Free Radical Biology & Medicine 38 (2005) 1392 – 1400 www.elsevier.com/locate/freeradbiomed
Original Contribution
Analysis of differentially expressed genes in nitric oxide-exposed human monocytic cells Kyril Turpaeva,b, Ce´cile Boutona, Alexandre Dieta, Annie Glatignyc, Jean-Claude Drapiera,T a
Institut de Chimie des Substances Naturelles, Centre National de la Recherche Scientifique, 91198 Gif-sur-Yvette Cedex, France b Engelhardt Institute of Molecular Biology, Russian Academy of Sciences, Moscow, Russia c Gif-Orsay DNA Microarray Platform, Centre National de la Recherche Scientifique, Gif-sur-Yvette, France Received 7 September 2004; revised 8 December 2004; accepted 2 February 2005 Available online 23 February 2005
Abstract
S
In this study we examined the gene expression pattern of NO-dependent genes in U937 and Mono Mac 6 monocytes exposed to the synthetic NO-donor DPTA-NO using microarray technology. cDNA microarray data were validated by Northern blot analysis and S quantitative real-time PCR. This approach allowed the identification of 17 NO-sensitive genes that showed at least a twofold difference in expression, in both U937 cells and Mono Mac 6 cells exposed to 500 AM DPTA-NO for 4 h. NO-stimulated genes belong to various functional groups, including transcription factors, signaling molecules, and cytokines. Among the selected genes, 11 (ATF-4, c-maf, SGK-1, PBEF, ATPase 8, NADH dehydrogenase 4, STK6, TRAF4-associated factor 1, molybdopterin synthase, CKS1, and CIDE-B) have not been S S previously reported to be sensitive to NO. Because several NO-stimulated genes are transcription factors, we analyzed the mRNA S expression profile in U937 cells exposed to DPTA-NO for 14 h. We found that long-term NO treatment influenced transcription rates of a rather limited set of genes, including CIDE-B, BNIP3, p21/Cip1, molybdopterin synthase, and TRAF4-associated factor 1. To accelerate formation of nitrosating species, U937 cells were exposed to DPTA-NO along with suboptimal concentrations of 2-phenyl-4,4,5,5S tetramethylimidazole-1-oxyl 3-oxide (PTIO). PTIO-mediated increase in nitrosating species remarkably enhanced NO-dependent induction of IL-8, p21/Cip1, and MKP-1 and built a specific gene expression profile. D 2005 Published by Elsevier Inc. Keywords: Nitric oxide; Gene expression; Monocytes; High-density DNA arrays; Free radicals
S
Nitric oxide ( NO) is a versatile molecule involved in many cell functions [1–4]. Over the past few years, S accumulating evidence has indicated that NO targets upstream sensors involved in signaling and transcription Abbreviations: ATPase 8, ATPase subunit 8; CDK, cyclin-dependent kinase; CKS1, CDK regulatory subunit 1; DPTA-NO, dipropylenetriamine NONOate; GAPDH, glyceraldehyde-3-phosphate dehydrogenase; H-ferritin, ferritin, heavy polypeptide; HOX, heme oxygenase 1; IL-8, interleukin 8; MKP-1, MAP kinase phosphatase 1; ND4, NADH dehydrogenase subunit 4; ODQ, 1H-(1,2,4)-oxadiazole(4,3-a)quinoxalon-1-one; PBS, phosphate-buffered saline; PTIO, 2-phenyl-4,4,5,5-tetramethylimidazole1-oxyl 3-oxide; RT-PCR, real-time PCR; SGK-1, serum- and glucocorticoid-inducible protein kinase 1; SSC, saline sodium citrate; STK6, serine/ threonine kinase 6; VEGF, vascular endothelial growth factor; TNF-a, tumor necrosis factor a; TRAF4, TNF-a receptor-associated factor 4. T Corresponding author. Fax: +33 1 69 07 72 47. E-mail address:
[email protected] (J.-C. Drapier). 0891-5849/$ - see front matter D 2005 Published by Elsevier Inc. doi:10.1016/j.freeradbiomed.2005.02.002
S
factors. Major NO-sensitive regulators include HIF-1, NF-nB, AP-1, and IRPs in mammalian cells [5–7], and SoxR, NorR, and Fur in bacteria [8,9]. Many functional S consequences of exposure to NO derive from a posttranslational modification of upstream sensors. Thus, the initial steps of SNO-dependent signaling pathways involve nitrosation, nitration, or oxidation of specific targets such as soluble guanylate cyclase, membraneassociated G proteins, and tyrosine kinases, resulting in up-regulation of MAP kinase cascades and activation of dependent transcription factors [10–13]. In addition to S transcriptional activation, the effects of NO at the gene level can also be mediated by alteration of specific mRNA lifetimes, in particular by p38 MAP kinasedependent RNA-binding proteins, which either stimulate or suppress mRNA cleavage by specific nucleases [14].
K. Turpaev et al. / Free Radical Biology & Medicine 38 (2005) 1392–1400
Monocyte and macrophage mRNA expression patterns are characterized by high plasticity that mediates monocyte differentiation and stimulation of macrophage involvement in immune responses. The highly regulated process of monocyte activation is under the control of specific growth factors, colony-stimulated factors, and cytokines [15,16]. Inflammatory and immune settings induce nitric oxide synthase (NOS2) in many cell types and tissues, including human macrophages under certain conditions [17,18]. Large screening tools like microarray analysis are well suited to S help determine the NO-responsive regulators which are S S central to the functional impact of NO. The effect of NO on macrophage expression of genes has already been studied by DNA microarray analyses in murine cells stimulated for NOS induction by cytokines and/or lipS opolysaccharide [16]. The intrinsic contribution of NO was deduced from the result obtained in cells either coexposed to SNO synthase inhibitors or explanted from NOS2 knockout mice [16]. The aim of this work was to study the global genomic response of human monocytic cells directly exposed to SNO. We determined the alterations of the mRNA expression pattern in two monocytic cell lines exposed to S the NO chemical donor dipropylenetriamine NONOate (DPTA-NO)1 and identified several previously unknown SNO-dependent genes. The spectrum of SNO-sensitive genes was comparable in the two cell lines (U937 and Mono Mac S 6). In addition, we examined the shift of the NO-dependent gene expression pattern under conditions that enhance SNO conversion to higher oxides of SNO. The results obtained by the microarray approach were validated by Northern blotting and real-time quantitative PCR analyses.
Materials and methods Cell treatments and reagents U937 cells were cultured under a 5% CO2 atmosphere in RPMI 1640 supplemented with 25 mM Hepes, pH 7.3, 5% fetal calf serum, 2 mM glutamine, penicillin (100 units/ml), and streptomycin (100 Ag/ml). Mono Mac 6 cells were cultured under 5% CO2 atmosphere in RPMI 1640 supplemented with 10% fetal calf serum, 2 mM sodium glutamate, 1 mM sodium pyruvate, 8 Ag/ml bovine insulin (Sigma), nonessential amino acids (1:100) (GIBCO), penicillin (100 units/ml), and streptomycin (100 Ag/ml). Cells were grown in fresh culture medium for 16 h before the experiments. The synthetic SNO donor DPTA-NO (Cayman) was dissolved in PBS (pH 7.3) and added to culture medium to a final concentration of 500 AM. The SNO trapping agent 2-phenyl-4,4,5,5-tetramethylimidazole1-oxyl 3-oxide (PTIO) was dissolved in ethanol. In these experiments the final ethanol concentration in the culture medium did not exceed 0.05%. PTIO was added to serumcontaining culture medium 15 min before incubation with
1393
DPTA-NO. Cells were disrupted and RNA was extracted using the Qiagen RNA isolation kit according to the manufacturerTs instructions. Microarray probe preparation, hybridization, and data processing cDNA microarray slides were purchased from the Ontario Cancer Institute Microarray Center, University Health Network (Toronto, ON, Canada) and were used for determination of relative levels of mRNAs that are expressed in U937 and Mono Mac 6 cells treated with DPTA-NO or untreated. Microarray glass slides (h19K) contained 19,008 cDNA sequences encompassing both known genes and ESTs (http://www.microarrays.ca). Each cDNA sequence was represented on the slide by two spots. In each experiment we used four parallel microarray slides for examination of RNA extracted from treated and control cells. After mRNA isolation, integrity and enrichment were ensured by hybridization with radiolabeled IL-8 and glyceraldehyde-3-phosphate dehydrogenase (GAPDH) cDNA using Northern blot analysis. For RNA labeling the following experimental procedures were performed using the FairPlay Microarray labeling kit supplied by Stratagene. Briefly, 12 Ag of RNA preparation was used for synthesis of cDNA using an oligo(dT) primer (0.5 Ag), AMV reverse transcriptase (50 U), and amino allyl-modified deoxynucleotide triphosphates. Purified amino allyl-modified cDNA probes were coupled to either Cy3 or Cy5 dye (Amersham Biosciences). The fluorescent dye-labeled cDNAs were purified using microspin columns and hybridized to the arrays coated to the glass for 18 h at 658C and then washed using an automated hybridization station (Genomics Solutions) according to [19]. The slides were scanned and quantified using a GenePix 4000B microarray scanner equipped with 635 and 532 nm excitation lasers for Cy5 and Cy3, respectively, and the resulting 16-bit TIFF images were analyzed with the accompanying software (Axon Instruments, Inc., Foster City, CA, USA). Expression ratios obtained from eight replicates were averaged for each experiment that was statistically significant ( p b 0.05, according to Student t test) and either fourfold up-regulated or twofold down-regulated. Functional annotation was established by the use of NCBI databases. cDNA probes Fragments of the genes encoding GAPDH, IL-8, HOX, c-fos, c-jun, MIP-1a, VEGF, mitochondrial ATPase subunit 8 (ATPase 8), and mitochondrial NADH dehydrogenase subunit 4 (ND4) were amplified from human cDNA. Primers are listed in Table I of the Supplementary material. The resulting PCR products were verified by sequencing. To synthesize radiolabeled cDNA probes, the PCR products were purified using a Qiagen kit according to the manufacturerTs instructions and used as a template. cDNA
1394
K. Turpaev et al. / Free Radical Biology & Medicine 38 (2005) 1392–1400
inserts were labeled with [a-32P]dCTP by a random-priming method using a commercial labeling kit (Eurobio) according to the manufacturerTs instructions. The [a-32P]dCTP was obtained from Pharmacia Biotech (Orsay, France). Northern blot analysis Ten micrograms of total cellular RNA was extracted from monocytic U937 or Mono Mac 6 cells, separated by electrophoresis in denaturing 1.2% agarose–formaldehyde gel, and transferred to Hybond N membrane (Pharmacia Biotech). Equal RNA loading and membrane transfer were confirmed by hybridization with a radiolabeled GAPDH cDNA probe. Membranes were hybridized overnight at 428C with 32P-labeled probes and washed first in 2 SSC, 0.1% SDS at room temperature, then washed in 0.5 SSC, 0.1% SDS at room temperature, and finally washed in 0.1 SSC, 0.1% SDS at 428C. Hybridization signals were scanned and quantified with a PhosphorImager (Molecular Dynamics). Quantitative real-time PCR analysis Quantitative real-time PCR (RT-PCR) was performed using the Roche Light Cycler System and the FastStart DNA Master DNA SYBR Green I kit (Roche Diagnostics). Total RNA was purified and treated with 20 units/Al RNase-free DNase (Promega) for 15 min at 378C. RNA was purified from DNase using the Qiagen RNeasy kit, and 2 Ag was used to prepare cDNAs with oligo(dT) primers and AMV reverse transcriptase (Eurobio). Primers for genes encoding serine/ threonine kinase (STK6), c-jun, MAP kinase phosphatase 1 (MKP-1), ATF-4, cyclin A2, p21/Cip1, serum- and glucocorticoid-inducible protein kinase 1 (SGK-1), CKS-1, TNF-a receptor-associated factor 4 (TRAF4)-associated factor 1, c-maf, molybdopterin synthase, pre-B-cell colony-enhancing factor (PBEF), ferritin heavy polypeptide (H-ferritin), BNIP3, and CIDE-B are listed in Table I of the Supplementary material. The generation of specific PCR products was confirmed by melting-curve analysis and agarose gel electrophoresis. For comparison of transcript levels between samples, a standard curve of cycle thresholds for serial dilutions of a cDNA sample was established and then used to calculate relative amounts of each gene. Values were then normalized to the relative amounts of GAPDH cDNA. All RT-PCR measurements were performed in triplicate.
decomposes in solution and releases SNO with a half-life of 3 h at 378C at neutral pH. According to absorbance at 253 nm, about 60% of DPTA-NO has decomposed after 4 h with S simultaneous NO generation. As previously reported [20], cell treatment with 500 AM DPTA-NO caused significant increase in IL-8 and HOX mRNA levels in both U937 and Mono Mac 6 cells. As shown in Fig. 1, in U937 monocytes, the accumulation of HOX continued to rise for 12 h after exposure to DPTA-NO. Total RNA was extracted from U937 or Mono Mac 6 cells incubated with DPTA-NO and was used to prepare Cy5-labeled cDNA. As a control, RNA extracted from untreated cells was used to prepare Cy3-labeled cDNA. Fluorescently labeled cDNAs representing mRNA from DPTA-NO-treated and control cells were then hybridized to microarrays containing ~19,000 cDNAs or ESTs printed in duplicate spots. After a first selection of up-or downregulated genes, the original clones were resequenced to avoid any uncertainty with the cDNA sequences printed on the slides. We found that several clones did not match the original database, and therefore, only sequence-verified cDNAs were taken into consideration. Moreover, a cDNA S was considered a putative NO-regulated target if the hybridization signal was increased more than fourfold or decreased more than twofold and reproduced at least across three array experiments (Table 1). Therefore, each display of the mRNA expression rate was the result of at least six of eight measurements. According to these criteria, 10 genes S were differently expressed in NO-treated Mono Mac 6 cells, among which 5 genes were up-regulated and 5 were down-regulated. In U937 cells, 13 genes were up-regulated
Results Microarray analysis and identification of differentially expressed genes in U937 and Mono Mac 6 cells treated with DPTA-NO Exponentially growing U937 cells were incubated with 500 AM DPTA-NO for 4 h. DPTA-NO spontaneously
Fig. 1. Time course of HOX mRNA accumulation in U937 cells exposed to DPTA-NO. Cells were incubated for various times in the presence of 500 AM DPTA-NO. Total RNA (10 Ag/lane) was extracted and analyzed by Northern blotting. Hybridization signals were quantified with a PhosphorImager. Data were normalized to corresponding GAPDH mRNA levels and represent the means of at least three independent experiments ( p b 0.005 compared with control). The basal content of HOX mRNA in untreated cells was assigned a value of 1.
K. Turpaev et al. / Free Radical Biology & Medicine 38 (2005) 1392–1400
1395
Table 1 NO-sensitive genes that were revealed by microarray analysis in U937 or Mono Mac 6 cells exposed to DPTA-NO for 4 h
S
Gene
Accession No.
Description
Differential display
c-jun c-fos ATF-4 c-maf MKP-1 SGK-1 IL-8 PBEF MIP-1-a VEGF H-ferritin ATPase 8 ND4 p21/Cip1 cyclin A2 STK6 TRAF4-associated factor 1 molybdopterin synthase CKS1
W33116 AA019816 W52394 W52922 H29136 AF153609 W40425 AA047110 R07513 N91060 N76562 AF347004 AF347004 H83378 AA001329 H58486 T84975 R17728 H89939
Transcription factor Transcription factor Transcription factor Transcription factor Signal transduction Signal transduction Cytokine Cytokine Cytokine Growth factor Iron homeostasis Oxidative phosphorylation Oxidative phosphorylation Cell cycle negative regulator Cell cycle positive regulator Cell cycle positive regulator Signal transduction Sulfur metabolism Cell cycle positive regulator
12.2 14.6 5.3 5.8 14.4 6.2 32.2 5.5
U937
Mono Mac 6 8.5 7.0
previously observed as SGenes NO-dependent (Ref.) [21] [22]
[23]
5.2 5.0 6.1 5.4 10.0
41.1
[24]
8.6
[25] [26] [27]
10.6 0.38 0.43 0.47 0.44 0.39
[28] [29]
Genes that were found to be either up-regulated more than fourfold or down-regulated more than twofold in response to treatment with 500 AM DPTA-NO for 4 h are listed. The fold induction of each gene was determined by the GenePix software and represents the mean of at least six measurements. Blanks mean no significant changes.
after treatment with DPTA-NO. Almost all the selected genes are involved in intercellular proinflammatory communications or in cellular signaling and settings related to cell cycle, apoptosis, and gene expression. In addition to the genes listed in Table 1, our cDNA microarray analysis S indicated several other putative NO-induced genes. HowS ever, their sensitivity to NO was not confirmed by independent Northern blot hybridization or RT-PCR analysis (data not shown). S As shown in Table 1, one class of NO-responsive genes identified in our study are genes encoding transcription factors, including c-fos, c-jun, c-maf, and ATF-4, and their up-regulation is expected to influence transcription of downstream target genes. In order to explore late genomic S consequences of NO action, we performed a microarray analysis of mRNA extracted from U937 cells treated with
500 AM DPTA-NO for 14 h. A differential expression was shown for 8 genes of which 5 were up-regulated and 3 were S down-regulated by NO (Table 2). Among the 12 genes whose expression was modulated after 4-h treatment of S U937 cells with the NO donor, only 5 retained substantial alteration of their expression rates within 14 h of exposure, whereas the mRNA level of the others declined to basal levels. Conversely, greater effects on BNIP3, CIDE-B, and molybdopterin synthase mRNA expression were achieved after long-term exposure to DPTA-NO. We validated the results found by the microarray screen by Northern blot and quantitative RT-PCR analyses of the expression of genes selected in U937 and Mono Mac 6 cells. In general, Northern blotting and RT-PCR confirmed the S sensitivity to NO of most of the genes examined (Fig. 2). However, mRNA expression assessed by RT-PCR analysis
Table 2 NO-sensitive genes that were revealed by microarray analysis in U937 cells exposed to DPTA-NO for 14 h
S
Gene
Accession No.
Description
Differential display
ATF-4 MIP-1a H-ferritin p21/Cip1
R81715 R07513 N76562 H83378
4.6 6.9 4.8 4.7
BNIP3 TRAF4-associated factor 1 molybdopterin synthase CIDE-B
W15599 T84975 R17728 H82611
Transcription factor Cytokine Iron homeostasis Cell cycle negative regulator Apoptosis regulator Cytokine receptor Sulfur metabolism Apoptosis positive regulator
5.2 0.23 0.28 0.19
Previously observed as NO-dependent (Ref.)
S
[25] [27] [28] [30]
Genes that were found to be either up-regulated more than fourfold or down-regulated more than twofold in U937 cells in response to treatment with 500 AM DPTA-NO for 14 h are listed. The fold induction of each gene was determined by the GenePix software and represents the mean at least six measurements. Blank means not detected.
1396
K. Turpaev et al. / Free Radical Biology & Medicine 38 (2005) 1392–1400
cells. RT-PCR allows precise quantification of gene expression, and it was therefore possible to compare accurately the gene expression profile of the cells under various experimental conditions. In Fig. 2, U937 cells were compared to Mono Mac 6 cells, and it was clear that results were similar in both cells lines. In Fig. 3A, the difference between early and late genomic responses is stressed. Analysis and identification of differentially expressed genes in U937 cells simultaneously treated with DPTA-NO and PTIO
S
We next assessed the role of higher oxides of NO in the S determination of the expression profile of NO-dependent
Fig. 2. Validation of the gene expression profiles of U937 and Mono Mac 6 cells exposed to DPTA-NO. Change in the mRNA levels of monocytic cells was determined by Northern blot real-time PCR analyses. Transcripts of U937 and Mono Mac 6 cells were analyzed 4 h after exposure to 500 AM DPTA-NO. (A) Northern blot analysis. Total RNA (10 Ag/lane) was extracted from cells, and 32P-labeled specific cDNAs were used as probes for the different mRNAs examined. Equal RNA loading and transfer efficiency were verified by hybridization with a radiolabeled cDNA probe specific for GAPDH mRNA. The results are representative of at least two independent experiments. (B) Quantification using real-time PCR and Northern blot (IL-8 only) analyses. The mRNA levels were normalized to the level of GAPDH mRNA. Genes that showed increased expression are given positive numbers and those with decreased expression negative numbers. Results represent means of at least three independent experiments (FSD).
(Fig. 2B and Supplementary Table II) seemed to be generally two times lower than that indicated by the microarray screen. For instance, microarray analysis showed up-regulation of ATF-4, MKP-1, and VEGF genes in U937 cells treated with DPTA-NO for 4 h of 5.3-, 14.4-, and 5.2fold, respectively (Table 1), versus 1.4-, 2.7-, and 3.9-fold, respectively, as determined by RT-PCR analysis. It is of note S that the level of NO-dependent induction of MKP-1 in Mono Mac 6 cells was about 2–3 times higher than in U937
S
Fig. 3. Validation by real-time Q-PCR of selected NO-responsive genes. Cells were incubated with 500 AM DPTA-NO alone or with 100 AM PTIO and analyzed by RT-PCR or Northern blotting (IL-8 only). (A) U937 cells were exposed to 500 AM DPTA-NO for 4 or 14 h. (B) Gene expression profiles in U937 cells after exposure to DPTA-NO and PTIO. The mRNA levels were normalized to the level of GAPDH mRNA. The basal mRNA contents in control cells were assigned a value of 1. Genes that showed increased expression are given positive numbers and those with decreased expression negative numbers. Results represent the means of at least three independent experiments (FSD).
K. Turpaev et al. / Free Radical Biology & Medicine 38 (2005) 1392–1400
genes (Table 3). To this purpose, we took advantage of the S ability of PTIO to form NO2 and, after reacting with the S remaining NO, to promote N2O3 formation according to the following reactions:
SNO
S
þ PTIO Y NO2 þ PTI;
SNO2
þ
ð1Þ
SNO Y N2 O3 :
ð2Þ
According to previous studies [20,31,32], PTIO at S suboptimal concentrations can mimic the reaction of NO oxidation by molecular oxygen, thus accelerating the formation of nitrosating species. U937 cells were exposed to both 500 AM DPTA-NO and 100 AM PTIO, and genes selected by the microarray analysis were analyzed by RTPCR. As shown in Fig. 3B, addition of PTIO to DPTA-NO caused a significant increase in the expression of IL-8, p21/ Cip1, and MKP-1 genes but slightly reduced SNO-dependent induction of ATPase 8 and ND4 genes. PTIO also reduced the down-regulation of STK6 and TRAF4-associated factor 1 genes (see also Supplementary Table II).
Discussion The aim of this study was to provide insight into the S regulation by NO of genes relevant to monocyte/macrophage functions. The results describe the expression profile S of NO-sensitive genes specific for two human monocytic cell lines using a microarray analysis. To present only the most responsive candidate genes, a criterion of sensitivity to SNO was chosen as a higher than twofold difference in mRNA level. Using this criterion, we obtained a handful of SNO-regulated genes. Exposure of U937 cells to SNO for 4 h caused up-regulation of 14 genes and down-regulation of S 4 genes. NO-treated Mono Mac 6 cells exhibited a similar pattern of regulated genes, with an induction of 14 and repression of 3 genes. Among the selected genes, 11 (ATFTable 3 NO-sensitive genes revealed by microarray analysis in U937 cells treated with DPTA-NO and PTIO
S
Symbol
Accession No.
Description
Differential display
c-jun c-fos c-maf MKP-1 IL-8 MIP-1a H-ferritin p21/Cip1 cyclin A2 STK6
W33116 AA019816 W52922 H29136 W40425 R07513 N76562 H83378 AA001329 H58486
Transcription factor Transcription factor Transcription factor Signal transduction Cytokine Cytokine Iron homeostasis Cell cycle negative regulator Cell cycle positive regulator Cell cycle positive regulator
6.7 5.9 4.6 9.2 111.2 10.4 5.5 7.8 0.4 0.45
Genes that were found to be either up-regulated more than fourfold or down-regulated more than twofold in U937 cells in response to treatment with 500 AM DPTA-NO for 4 h in the presence of 100 AM PTIO are listed. The fold induction of each gene was determined by the GenePix software and represents the mean of at least six measurements.
1397
4, c-maf, SGK-1, PBEF, ATPase 8, ND4, STK6, TRAF4associated factor 1, molybdopterin synthase, CKS1, and S CIDE-B) have not been reported to be sensitive to NO. The SNO-responsive genes revealed here may be classified into several functional groups. The major group identified in our study is composed of genes encoding transcription factors, including c-maf, ATF-4, c-fos, and c-jun. c-Fos and c-Jun are the main constituents of the dimeric factor AP-1, a major activator of genes related to cell proliferation and response to stress conditions [33]. Two other members of the bZip family, c-Maf and ATF-4 factors, are negative modulators of DNA binding and activities of various transcription factors of the C/EBP, CREB, and AP-1 families [34,35]. The induction of the c-maf and ATF-4 genes may constitute a S negative feedback loop that mitigates the NO-induced S genomic effects. Another NO-sensitive regulator of cellular signaling systems is MKP-1, a phosphatase which exhibits dual catalytic activity toward phosphotyrosine and phosphothreonine residues in substrate proteins [23]. The MKP-1 gene can be induced by multiple intracellular signals, including MAP kinase cascades and the cGMP S pathway, which are well-known mediators of NO-dependent gene induction. MKP-1 in turn inactivates MAP kinases JNK, ERK, and p38 [23,36]. The second class of SNO-dependent genes identified in this screening are cell cycle regulators. It is well established that cell cycle progression results from the action of cyclindependent kinases (CDKs), which are under the control of periodic expression of their regulatory subunits [29,37]. Our S study revealed that exposure of monocytic cells to NO leads to a roughly twofold decrease in mRNA level of cyclin A2, which is a positive regulator of CDK1 and CDK2. Moreover, SNO affects expression of CDK inhibitors Cip/Kip proteins, which negatively regulate cell progression through G1 phase [29,38]. In both U937 and Mono Mac 6 monocytic cell lines, S the action of NO led to the induction of the gene encoding S the CDK2 inhibitor p21/Cip1. In addition, we show that NO down-regulated the expression of the gene encoding CKS1, a CDK-binding protein which directs ubiquitin-mediated proteolysis of p27/Kip1, an inhibitor of CDK [39]. Therefore, SNO is expected to cause stabilization of p27/Kip1 protein. A member of another group of proliferation regulators indicated by our cDNA microarray analysis is the mitotic doublespecific protein kinase STK6 (also called aurora-A) [40]. The function of STK6, which is associated with centrosomes and spindle microtubules, is to control chromosome segregation. Taken together, these data are in keeping with the well-known SNO cytostatic ability and indicate that SNO may affect all the main stages of the cell cycle. S Another class of NO-dependent genes are factors responsible for intercellular communications. Monocytic cells produce a vast array of cytokines and chemokines, S and regulation by NO of cytokine gene expression in cells of the myeloid lineage has long been documented [41, 42]. In addition to the up-regulation of gene expression of the inflammatory cytokine IL-8 and the growth factor VEGF,
1398
K. Turpaev et al. / Free Radical Biology & Medicine 38 (2005) 1392–1400
which are highly inducible by a variety of extracellular stimuli [43,44], we report the down-regulation of gene expression of TRAF4-associated factor 1. At present, the exact function of this factor remains unclear but its sensitivity S to NO may be related to the fact that monocytes are both TNF-a-producing and TNF-a-target cells. The function of another gene identified by our microarray analysis, PBEF, is not clear either. PBEF was initially referred to as a cytokinelike secreted protein ubiquitously expressed in various tissues. However, recent findings indicate that PBEF possesses nicotinamide phosphoribosyltransferase activity involved in the biosynthesis of NAD [45]. SNO has been attributed both pro- and antiapoptotic roles S [46]. Our microarray analysis indicates that NO influences various parameters controlling apoptosis in different ways. S Thus, NO up-regulates the mRNA level of proapoptotic mitochondrial protein BNIP3 (also known as NIP3), a member of the Bcl-2 family of cell death factors. BNIP3 interacts with Bcl-2, a major apoptosis suppressor, and S inhibits its functional activity [47]. Another putative NOdependent cell death suppressive pathway may be revealed by the significant inhibition of CIDE-B gene expression. Like BNIP-3, CIDE-B is associated with the mitochondrial membrane. This factor is an effector of apoptotic DFF40 endonuclease [48]. Our findings indicate that both CIDE-B down-regulation and BNIP3 up-regulation occur late after S cell treatment with NO. Another apoptosis-related factor revealed in our study is SGK-1, a serum and glucocorticoidinduced kinase which belongs to a family of serine/threonine kinases. SGK-1 is closely related to the phosphoinositidedependent protein kinase Akt. Like Akt, SGK-1 is capable of promoting cell survival in response to a variety of extracellular stimuli [49]. In brief, our data indicate that SNO may influence the activity of various signaling systems which affect cell sensitivity to proapoptotic stimuli. Interestingly, our screening study also points to the S sensitivity to NO of two mitochondrial genes, encoding ATPase subunit 8 and NADH dehydrogenase subunit 4. As recently emphasized by Nisoli et al., mammalian cells S (including U937 cells) exposed to exogenous NO exhibit higher expression of PGC-1a, NRF-1, and mtTFA, three transcription factors which control mitochondrial gene expression [50]. Increased expression or activity of these regulators may explain up-regulation of the ATPase 8 and ND4 genes. Yet in our experiments, a global up-regulation of mitochondria-coded genes is unlikely as the level of transcripts of two other mitochondrial genes, i.e., 16S rRNA S and cytochrome c oxidase subunit 1, was insensitive to NO (data not shown). S Finally, our microarray analysis indicates NO-dependent alteration in the expression rates of two genes involved in homeostasis of inorganic compounds, H-ferritin and molybdopterin synthase. H-ferritin is responsible for immoS bilization of highly deleterious free iron ion, which in NOtreated cells likely results from disruption of Fe-S proteins and enzymatic oxidation of heme prosthetic groups by HOX
[51]. Molybdopterin synthase is involved in conversion of an unstable precursor into a molybdenum-containing cofactor essential for the activity of all human molybdoenzymes, including mitochondria-associated sulfite oxidase whose deficiency results in neurological damage [52]. We then addressed the temporal profile of the U937 cell S S response to NO. In general, after 14-h exposure to the NO donor, the number of affected genes declined remarkably. VEGF, H-ferritin, and p21/Cip1 mRNA levels were similar to those seen after a short-term (4 h) exposure, whereas only BNIP3 gene expression was higher. Moreover, downS regulation by NO of TRAF4-associated factor 1, CIDEB, and molybdopterin synthase genes was significantly strengthened by long-term DPTA-NO treatment (Fig. 3A). It thus seems that modulation of apoptosis-related genes S would be a remote consequence of NO challenge. SNO itself is able to interact directly with various targets, mostly with other radicals and metalloproteins, in particular soluble guanylate cyclase. It is known that several NOresponsive genes revealed in this study are regulated by cGMP in various cells [53]. However, preliminary experiments using up to 105 M ODQ, a well-known inhibitor of soluble guanylate cyclase, suggest that the increase in c-jun, c-fos, VEGF, BNIP3, and HO-1 expression is not related to guanylate cyclase activation under our experimental conditions (not shown). Alternatively, high-output NO production and higher nitrogen oxides can regulate gene expression via cGMP-independent mechanisms, by targeting a broader range of intracellular compounds. The main targets of nitrosating species are reactive thiol-containing peptides and proteins, i.e., cysteine residues located in the vicinity of basic and acidic amino acids [54]. In this study, we also investigated alterations of the genomic response of U937 cells to nitrosation by accelerating formation of nitrosating species. In a previous study, we reported that stimulation of SNO autoxidation by PTIO led to a marked increase in the expression levels of IL-8 and HOX genes [20]. Here, PTIO S was also added to the NO donor under conditions under which it promotes N2O3 generation, as indicated by an increase in DAF nitrosation, thus mimicking nitrosative stress [20,32]. The microarray analysis and precise determination of the expression rates by Northern blot and RT-PCR S analysis showed here that PTIO markedly boosted NOdependent induction of IL-8, p21/Cip1, and MKP-1 genes. S In addition, PTIO enhanced NO-dependent induction of VEGF, H-ferritin, and c-maf genes. In contrast, the upS regulation by NO of c-fos, ATPase 8, and ND4 and downregulation of cyclin A2, TRAF4-associated factor 1, CKS1, and STK6 genes were reduced by the addition of PTIO, which suggests that these genes are dependent on pathways S sensitive to NO itself rather than to its oxidation products. Some genes (BNIP3, c-jun) were not affected by PTIO addition (Fig. 3B). The most likely explanation is that the expression of these genes is dependent on the low concentration of nitrosating species which spontaneously arise in the presence of oxygen. These data suggest that PTIO may
K. Turpaev et al. / Free Radical Biology & Medicine 38 (2005) 1392–1400
be used as a tool to discriminate clusters of genes sensitive to S different NO-derived species. As documented previously [20], it is tempting to propose that the PTIO hydrophobic property is required for achieving its transcriptional effect and that NO sensors implicated in transcription of NOsensitive genes are associated with the plasma membrane. S The pattern of NO-modified genes has recently been reported in various cells, including murine hepatocytes transfected with NOS2 [55], murine 3T3 fibroblasts exposed to S-nitroso-N-acetylpenicillamine [56], murine macrophages infected by Mycobacterium tuberculosis or stimulated with IFN-g [16], and a cytokine-stimulated rat h cell line [57]. Some NO-responsive genes identified in this study were also revealed in these previous large screenings, including IL-8, c-fos, c-jun, MIP-1a, BNIP3, and p21/ S Cip1. However, the set of NO-sensitive genes reported here, as well as the time course of their alterations, seems specific to monocytic cell lines. Thus, in contrast to data S obtained previously on NO-treated NIH3T3 cells [56], our S study did not reveal sensitivity to NO of cell cycle-related genes encoding p53-binding protein Mdm2, c-Myc transcription factor, and cyclin G. S In summary, our data demonstrate that NO has multiple effects on gene expression profiling of monocytic cells. Several distinct functional groups of SNO-sensitive genes (up- or down-regulated) were distinguished, with notably a specific pattern of late response genes. Moreover, combiS nation of a subthreshold concentration of PTIO with NO allowed discrimination between subsets of genes differently sensitive to various reactive nitrogen oxides.
Acknowledgments This study was partly supported by grants from ARC, Villejuif, France (No. 5856); the Fondation pour la Recherche Me´dicale (to K.T.); and the Russian Foundation for Basic Research (No. 03-04-49367). We thank Drs. David Rickman and Lawrence Aggerbeck for valuable advice in performing the microarray experiments.
Appendix A. Supplementary data Supplementary data for this article may be found on ScienceDirect.
References [1] Grisham, M. B.; JourdTheuil, D.; Wink, D. A. Nitric oxide. I. Physiological chemistry of nitric oxide and its metabolites: implications in inflammation. Am. J. Physiol. 276:G315 – G321; 1999. [2] Drapier, J. C.; Bouton, C. Modulation by nitric oxide of metalloprotein regulatory activities. Bioessays 7:549 – 556; 1996. [3] Bogdan, C. Nitric oxide and the regulation of gene expression. Trends Cell. Biol. 11:66 – 75; 2001.
1399
[4] Marshall, H. E.; Merchant, K.; Stamler, J. S. Nitrosation and oxidation in the regulation of gene expression. FASEB J. 14:1889 – 1900; 2000. [5] Metzen, E.; Zhou, J.; Jelkmann, W.; Fandrey, J.; Brqne, B. Nitric oxide impairs normoxic degradation of HIF-1a by inhibition of prolyl hydroxylases. Mol. Biol. Cell. 14:3470 – 3481; 2003. [6] Kim, S. J.; Hwang, S. G.; Shin, D. Y.; Kang, S. S.; Chun, J. S. p38 kinase regulates nitric oxide-induced apoptosis of articular chondrocytes by accumulating p53 via NF-nB-dependent transcription and stabilization by serine 15 phosphorylation. J. Biol. Chem. 277: 33501 – 33508; 2002. [7] Gudi, T.; Casteel, D. E.; Vinson, C.; Boss, G. R.; Pilz, R. B. NO activation of fos promoter elements requires nuclear translocation of G-kinase I and CREB phosphorylation but is independent of MAP kinase activation. Oncogene 19:6324 – 6333; 2000. [8] Nunoshiba, T.; deRojas-Walker, T.; Wishnok, J. S.; Tannenbaum, S. R.; Demple, B. Activation by nitric oxide of an oxidative-stress response that defends Escherichia coli against activated macrophages. Proc. Natl. Acad. Sci. USA 90:9993 – 9997; 1993. [9] Mukhopadhyay, P.; Zheng, M.; Bedzyk, L. A.; LaRossa, R. A.; Storz, G. Prominent roles of the NorR and Fur regulators in the Escherichia coli transcriptional response to reactive nitrogen species. Proc. Natl. Acad. Sci. USA 101:745 – 750; 2004. [10] Friebe, A.; Koesling, D. Regulation of nitric oxide-sensitive guanylyl cyclase. Circ. Res. 93:96 – 105; 2003. [11] Soum, E.; Brazzolotto, X.; Goussias, C.; Bouton, C.; Moulis, J. M.; Mattioli, T. A.; Drapier, J. C. Peroxynitrite and nitric oxide differently target the iron-sulfur cluster and amino acid residues of human iron regulatory protein 1. Biochemistry 42:7648 – 7654; 2003. [12] Williams, J. G.; Pappu, K.; Campbell, S. L. Structural and biochemical studies of p21Ras S-nitrosylation and nitric oxidemediated guanine nucleotide exchange. Proc. Natl. Acad. Sci. USA 100:6376 – 6381; 2003. [13] Vondriska, T. M.; Zhang, J.; Song, C.; Tang, X. L.; Cao, X.; Baines, C. P.; Pass, J. M.; Wang, S.; Bolli, R.; Ping, P. Protein kinase CE–Src modules direct signal transduction in nitric oxide-induced cardioprotection: complex formation as a means for cardioprotective signaling. Circ. Res. 88:1306 – 1313; 2001. [14] Dean, J. L.; Sarsfield, S. J.; Tsounakou, E.; Saklatvala, J. p38 mitogenactivated protein kinase stabilises mRNAs that contain cyclooxygenase-2 and tumour necrosis factor AU-rich elements by inhibiting deadenylation. J. Biol. Chem. 278:39470 – 39476; 2003. [15] Klimp, A. H.; de Vries, E. G.; Scherphof, G. L.; Daemen, T. A potential role of macrophage activation in the treatment of cancer. Crit. Rev. Oncol. Hematol. 44:143 – 161; 2002. [16] Ehrt, S.; Schnappinger, D.; Bekiranov, S.; Drenkow, J.; Shi, S.; Gingeras, T. S.; Gaasterland, T.; Schoolnik, G.; Nathan, C. Reprogramming of the macrophage transcriptome in response to interferon-g and Mycobacterium tuberculosis: signaling roles of nitric oxide synthase-2 and phagocyte oxidase. J. Exp. Med. 194:1123 – 1139; 2001. [17] McMicking, J.; Xie, Q. W.; Nathan, C. Nitric oxide and macrophage function. Annu. Rev. Immunol. 15:323 – 350; 1997. [18] Weinberg, J. B. Nitric oxide production and nitric oxide synthase type 2 expression by human mononuclear phagocytes: a review. Mol. Med. 4:557 – 591; 1998. [19] Rickman, D. S.; Herbert, C. J.; Aggerbeck, L. P. Optimizing spotting solutions for increased reproducibility of cDNA microarrays. Nucleic Acids Res. 31:e109; 2003. [20] Turpaev, K.; Bouton, C.; Drapier, J. C. Nitric oxide-derived nitrosating species and gene transcriptional activation in human monocytic cells. Biochemistry 43:10844 – 10850; 2004. [21] Janssen, Y. M.; Matalon, S.; Mossman, B. T. Differential induction of c-fos, c-jun, and apoptosis in lung epithelial cells exposed to ROS or RNS. Am. J. Physiol. 273:L789 – L796; 1997. [22] Morris, B. J. Stimulation of immediate early gene expression in striatal neurons by nitric oxide. J. Biol. Chem. 270:24740 – 24744; 1995.
1400
K. Turpaev et al. / Free Radical Biology & Medicine 38 (2005) 1392–1400
[23] Pervin, S.; Singh, R.; Freije, W. A.; Chaudhuri, G. MKP-1-induced dephosphorylation of extracellular signal-regulated kinase is essential for triggering nitric oxide-induced apoptosis in human breast cancer cell lines: implications in breast cancer. Cancer Res. 63:8853 – 8860; 2003. [24] Villarete, L. H.; Remick, D. G. Nitric oxide regulation of interleukin-8 gene expression. Shock 7:29 – 35; 1997. [25] Sherry, B.; Schmidtmayerova, H.; Zybarth, G.; Dubrovsky, L.; Raabe, T.; Bukrinsky, M. Nitric oxide regulates MIP-1alpha expression in primary macrophages and T lymphocytes: implications for anti-HIV-1 response. Mol. Med. 6:542 – 549; 2000. [26] Turpaev, K.; Litvinov, D.; Dubovaya, V.; Panasyuk, A.; Ivanov, D.; Prassolov, V. Induction of vascular endothelial growth factor by nitric oxide in cultured human articular chondrocytes. Biochimie 83:515 – 522; 2001. [27] Recalcati, S.; Taramelli, D.; Conte, D.; Cairo, G. Nitric oxidemediated induction of ferritin synthesis in J774 macrophages by inflammatory cytokines: role of selective iron regulatory protein-2 downregulation. Blood 91:1059 – 1066; 1998. [28] Ishida, A.; Sasaguri, T.; Kosaka, C.; Nojima, H.; Ogata, J. Induction of the cyclin-dependent kinase inhibitor p21 (Sdi1/Cip1/Waf1) by nitric oxide-generating vasodilator in vascular smooth muscle cells. J. Biol. Chem. 272:10050 – 10057; 1997. [29] Tanner, F. C.; Meier, P.; Greutert, H.; Champion, C.; Nabel, E. G.; Lqscher, T. F. Nitric oxide modulates expression of cell cycle regulatory proteins: a cytostatic strategy for inhibition of human vascular smooth muscle cell proliferation. Circulation 101:1982 – 1989; 2000. [30] Zamora, R.; Alarcon, L.; Vodovotz, Y.; Betten, B.; Kim, P. K.; Gibson, K. F.; Billiar, T. R. Nitric oxide suppresses the expression of Bcl-2 binding protein BNIP3 in hepatocytes. J. Biol. Chem. 276:46887 – 46895; 2001. [31] Akaike, T.; Yoshida, M.; Miyamoto, Y.; Sato, K.; Kohno, M.; Sasamoto, K.; Miyazaki, K.; Ueda, S.; Maeda, H. Antagonistic action of imidazolineoxyl N-oxides against endothelium-derived relaxing factor/ NO through a radical reaction. Biochemistry 32:827 – 832; 1993. [32] Espey, M. G.; Miranda, K. M.; Thomas, D. D.; Wink, D. A. Distinction between nitrosating mechanisms within human cells and aqueous solution. J. Biol. Chem. 276:30085 – 30091; 2001. [33] Herdegen, T.; Leah, J. D. Inducible and constitutive transcription factors in the mammalian nervous system: control of gene expression by Jun, Fos and Krox, and CREB/ATF proteins. Brain Res. Rev. 28:370 – 490; 1998. [34] Cao, S.; Liu, J.; Chesi, M.; Bergsagel, P. L.; Ho, I. C.; Donnelly, R. P.; Ma, X. Differential regulation of IL-12 and IL-10 gene expression in macrophages by the basic leucine zipper transcription factor c-Maf fibrosarcoma. J. Immunol. 169:5715 – 5725; 2002. [35] Hai, T.; Hartman, M. G. The molecular biology and nomenclature of the activating transcription factor/cAMP responsive element binding family of transcription factors: activating transcription factor proteins and homeostasis. Gene 25:1 – 11; 2001. [36] Ryser, S.; Massiha, A.; Piuz, I.; Schlegel, W. Stimulated initiation of mitogen-activated protein kinase phosphatase-1 (MKP-1) gene transcription involves the synergistic action of multiple cis-acting elements in the proximal promoter. Biochem. J. 378:473 – 484; 2004. [37] Stevens, C.; La Thangue, N. B. E2F and cell cycle control: a doubleedged sword. Arch. Biochem. Biophys. 412:157 – 169; 2003. [38] Bauer, P. M.; Buga, G. M.; Ignarro, L. J. Role of p42/p44 mitogenactivated-protein kinase and p21waf1/cip1 in the regulation of vascular smooth muscle cell proliferation by nitric oxide. Proc. Natl. Acad. Sci. USA 98:12802 – 12807; 2001. [39] Sato, J.; Nair, K.; Hiddinga, J.; Eberhardt, N. L.; Fitzpatrick, L. A.; Katusic, Z. S.; OTBrien, T. eNOS gene transfer to vascular smooth muscle cells inhibits cell proliferation via upregulation of p27 and p21 and not apoptosis. Cardiovasc. Res. 47:697 – 706; 2000. [40] Dutertre, S.; Descamps, S.; Prigent, C. On the role of aurora-A in centrosome function. Oncogene 21:6175 – 6183; 2002.
S
[41] Muhl, H.; Nold, M.; Chang, J. H.; Frank, S.; Eberhardt, W.; Pfeilschifter, J. Expression and release of chemokines associated with apoptotic cell death in human promonocytic U937 cells and peripheral blood mononuclear cells. Eur. J. Immunol. 29:3225 – 3235; 1999. [42] Magrinat, G.; Mason, S. N.; Shami, P. J.; Weinberg, J. B. Nitric oxide modulation of human leukemia cell differentiation and gene expression. Blood 80:1880 – 1884; 1992. [43] Holtmann, H.; Winzen, R.; Holland, P.; Eickemeier, S.; Hoffmann, E.; Wallach, D.; Malinin, N. L.; Cooper, J. A.; Resch, K.; Kracht, M. Induction of interleukin-8 synthesis integrates effects on transcription and mRNA degradation from at least three different cytokine- or stress-activated signal transduction pathways. Mol. Cell. Biol. 19:6742 – 6753; 1999. [44] Dulak, J.; Jozkowicz, A. Regulation of vascular endothelial growth factor synthesis by nitric oxide: facts and controversies. Antioxid. Redox Signaling 5:123 – 132; 2003. [45] Rongvaux, A.; Shea, R. J.; Mulks, M. H.; Gigot, D.; Urbain, J.; Leo, O.; Andris, F. Pre-B-cell colony-enhancing factor, whose expression is up-regulated in activated lymphocytes, is a nicotinamide phosphoribosyltransferase, a cytosolic enzyme involved in NAD biosynthesis. Eur. J. Immunol. 32:3225 – 3234; 2002. [46] Brqne, B. Nitric oxide: NO apoptosis or turning it ON? Cell Death Differ. 10:864 – 869; 2003. [47] Ray, R.; Chen, G.; Vande Velde, C.; Cizeau, J.; Park, J. H.; Reed, J. C.; Gietz, R. D.; Greenberg, A. H. BNIP3 heterodimerizes with Bcl2/Bcl-X(L) and induces cell death independent of a Bcl-2 homology 3 (BH3) domain at both mitochondrial and nonmitochondrial sites. J. Biol. Chem. 275:1439 – 1448; 2000. [48] Chen, Z.; Guo, K.; Toh, S. Y.; Zhou, Z.; Li, P. Mitochondria localization and dimerization are required for CIDE-B to induce apoptosis. J. Biol. Chem. 275:22619 – 22622; 2000. [49] Leong, M. L.; Maiyar, A. C.; Kim, B.; OTKeeffe, B. A.; Firestone, G. L. Expression of the serum- and glucocorticoid-inducible protein kinase, Sgk, is a cell survival response to multiple types of environmental stress stimuli in mammary epithelial cells. J. Biol. Chem. 278:5871 – 5882; 2003. [50] Nisoli, E.; Clementi, E.; Paolucci, C.; Cozzi, V.; Tonello, C.; Sciorati, C.; Bracale, R.; Valerio, A.; Francolini, M.; Moncada, S.; Carruba, M. O. Mitochondrial biogenesis in mammals: the role of endogenous nitric oxide. Science 299:896 – 899; 2003. [51] Kim, S.; Ponka, P. Nitrogen monoxide-mediated control of ferritin synthesis: implications for macrophage iron homeostasis. Proc. Natl. Acad. Sci. USA 99:12214 – 12239; 2002. [52] Reiss, J. Genetics of molybdenum cofactor deficiency. Hum. Genet. 106:157 – 163; 2000. [53] Pilz, R. B.; Casteel, D. E. Regulation of gene expression by cyclic GMP. Circ. Res. 93:1034 – 1046; 2003. [54] Bauer, P. M.; Buga, G. M.; Fukuto, J. M.; Pegg, A. E.; Ignarro, L. J. Nitric oxide inhibits ornithine decarboxylase via S-nitrosylation of cysteine 360 in the active site of the enzyme. J. Biol. Chem. 276:34458 – 34464; 2001. [55] Zamora, R.; Vodovotz, Y.; Aulak, K. S.; Kim, P. K.; Kane, J. M.; Alarcon, L.; Stuehr, D. J.; Billiar, T. R. A DNA microarray study of nitric oxide-induced genes in mouse hepatocytes: implications for hepatic heme oxygenase-1 expression in ischemia/reperfusion. Nitric Oxide 7:165 – 186; 2002. [56] Hemish, J.; Nakaya, N.; Mittal, V.; Enikolopov, G. Nitric oxide activates diverse signaling pathways to regulate gene expression. J. Biol. Chem. 278:42321 – 42329; 2003. [57] Kutlu, B.; Cardozo, A. K.; Darville, M. I.; Kruhoffer, M.; Magnusson, N.; Orntoft, T.; Eizirik, D. L. Discovery of gene networks regulating cytokine-induced dysfunction and apoptosis in insulin-producing INS-1 cells. Diabetes 52:2701 – 2719; 2003.