. Rev. 13**:**293-307. [PubMed] [Google Scholar]

17. Evans, J. L., I. D. Goldfine, B. A. Maddux, and G. M. Grodsky. 2003. Are oxidative stress-activated signaling pathways mediators of insulin resistance and beta-cell dysfunction? Diabetes 52**:**1-8. [PubMed] [Google Scholar]

18. Evans, J. L., I. D. Goldfine, B. A. Maddux, and G. M. Grodsky. 2002. Oxidative stress and stress-activated signaling pathways: a unifying hypothesis of type 2 diabetes. Endocrinol. Rev. 23**:**599-622. [PubMed] [Google Scholar]

19. Flodström, M., B. Tyrberg, D. L. Eizirik, and S. Sandler. 1999. Reduced sensitivity of inducible nitric oxide synthase-deficient mice to multiple low-dose streptozotocin-induced diabetes. Diabetes 48**:**706-713. [PubMed] [Google Scholar]

20. Gepts, W. 1965. Pathologic anatomy of the pancreas in juvenile diabetes mellitus. Diabetes 14**:**619-633. [PubMed] [Google Scholar]

21. Gerrish, K., M. Gannons, D. Shih, E. Henderson, M. Stoffel, C. V. E. Wright, and R. Stein. 2000. Pancreatic β cell-specific transcription of the pdx-1 gene. J. Biol. Chem. 275**:**3485-3492. [PubMed] [Google Scholar]

22. Grey, S. T., M. B. Arvelo, W. Hasenkamp, F. H. Bach, and C. Ferran. 1999. A20 inhibits cytokine-induced apoptosis and nuclear factor κB-dependent gene activation in islets. J. Exp. Med. 190**:**1135-1146. [PMC free article] [PubMed] [Google Scholar]

23. Guttridge, D. C., C. Albanese, J. Y. Reuther, R. G. Pestell, and A. S. Baldwin, Jr. 1999. NF-κB controls cell growth and differentiation through transcriptional regulation of cyclin D1. Mol. Cell. Biol. 19**:**5785-5799. [PMC free article] [PubMed] [Google Scholar]

24. Hai, T., and T. Curran. 1991. Cross-family dimerization of transcription factors Fos/Jun and ATF/CREB alters DNA binding specificity. Proc. Natl. Acad. Sci. USA 88**:**3720-3724. [PMC free article] [PubMed] [Google Scholar]

25. Hai, T., and M. G. Hartman. 2001. The molecular biology and nomenclature of the ATF/CREB family of transcription factors: ATF proteins and homeostasis. Gene 273**:**1-11. [PubMed] [Google Scholar]

26. Hai, T., F. Liu, W. J. Coukos, and M. R. Green. 1989. Transcription factor ATF cDNA clones: an extensive family of leucine zipper proteins able to selectively form DNA-binding heterodimers. Genes Dev. 3**:**2083-2090. [PubMed] [Google Scholar]

27. Hai, T., C. D. Wolfgang, D. K. Marsee, A. E. Allen, and U. Sivaprasad. 1999. ATF3 and stress responses. Gene Expr. 7**:**321-335. [PMC free article] [PubMed] [Google Scholar]

28. Hanafusa, T., J.-I. Miyagawa, H. Nakajima, K. Tomita, M. Kuwajima, Y. Matsuzawa, and S. Tarui. 1994. The NOD mouse. Diabet. Res. Clin. Pract. 24**:**S307-S311. [PubMed] [Google Scholar]

29. Herold, K. C., V. Vezys, A. Koons, D. Lenschow, C. Thompson, and J. A. Bluestone. 1997. CD28/B7 costimulation regulates autoimmune diabetes induced with multiple low doses of streptozotocin. J. Immunol. 158**:**984-991. [PubMed] [Google Scholar]

30. Hohmeier, H. E., H. Mulder, G. Chen, R. Henkel-Rieger, M. Prentki, and C. B. Newgard. 2000. Isolation of INS-1-derived cell lines with robust ATP-sensitive K+ channel-dependent and -independent glucose-stimulated insulin secretion. Diabetes 49**:**424-430. [PubMed] [Google Scholar]

31. Jiang, H. Y., S. A. Wek, B. C. McGrath, D. Lu, T. Hai, H. P. Harding, X. Wang, D. Ron, D. R. Cavener, and R. C. Wek. 2004. Activating transcription factor 3 is integral to the eukaryotic initiation factor 2 kinase stress response. Mol. Cell. Biol. 24**:**1365-1377. [PMC free article] [PubMed] [Google Scholar]

32. Jonas, J. C., A. Sharma, W. Hasenkamp, H. Ilkova, G. Patanè, R. Laybutt, S. Bonner-Weir, and G. C. Weir. 1999. Chronic hyperglycemia triggers loss of pancreatic β cell differentiation in an animal model of diabetes. J. Biol. Chem. 274**:**14112-14121. [PubMed] [Google Scholar]

33. Kleemann, R., H. Rothe, V. Kolb-Bachofen, Q. W. Xie, C. Nathan, S. Martin, and H. Kolb. 1993. Transcription and translation of inducible nitric oxide synthase in the pancreas of prediabetic BB rats. FEBS Lett. 328**:**9-12. [PubMed] [Google Scholar]

34. Kutlu, B., A. K. Cardozo, M. I. Darville, M. Kruhøffer, N. Magnunsson, T. Ørntoft, and D. L. Eizirik. 2003. Discovery of gene networks regulating cytokine-induced dysfunction and apoptosis in insulin-producing INS-1 cells. Diabetes 52**:**2701-2719. [PubMed] [Google Scholar]

35. Lee, Y., H. Hirose, M. Ohneda, J. H. Johnson, J. D. McGarry, and R. H. Unger. 1994. β-cell lipotoxicity in the pathogenesis of non-insulin-dependent diabetes mellitus of obese rats: Impairment in adipocyte-β-cell relationships. Proc. Natl. Acad. Sci. USA 91**:**10878-10882. [PMC free article] [PubMed] [Google Scholar]

36. Lee, Y. H., J. Giraud, R. J. Davis, and M. F. White. 2003. c-Jun N-terminal kinase (JNK) mediates feedback inhibition of the insulin signaling cascade. J. Biol. Chem. 278**:**2896-2902. [PubMed] [Google Scholar]

37. Liang, G., C. D. Wolfgang, B. P. C. Chen, T. H. Chen, and T. Hai. 1996. ATF3 gene: Genome organization, promoter and regulation. J. Biol. Chem. 271**:**1695-1701. [PubMed] [Google Scholar]

38. Like, A. A., M. C. Appel, R. M. Williams, and A. A. Rossini. 1978. Streptozotocin-induced pancreatic insulitis in mice. Morphologic and physiologic studies. Lab. Investig. 38**:**470-486. [PubMed] [Google Scholar]

39. Like, A. A., and A. A. Rossini. 1976. Streptozotocin-induced pancreatic insulitis: new model of diabetes mellitus. Science 193**:**415-417. [PubMed] [Google Scholar]

40. Lin, Y.-Z., S. Yao, R. A. Veach, T. R. Torgerson, and J. Hawiger. 1995. Inhibition of nuclear translocation of transcription factor NF-κB by a synthetic peptide containing a cell membrane-permeable motif and nuclear localization sequence. J. Biol. Chem. 270**:**14255-14258. [PubMed] [Google Scholar]

41. Liu, D., D. Pavlovic, M.-C. Chen, M. Flodström, S. Sandler, and D. L. Eizirik. 2000. Cytokines induce apoptosis in β-cells isolated from mice lacking the inducible form of nitric oxide synthase (iNOS−/−). Diabetes 49**:**1116-1122. [PubMed] [Google Scholar]

42. Mandrup-Poulsen, T. 1996. The role of interleukin-1 in the pathogenesis of IDDM. Diabetologia 39**:**1005-1029. [PubMed] [Google Scholar]

43. Mandrup-Poulsen, T. 2001. β-Cell apoptosis: stimuli and signaling. Diabetes 50**:**S58-S63. [PubMed] [Google Scholar]

44. Mashima, T., S. Udagawa, and T. Tsuruo. 2001. Involvement of transcriptional repressor ATF3 in acceleration of caspase protease activation during DNA damaging agent-induced apoptosis. J. Cell. Physiol. 188**:**352-358. [PubMed] [Google Scholar]

45. Mathis, D., L. Vence, and C. Benoist. 2001. Beta cell death during progression to diabetes. Nature 414**:**792-798. [PubMed] [Google Scholar]

46. Nakagomi, S., Y. Suzuki, K. Namikawa, S. Kiryu-Seo, and H. Kiyama. 2003. Expression of the activating transcription factor 3 prevents c-Jun N-terminal kinase-induced neuronal death by promoting heat shock protein 27 expression and Akt activation. J. Neurosci. 23**:**5187-5196. [PMC free article] [PubMed] [Google Scholar]

47. Nobori, K., H. Ito, M. Tamamori-Adachi, S. Adachi, Y. Ono, J. Kawauchi, S. Kitajima, F. Marumo, and M. Isobe. 2002. ATF3 inhibits doxorubicin-induced apoptosis in cardiac myocytes: a novel cardioprotective role of ATF3. J. Mol. Cell. Cardiol. 34**:**1387-1397. [PubMed] [Google Scholar]

48. Okamoto, Y., A. Chaves, J. Chen, R. Kelley, K. Jones, H. G. Weed, K. L. Gardner, L. Gangi, M. Yamaguchi, W. Klomkleaw, T. Nakayama, R. L. Hamlin, C. A. Carnes, R. A. Altschuld, J. A. Bauer, and T. Hai. 2001. Transgenic mice expressing ATF3 in the heart have conduction abnormalities and contractile dysfunction. Am. J. Pathol. 159**:**639-650. [PMC free article] [PubMed] [Google Scholar]

49. O'Neill, L. A., and C. Greene. 1998. Signal transduction pathways activated by the IL-1 receptor family: ancient signaling machinery in mammals, insects, and plants. J. Leukoc. Biol. 63**:**650-657. [PubMed] [Google Scholar]

50. Oyadomari, S., A. Koizumi, K. Takeda, T. Gotoh, S. Akira, E. Araki, and M. Mori. 2002. Targeted disruption of the Chop gene delays endoplasmic reticulum stress-mediated diabetes. J. Clin. Investig. 109**:**525-532. [PMC free article] [PubMed] [Google Scholar]

51. Oyadomari, S., K. Takeda, M. Takiguchi, T. Gotoh, M. Matsumoto, I. Wada, S. Akira, E. Araki, and M. Mori. 2001. Nitric oxide-induced apoptosis in pancreatic beta cells is mediated by the endoplasmic reticulum stress pathway. Proc. Natl. Acad. Sci. USA 98**:**10845-10850. [PMC free article] [PubMed] [Google Scholar]

52. Pierce, J. W., R. Schoenleber, G. Jesmok, J. Best, S. A. Moore, T. Collins, and M. E. Gerritsen. 1997. Novel inhibitors of cytokine-induced IκBα phosphorylation and endothelial cell adhesion molecule expression show anti-inflammatory effects in vivo. J. Biol. Chem. 272**:**21096-21103. [PubMed] [Google Scholar]

53. Prentki, M., E. Joly, W. El-Assaad, and R. Roduit. 2002. Malonyl-CoA signaling, lipid partitioning, and glucolipotoxicity: role in beta-cell adaptation and failure in the etiology of diabetes. Diabetes 51(Suppl. 3)**:**S405-S413. [PubMed] [Google Scholar]

54. Rabinovitch, A. 1998. An update on cytokines in the pathogenesis of insulin-dependent diabetes mellitus. Diabet. Metab. Rev. 14**:**129-151. [PubMed] [Google Scholar]

55. Rabinovitch, A., W. L. Suarez-Pinzon, O. Sorensen, and R. C. Bleackley. 1996. Inducible nitric oxide synthase (iNOS) in pancreatic islets of nonobese diabetic mice: identification of iNOS-expressing cells and relationships to cytokines expressed in the islets. Endocrinology 137**:**2093-2099. [PubMed] [Google Scholar]

56. Roduit, R., J. Morin, F. Masse, L. Segall, E. Roche, C. B. Newgard, F. Assimacopoulos-Jeannet, and M. Prentki. 2000. Glucose down-regulates the expression of the peroxisome proliferator-activated receptor-alpha gene in the pancreatic beta-cell. J. Biol. Chem. 275**:**35799-35806. [PubMed] [Google Scholar]

57. Rosen, P., P. P. Nawroth, G. King, W. Moller, H. J. Tritschler, and L. Packer. 2001. The role of oxidative stress in the onset and progression of diabetes and its complications: a summary of a congress series sponsored by UNESCO-MCBN, the American Diabetes Association and the German Diabetes Society. Diabetes Metab. Res. Rev. 17**:**189-212. [PubMed] [Google Scholar]

58. Rothe, H., A. Faust, U. Schade, R. Kleemann, G. Bosse, T. Hibino, S. Martin, and H. Kolb. 1994. Cyclophosphamide treatment of female non-obese diabetic mice causes enhanced expression of inducible nitric oxide synthase and interferon-gamma, but not of interleukin-4. Diabetologia 37**:**1154-1158. [PubMed] [Google Scholar]

59. Southern, C., D. Schulster, and I. C. Green. 1990. Inhibition of insulin secretion by interleukin-1β and tumor necrosis factor-α via an L-arginine-dependent nitric oxide generating mechanisms. FEBS Lett. 276**:**42-44. [PubMed] [Google Scholar]

60. Takamura, T., I. Kato, N. Kimura, T. Nakazawa, H. Yonekura, S. Takasawa, and H. Okamoto. 1998. Transgenic mice overexpressing type 2 nitric-oxide synthase in pancreatic beta cells develop insulin-dependent diabetes without insulitis. J. Biol. Chem. 273**:**2493-2496. [PubMed] [Google Scholar]

61. Thomas, H. E., W. Irawaty, R. Darwiche, T. C. Brodnicki, P. Santamaria, J. Allison, and T. W. H. Kay. 2004. IL-1 receptor deficiency slows progression to diabetes in the NOD mouse. Diabetes 53**:**113-121. [PubMed] [Google Scholar]

62. Unger, R. H. 1995. Lipotoxicity in the pathogenesis of obesity-dependent NIDDM. Genetic and clinical implications. Diabetes 44**:**863-870. [PubMed] [Google Scholar]

63. Waeber, G., J. Delplanque, C. Bonny, V. Mooser, M. Steinmann, C. Widmann, A. Maillard, J. Miklossy, C. Dina, E. H. Hani, N. Vionnet, P. Nicod, P. Boutin, and P. Froguel. 2000. The gene MAPK8IP1, encoding islet-brain-1, is a candidate for type 2 diabetes. Nat. Genet. 24**:**291-295. [PubMed] [Google Scholar]

64. Wajant, H. 2002. The Fas signaling pathway: more than a paradigm. Science 296**:**1635-1636. [PubMed] [Google Scholar]

65. Wang, J., T. Takeuchi, S. Tanaka, S.-K. Kubo, T. Kayo, D. Lu, K. Takata, A. Koizumi, and T. Izumi. 1999. A mutation in the insulin 2 gene induces diabetes with severe pancreatic β-cell dysfunction in the Mody mouse. J. Clin. Investig. 103**:**27-37. [PMC free article] [PubMed] [Google Scholar]

66. Wang, J., G. Webb, Y. Cao, and D. F. Steiner. 2003. Contrasting patterns of expression of transcription factors in pancreatic α and β cells. Proc. Natl. Acad. Sci. USA 100**:**12660-12665. [PMC free article] [PubMed] [Google Scholar]

67. Wang, X. Z., H. P. Harding, Y. Zhang, E. M. Jolicoeur, M. Kuroda, and D. Ron. 1998. Cloning of mammalian Ire1 reveals diversity in the ER stress responses. EMBO J. 17**:**5708-5717. [PMC free article] [PubMed] [Google Scholar]

68. Weston, C. R., and R. J. Davis. 2002. The JNK signal transduction pathway. Curr. Opin. Genet. Dev. 12**:**14-21. [PubMed] [Google Scholar]

69. Wolfgang, C. D., B. P. C. Chen, J. L. Martindale, N. J. Holbrook, and T. Hai. 1997. gadd153/Chop10, a potential target gene of the transcriptional repressor ATF3. Mol. Cell. Biol. 17**:**6700-6707. [PMC free article] [PubMed] [Google Scholar]

70. Wu, K.-L., M. Gannon, M. Peshavaria, M. F. Offield, E. Henderson, M. Ray, A. Marks, L. W. Gamer, C. V. E. Wright, and R. Stein. 1997. Hepatocyte nuclear factor 3β is involved in pancreatic β-cell-specific transcription of the pdx-1 gene. Mol. Cell. Biol. 17**:**6002-6013. [PMC free article] [PubMed] [Google Scholar]

71. Yuan, M., N. Konstantopoulos, J. Lee, L. Hansen, Z. W. Li, M. Karin, and S. E. Shoelson. 2001. Reversal of obesity- and diet-induced insulin resistance with salicylates or targeted disruption of Ikkβ. Science 293**:**1673-1677. [PubMed] [Google Scholar]

72. Zumsteg, U., S. Frigerio, and G. A. Hollander. 2000. Nitric oxide production and Fas surface expression mediate two independent pathways of cytokine-induced murine β-cell damage. Diabetes 49**:**39-47. [PubMed] [Google Scholar]

Articles from Molecular and Cellular Biology are provided here courtesy of Taylor & Francis