The intracellular molecular events involved in the β-cell death process are complex but poorly understood. Cytokines, e.g., interleukin (IL)-1β, may play a crucial role in inducing this process. Protein synthesis is necessary for the deleterious effect of IL-1, and induction of both protective and deleterious proteins has been described. To characterize the rather complex pattern of islet protein expression in rat islets in response to IL-1, we have attempted to identify proteins of altered expression level after IL-1 exposure by 2D gel electrophoresis and mass spectrometry. Of 105 significantly changed (i.e., up- or downregulated or de novo–induced) protein spots, we obtained positive protein identification for 60 protein spots. The 60 identifications corresponded to 57 different proteins. Of these, 10 proteins were present in two to four spots, suggesting that posttranslatory modifications had occurred. In addition, 11 spots contained more than one protein. The proteins could be classified according to their function into the following groups: 1) energy transduction; 2) glycolytic pathway; 3) protein synthesis, chaperones, and protein folding; and 4) signal transduction, regulation, differentiation, and apoptosis. In conclusion, valuable information about the molecular mechanisms involved in cytokine-mediated β-cell destruction was obtained by this approach.
Selective β-cell destruction accompanied by mononuclear cell infiltration of the islets of Langerhans (insulitis) is the hallmark of recent-onset type 1 diabetes in humans (1,2) and animal models of type 1 diabetes (e.g., the NOD mouse [3] and the BB rat [4]). Cytokines and cytotoxic T-cells are likely to be the most important mediators of selective β-cell destruction (rev. in 5,6,7). The intracellular molecular events involved in the β-cell death process are complex and poorly understood. Cytokines (interleukin [IL]-1 in particular) induce the synthesis of the enzyme inducible nitric oxide synthase (iNOS) (rev. in 8) and the expression of Fas (9) in β-cells. l-Arginine analogs, like NMMA, which prevent nitric oxide (NO) production from iNOS, partially protect β-cells from cytokine toxicity (rev. in 8), and hyperexpression of the antiapoptotic molecule Bcl-2 (10) or scavengers of oxygen free radicals (e.g., catalase, glutathione peroxidase, and Cu/Zn super oxide dismutase [SOD] [11]) markedly reduce the destruction of β-cells induced by cytokines.
Protein synthesis inhibitors (e.g., cycloheximide) effectively protect IL-1–exposed islets from destruction (12). Hence, protein synthesis is necessary for the deleterious effect of IL-1. Previous studies have shown that IL-1 induces the synthesis of members of the heat shock protein (HSP) family, like heme oxygenase (13) and HSPs 70 and 90 (14,15), and hyperexpression of HSPs in islets is partially protective against cytokine-induced β-cell destruction (16). Furthermore, it has been reported that IL-1 may induce the synthesis of unknown proteins with molecular weights of 45, 50, 75, 85, 95, and 120 kDa (17). It has previously been shown that rat islets exposed to IL-1 release NO into the culture media (18). iNOS has been cloned from islets (19) in which it has been shown to be inducible in β-cells only (20). We have further shown that IL-1 also upregulates IL-1 converting enzyme mRNA transcription (21) in rat islets. Also, SOD is shown to be upregulated in islets by cytokines (15,22).
Based on β-cell–selective toxic effects, we hypothesized that IL-1 induces a rather complex pattern of both protective and deleterious events and mechanisms in islets cells, and that in β-cells the deleterious events seem to prevail (23). We further suggested that this might be reflected at the level of islet protein expression (24). To examine this hypothesis, we used 2D gel electrophoresis to produce a database of rat islet proteins containing about 2,200 protein spots characterized by molecular weight and isoelectric point (pI). The data presented here provide the first global assessment of the IL-1–mediated β-cell–damaging processes at the protein level. We could demonstrate that IL-1 exposure of the islets in culture resulted in reproducible and statistically significant modulation of protein expression levels or de novo synthesis of 105 of these proteins. A total of 52 proteins were upregulated, 47 downregulated, and 6 synthesized de novo (25).
The aim of the present study was to positively identify these 105 proteins in order to better understand β-cell destruction and type 1 diabetes at the molecular level.
RESEARCH DESIGN AND METHODS
Reagents.
Dulbecco’s modified Eagle’s medium (DMEM), RPMI-1640, and Hanks’ balanced salt solution (HBSS) were purchased from Life Technologies (Paisley, Scotland). RPMI-1640 was supplemented with 20 mmol/l HEPES buffer, 100,000 IU/l penicillin, and 100 mg/l streptomycin. Authentic recombinant human IL-1 was provided by Novo Nordisk (Bagsværd, Denmark). The specific activity was 400 U/ng. The following other reagents were used: 2-mercaptoethanol, bovine serum albumin (BSA), Tris-HCl, Tris-base, glycine, (Sigma, St. Louis, MO); trichloroacetic acid (TCA), phosphoric acid, NaOH, glycerol, n-butanol, Bromophenol blue (Merck, Darmstadt, Germany); 35S-methionine (SJ 204, specific activity: >1,000 Ci/mmol, containing 0.1% 2-mercapthoethanol), Amplify (Amersham International, Amersham, U.K.); filters (HAWP 0.25-mm pore size; Millipore, Boston, MA); RNase A, DNase I (Worthington, Freehold, NJ); urea (ultra pure) (Schwarz/Mann, Cambridge, MA); acrylamide, N,N1-methylenebisacrylamide, TEMED, ammonium persulfate (Bio-Rad, Richmond, CA); carrier ampholytes (pH 5–7, pH 3.5–10, pH 7–9, pH 8–9.5) (Pharmacia, Uppsala, Sweden); Nonidet P-40 (BDH, Poole, UK); carrier ampholytes (pH 5–7 and sodium dodecyl sulfate) (Serva, Heidelberg, Germany); agarose (Litex, Copenhagen); ethanol (absolute 96%) (Danish Distillers, Aalborg, Denmark); methanol (Prolabo, Brione Le Blanc, France); acetic acid (technical quality, 99% glacial) (Bie & Berntsen, Århus, Denmark); and X-ray film (Curix RP-2; AGFA).
Islet isolation and culture.
For the database and assay variation experiments, 12 different islet isolations were performed, 10 for the databases, 1 for intra-assay, and 1 for interassay analysis (24). For the protein identification studies, additional islet isolations were performed (a total of ∼200,000 islets). Islets from pancreata of 4-day-old inbred Wistar Furth rats (Møllegård, Lille Skensved, Denmark) were isolated after collagenase digestion (26). After a preculture period of 4 days in RPMI-1640 plus 10% fetal calf serum, sets of 150 islets were incubated for 24 h in 300 μl RPMI-1640 plus 0.5% normal human serum (NHS) for labeling with 35S-methionine (tracer islets) (24). The remaining islets were used for preparatory gels for protein identification. In parallel experiments, 150 islets were incubated for 24 h in 300 μl RPMI-1640 plus 0.5% NSH with or without the addition of 150 pg/ml IL-1 for functional studies of NO and insulin secretion. To reduce variation, the same batches of fetal calf serum and NHS were used throughout the experiment.
Islet protein labeling.
After 24 h in culture, the 150 islets were harvested, washed twice in HBSS, and labeled for 4 h in 200 μl methionine-free Dulbecco’s modified Eagle’s medium (DMEM) with 10% NHS dialysed for amino acids and 200 μCi 35S-methionine (24). To reduce variation, the same batch of NHS has been used throughout the experiment. To eliminate 2-mercaptoethanol, 35S-methionine was freeze-dried for at least 4 h before labeling. After labeling, islets were washed thrice in HBSS, pelleted, and frozen at −80°C.
Sample preparation.
The frozen islets were resuspended in 100 μl DNase/RNase A solution and lysed by freeze-thawing twice. After the second thawing, they were left on ice for 30 min for the digestion of nucleic acids. The lysed sample was then freeze-dried overnight. The samples were dissolved by shaking in 120 μl lysis buffer (8.5 mol/l urea, 2% Nonidet P-40, 5% 2-mercaptoethanol, and 2% carrier ampholytes, pH range 7–9) for a minimum of 4 h.
Determination of 35S-methionine incorporation.
The amount of 35S-methionine incorporation was quantitated in duplicate by adding 10 μl BSA (0.2 μg/ml H2O) as a carrier to 5 μl of a 1:10 dilution of each sample, followed by 0.5 ml of 10% TCA. This was left to precipitate for 30 min at 4°C before being filtered through 0.25-μm filters. The HAWP filters were dried and placed into scintillation liquid for counting.
2D gel electrophoresis.
The procedure was essentially performed as previously described (27,28 29). Briefly, first-dimensional gels contained 4% acrylamide, 0.25% bisacrylamide, and carrier ampholytes (the actual ratio depending on the batch) and were 175 mm long and 1.55 mm in diameter. An equal number of counts (106 cpm) of each sample was applied to the gels. In case of lower amounts of radioactivity, it was necessary to regulate the exposure time of the gel so that comparable total optical densities were obtained. The samples were analyzed on both isoelectric focusing (IEF) (pH 3.5–7) and nonequilibrium pH-gradient electrophoresis (NEPHGE) (pH 6.5–10.5) gels. IEF gels were prefocused for ∼4 h at 140 μA/gel (limiting current); the sample was then applied and focused for 18 h at 1,200 V (limiting voltage). NEPHGE gels were focused for ∼6.5 h using 140 μA/gel and 1,200 V as the limiting parameters. Second-dimension gels, 1 × 200 × 185 mm, contained either 15% acrylamide and 0.075% Bis or 10% acrylamide and 0.05% Bis, and they were run overnight. This separation protocol was optimized for hydrophilic proteins; thus, a detailed characterization of hydrophobic (membrane) proteins is not possible. After electrophoresis, the gels were fixed in 45% methanol and 7.5% acetic acid for 45 min and treated for fluorography with Amplify for 45 min before being dried. The gels were placed in contact with X-ray films and exposed at −70°C for 1–40 days. Each gel was exposed for at least three time periods to compensate for the lack of dynamic range of X-ray films.
Determination of Mr and pI.
Mr and pI for individual proteins on the gels was determined by the use of internal standards. Theoretical Mr and pI were calculated using the Compute pI/Mw tool at the ExPASy Molecular Biology Server (www.expasy.ch/tools/pi_tool.html).
Protein characterization.
Preparatory 2D gels were produced from a pool of ∼200,000 neonatal WF islets that were isolated, prepared, and separated on gels as described above. For localization of the spots, radioactively labeled tracer islets were mixed with the nonlabeled islets. Because initial attempts to identify the proteins in the gel resulted in very few positive identifications by direct microsequencing, the method of choice became mass spectrometry (MS).
Protein identification by MS.
Briefly, protein spots of interest were obtained by cutting them from the dried gel using a scalpel. A total of 103 spots could technically be cut from the gels for analysis. The proteins were enzymatically digested in the gel as described (30,31), with minor modifications (32). The excised gel plugs were washed in 50 mmol/l NH4HCO3/acetonitrile (60/40) and dried by vacuum centrifugation. Modified porcine trypsin (12 ng/μl, sequencing grade; Promega) in digestion buffer (50 mmol/l NH4HCO3) was added to the dry gel pieces and incubated on ice for 1 h for reswelling. After removing the supernatant, 20–40 μl digestion buffer was added and the digestion was continued at 37°C for 4–18 h. The peptides were extracted as described (31) and dried in a vacuum centrifuge. The residue was dissolved in 5% formic acid and analyzed by matrix-assisted laser desorption/ionization (MALDI) MS. Delayed extraction MALDI mass spectra of the peptide mixtures resulting from in-gel digestion were acquired using a PerSeptive Biosystems Voyager Elite reflector time-of-flight mass spectrometer (Perseptive Biosystems, Framingham, MA). Samples were prepared using αa-cyano-4-hydroxy cinnamic acid as matrix. When appropriate, nitrocellulose was mixed with the matrix (33). Protein identification was performed by searching the peptide-mass maps in a comprehensive nonredundant protein sequence database (NRDB; European Bioinformatics Institute, Hinxton, U.K.) using the PeptideSearch software (34) further developed at EMBL (Heidelberg, Germany). The protein identifications were examined using the “second pass search” feature of the software and critical evaluation of the peptide mass map as described (35).
In cases in which protein spots to be examined by MS were present at low abundance in the 2D gels of islet material, and in cases in which results were ambiguous, confirmatory MS analysis was performed on protein spots with the same coordinates (Mr and pI) from large batches of rat insulinoma (RIN) cells treated as described for islets.
The following protein databases were searched for matches: SWISS-PROT, PIR, National Institutes of Health, and Genebank.
RESULTS
As might have been anticipated from the available amount of protein in each protein spot, which was estimated to range from <50 to 500 ng with very large variation from spot to spot, initial attempts to identify the proteins by microsequencing were of little use (data not shown). However, using MS, 57 positive protein identifications were obtained (Table 1). Six proteins (GAPDH, pyruvate kinase M, ATP synthase regulatory subunit A, acetyl-CoA acetyl transferase, mortalin [GRP75] and protein disulfide isomerase ER60) were present in two protein spots, three proteins (pyruvate kinase M2, glutamate dehydrogenase and neuroendocrine convertase two precursor) were present in three spots, and one protein (glucose-regulated protein precursor [GRP78]) was present in four spots, suggesting that these proteins may be posttranslatory modified. Nine protein spots (IEF 28, 329, 614, and 908 and NEPHGE 130, 174, 182, 231, and 269) contained two identified proteins, and two spots (IEF 387 and NEPHGE 102) contained three identified proteins. Neither positive protein identification nor useful mass spectra could be obtained from the remaining 43 protein spots.
For proteins with observed Mr’s >100 kDa (spots IEF 11, 85, 173, 186, 187, 194, 201, and 265), the calculated Mr’s are considerably lower (20–50%). This may be explained by relatively imprecise Mr determination of high–molecular weight proteins on the gels. Only minor inconsistencies are found between observed and theoretical Mr’s for proteins with molecular weights <100 kDa, with two exceptions (spots IEF 665 and NEPHGE 306). These proteins also show major inconsistencies between observed and theoretical pI values. If identifications are correct, this suggests that these proteins have been subject to posttranslational modification. Where available from the Compute pI/Mw tool at the ExPASy Molecular Biology Server, we have included the theoretical calculated values for both the native and posttranslationally modified protein.
Some protein spots (IEF 25, IEF 186, IEF 194, IEF 310, IEF 387, IEF 506, IEF 507, and NEPHGE 102) contained proteins that were identified through homology to other species (Table 1); at the time of the database search, their rat sequences were not entered.
Hence, the success rate for MALDI MS identification of proteins was 58% for positive identification. It’s beyond the space available to the present article to describe the functions and possible importance for type 1 diabetes pathogenesis for each of the identified proteins in detail. However, in Table 1, the identified proteins are assigned to broad classes according to their known or putative functions: 1) energy transduction and redox potentials; 2) glycolytic pathway; 3) protein synthesis, chaperones, and protein folding; and 4) signal transduction, regulation, differentiation, and apoptosis.
DISCUSSION
A success rate of 58% for positive identification by MS is acceptable considering the rather minute amounts of islet tissue on average in the low nanogram amounts. It demonstrates that MS can be used for large-scale screening of protein identities. The high-resolution 2D gel technology can effectively separate proteins and offers the possibility of relative quantification of changes in protein expression as well as identification of certain posttranslatory protein modifications (e.g., induced by cleavage and/or chemically like phosphorylation) induced by interventions. Combined 2D gel and MS technology may be useful tools for dynamic studies of molecular processes underlying complex disease processes, like the ones leading to β-cell destruction and type 1 diabetes.
Messenger RNA differential display techniques have been applied to various target tissues, including islet tissue, in an attempt to identify differences in specific mRNA expression and genes of relevance for type 1 diabetes as well as for type 2 diabetes (36,37 38). However, in contrast to mRNA (cDNA) expression studies, protein expression studies allow assessment of posttranslational modifications. Posttranslational modifications are often necessary for protein function and, hence, may also be of pathogenetic relevance. Thus, for the time being, the combination of high-resolution 2D gel technology and MS seems to be the method of choice for investigating disease processes.
Several studies have documented the deleterious functional and morphologic effects of IL-1β on isolated rat islets in vitro (5,6,7) and the finding that both protein synthesis and suppression (25) take place in islets exposed to IL-1 in vitro. On the basis of these and other observations, we proposed (5) that cytokines, and IL-1 in particular, were responsible for initiation of the processes producing spontaneous type 1 diabetes in animals and humans. We further hypothesized that islet exposure to cytokines would induce a complex response pattern in islets and β-cells comprising protective (e.g., upregulation of stress proteins) as well as deleterious (e.g., iNOS induction and NO production) events. The findings presented here support this hypothesis. Islet function as measured by NO production and insulin release was impaired by exposure to IL-1 (data not shown) in complete accordance with previous studies from our group (24). As evidenced in Table 1, expression of several proteins of known or putative importance for glycolysis and mitochondrial energy production, gene transcription, protein synthesis, and apoptosis are affected by cytokine exposure.
In Table 1, we have grouped the identified proteins according to the cellular pathways in which they may be involved. In the following sections, only some of these findings are discussed in relation to previous observations of cytokine effects on islets. Proteins that were identified in this study are indicated by italic letters.
Energy transduction and redox potentials.
It is well documented that IL-1 has an inhibitory effect on mitochondrial energy production in islets (39,40). Here, two components of the five, which form the mitochondrial ATP synthase, were identified: the regulatory subunit A and the catalytic subunit B. Both are strongly downregulated in response to IL-1 (by the ratios of 0.21 and 0.29, respectively).
Adenylate kinase, another mitochondrial enzyme (outer compartment), which can interconvert ATP and AMP to ADP, was also strongly downregulated (integrated optical density ratio 0.09). Thus, there appears to be less ATP synthesized or available in the mitochondria, in accordance with previous observations (41,42).
In contrast, both the ATPase of the transitional endoplasmic reticulum and the vacuolar ATPase are upregulated.
Further, acetyl-CoA acetyl transferase and 3-hydroxyacyl-CoA dehydrogenase, which are both involved in fatty acid oxidation, were found to be downregulated.
NADH cytochrome B5 reductase, which is located to the cytoplasmic side of the endoplasmic reticulum and mitochondrial outer membranes, is known to be involved in the desaturation and elongation of fatty acids, and cholesterol biosynthesis was shown to be strongly downregulated.
The glycolytic pathway.
Whereas cytokines have been shown to decrease the mitochondrial oxidation of glucose, the effect on the glycolytic pathway is not clear (6). It has been described that cytokines neither decrease function of the glycolytic pathway nor affect the activity of several key glycolytic enzymes (41,42). How this observation correlates to the changed expression of nine enzymes in the glycolytic pathway in response to IL-1 found in the present study remains to be clarified (Table 1).
Several of the enzymes involved in substrate metabolism have been demonstrated to have their synthesis and activity regulated by cellular ATP levels (42). This may in part explain the observed changes in expression levels of proteins involved in the metabolic pathways.
Protein synthesis, chaperones, and protein folding.
Several studies have shown IL-1 to influence gene transcription and protein synthesis (12,14,25,43) and have shown IL-1 to cause a fall in the overall rate of protein synthesis and of preproinsulin biosynthesis in particular (44).
Three separate components of the TCP-1 T-complex were identified (gamma, epsilon, and zeta). TCP-1 (CCT) is a ring shaped 950-kDa complex of eight polypeptides encoded by different genes. This complex functions as a type II chaperone and works specifically to fold and produce native actin, tubulin, and probably a few other proteins (45).
Formation and maintenance of the actin cytoskeleton is important, and proteins with functions in this context were identified. Caldesmon is a protein that binds both actin and myosin and promotes the binding of yet another protein detected here, tropomyosin gamma, to the actinomyosin bundles. Annexin II has been shown to interact with CD44 (one of the major cell surface receptors for hyaluronic acid) in the formation of cholesterol-rich lipid rafts (46). Annexin II is located on the cytoplasmic side of these rafts and appears to play a role in the binding to the actin bundles.
IL-1–induced altered expression levels of seven members of the HSP family were identified. HSPs are well established as molecular chaperones, with functions within folding/unfolding, assembly/disassembly, and transport of protein (47). Furthermore, upregulated HSP70 expression has been demonstrated to prevent the inhibitory effect of IL-1 on islet insulin secretion and NO-induced mitochondrial impairment (48,49). In the diabetes-prone BB rat, insufficient HSP70 expression has been correlated to high vulnerability to NO and oxygen radicals (50). The newly described irp94 (ischemia-responsive protein 94 kDa), a member of the HPS110 family, which is believed to be a homologue of the mouse and human apg-2 protein, is upregulated under ischemic stress (51,52). This protein was found to be strongly upregulated by IL-1 in the rat islets. Irp94 is not involved in the heat shock signaling mechanism, but is involved in neuronal stress responses following transient forebrain ischemia. Several genes and proteins are involved (including irp94), and it has been suggested that some of them may have a deleterious or protective role in the support of neuronal survival. (52). It is tempting to speculate that irp94 may play similar roles in the β-cell.
Signal transduction, regulation, differentiation, and apoptosis.
Different cytokines may induce β-cell death by different interacting pathways. Thus, the decreased amounts of phosphatidylethanol-amine-binding protein in IL-1–exposed islets may result in an increased sensitivity to tumor necrosis factor-α (53). Several studies have demonstrated that one IL-1–induced death pathway in islets and β-cells is mediated through apoptosis (54,55,56). This is in accordance with our finding of altered expression levels of several proteins involved in this complex pathway, e.g., lamins A and B (rev. in 57,58) and TGF-β receptor interacting protein (59).
The overall picture is complex and reflects the range of cellular responses to the IL-1 challenge. Thus, we do not know which protein changes may be considered “primary” or “secondary” in importance, time, and sequence. Furthermore, it has been shown that IL-1 may induce both NO-dependent as well as NO-independent β-cell impairment (8,24). In this context, we have recently demonstrated that of the 105 protein spots, which changed expression levels in response to IL-1 (25), 23 were dependent of NO production (60) (Table 1). In addition, the effect of the chemical NO donor S-nitroglutathione on protein expression was addressed, demonstrating altered expression levels of 19 islet protein spots, which were not MS identified (60). This suggests that the majority of protein changes observed in response to IL-1 are independent of NO production. It is not clear whether the plethora of changes observed might be secondary to the early effects of just one or very few proteins (e.g., one involved in mitochondrial energy generation). In addition, the cellular specificity of the observed changes in islet protein expression levels remains to be elucidated. However, most of the changes seen in the rat islets are reproduced in cytokine-exposed RIN-cells (data not shown). Our findings may offer relevant mechanisms for β-cell death after cytokine exposure.
Whether IL-1–induced changes in protein expression levels in rat islets in vitro will also reflect pathogenically important changes in β-cells in rats spontaneously developing type 1 diabetes remains to be determined. Preliminary observations using 2D gel studies of excised syngeneic islet transplants from different time points posttransplantation in BB-DP rats (61) suggest that the approach may be useful for studies of type 1 diabetes pathogenesis in vivo. Interestingly, the exquisite β-cell sensitivity to cytokine toxicity may be an acquired trait developed during β-cell maturation (62).
The data presented should be interpreted with some caution. As described, the level of expression of 105 rat islet protein spots was consistently and significantly changed after IL-1 exposure. Only these proteins have been analyzed further. In vitro 58% of these have been positively identified so far, and the cellular specificity and significance of the changes remain to be elucidated. Identification of the remaining 42% should thus consolidate the dynamic picture that is emerging. Surprisingly, some proteins previously shown to be modified by IL-1 (e.g., iNOS, insulin, GAD, and MnSOD) were not identified in the present experiments. It should be noted that protein expression changes that did not meet strict statistical criteria defined in our previous article (25) might be important. IL-1 effects, which result in posttranslational modifications (e.g., a phosphorylation without an overall change in expression level of that protein) were picked up in the present study. Some of the changes detected could also be due to changes in the rate of protein turnover. Furthermore, it should be noted that the protein expression changes were studied using only one IL-1β concentration and one set of experimental conditions. Using various concentrations of IL-1β or combinations of cytokines as well as incubation and labeling periods of different length may well produce additional results describing an even more complicated and dynamic picture. Nevertheless, the data presented represent the hitherto most detailed and complex picture of the molecular processes leading to β-cell destruction in vitro. Although the picture is complicated and far from complete, we are looking at the ailing β-cell through a new window, and the challenge now is to learn to fully understand what we see.
Gel spot no. . | % IOD . | IOD ratio . | Protein . | Database accession no. . | Mr . | Theoretical Mr . | pI . | Theoretical pI . |
---|---|---|---|---|---|---|---|---|
I. Energy transduction and redox potentials | ||||||||
NEPHGE 171/176† | 0.41/0.91 | 0.21/0.21 | ATP synthase regulatory subunit (A) | P15999 | 52.8/52.6 | 58.8/55.3 | 8.2/8.0 | 9.2/8.3 |
NEPHGE 174 | 0.27 | 0.43 | Rat mitochondrial H+-ATP synthase alpha subunit | J05266 | 53.8 | 58.8 | 7.9 | 9.2 |
IEF 614 | 0.93 | 0.27 | ATP synthase catalytic subunit (B) | P10719 | 52.9 | 56.4/51.7 | 4.8 | 5.2/5.0 |
NEPHGE 306 | 0.29 | 0.09 | Adenylate kinase isoenzyme 2 (mitochondrial) | P29410 | 35.7 | 26.2 | 8.1 | 6.4 |
IEF 265 | 0.05 | 3.31 | Transitional endoplasmic reticulum ATPase | P46462 | 120.0 | 89.3 | 5.0 | 5.1 |
IEF 25 | 0.16 | 1.79 | Vacuolar ATP synthase subunit B, brain isoform, bovine | P31408 | 54.2 | 56.6 | 5.5 | 5.7 |
NEPHGE 102 | 0.26 | 0.45 | 5-aminoimidazole-4-carboxamid ribonucleotide formyl transferase IMP cyclohydrolase | BAA22837 | 63.6 | 64.2 | 7.3 | 6.7 |
NEPHGE 227/231 | 0.34/0.19 | 0.29/0.41 | Acetyl-CoA acetyl transferase (ACAT2) | P17764 | 43.9/44.4 | 44.7/41.4 | 8.4/8.3 | 8.9/8.4 |
NEPHGE 182 | 1.78 | 0.55 | l-3-hydroxyacyl-CoA dehydrogenase | ADD42162 | 54.1 | 34.4 | 7.6 | 9.4 |
NEPHGE 296† | 5.80 | 0.07 | NADH-cytochrome B5 reductase | P20070 | 36.2 | 34.0 | 8.3 | 8.6 |
IEF 329 | 0.10 | 2.23 | NADPH-cytochrome P450 reductase | P00388 | 72.9 | 76.8 | 5.3 | 5.3 |
NEPHGE 231 | 0.19 | 0.41 | Creatin kinase (ubiquitous mitochondrial form) | P25809 | 44.3 | 47.0/43.1 | 8.3 | 8.7/7.8 |
NEPHGE 18/174/182 | 3.57/0.27/1.78 | 0.27/0.43/0.55 | Glutamate dehydrogenase (GDH) | P10860 | 36.4/53.8/54.1 | 61.4/58.8/55.9 | 8.4/7.9/7.6 | 8.1/9.2/6.7 |
NEPHGE 130 | 0.24 | 0.45 | Methylmalonate-semialdehyde dehydrogenase | Q02253 | 55.7 | 57.8 | 8.1 | 8.5 |
II. Glycolytic pathway | ||||||||
NEPHGE 169 | 0.28 | 2.16 | 6-phospho fructo-2-kinase | P07953 | 55.6 | 54.6 | 8.2 | 6.3 |
NEPHGE 668† | DN* | Fructose 1,6-biphosphate aldolase A | P05065 | 42.8 | 39.2 | 8.5 | 8.4 | |
NEPHGE 334 | 0.57 | 0.11 | Triose phosphate isomerase (TPI1) | P48500 | 30.9 | 26.8 | 8.2 | 6.5 |
NEPHGE 17†/269 | 0.40/0.15 | 5.09/9.04 | Glyceraldehyde-3-p-dehydrogenase (GAPDH) | P04797 | 40.0/40.0 | 35.7 | 8.3/8.0 | 8.4 |
NEPHGE 203 | 1.64 | 1.51 | Phospho-glycerate kinase | P16617 | 44.4 | 44.4 | 8.2 | 7.5 |
IEF 908† | 0.08 | 0.27 | Phospho-glycerate mutase (PGAM), brain form | P25113 | 25.8 | 28.5 | 6.3 | 6.2 |
IEF 561 | 0.12 | 2.52 | Enolase α/1 | P04764 | 49.3 | 47.0 | 6.0 | 6.2 |
NEPHGE 102/130 | 0.26/0.24 | 0.45/0.45 | Pyruvate kinase M (processed pseudogene) | M24361 | 63.6/55.7 | 141.4 | 7.3/8.1 | 4.9 |
NEPHGE 1/123/129 | 0.43/0.72/0.62 | 0.17/0.29/0.32 | Pyruvate kinase M2, early fetal tissue | P11981 | 58.8/57.0/57.6 | 57.6 | 8.3/8.2/7.7 | 7.4 |
IEF 85 | 0.004 | 4.21 | Pyruvate carboxylase | P52873 | 164.1 | 127.4 | 6.3 | 6.1 |
III. Protein synthesis, chaperones, and protein folding | ||||||||
IEF 11 | 0.03 | 2.34 | Elongation factor 2 (EF2, polypeptidyl–tRNA translocase) | P05197 | 118.7 | 95.3 | 6.6 | 6.4 |
IEF 28† | 0.11 | 2.65 | Heterogeneous nuclear ribonucleoprotein K/ROK | S41495 | 65.8 | 60.0 | 5.1 | 5.4 |
NEPHGE 156† | 0.01 | 3.22 | Polypyrimidine tract-binding protein (PTB) | Q00438 | 55.6 | 59.3 | 8.7 | 9.2 |
IEF 173 | 0.05 | 2.47 | Major vault protein (RNP protein for nucleocytoplasmic transport) | Q62667 | 136.7 | 98.5 | 5.4 | 5.7 |
IEF 310 | 0.10 | 1.88 | Glycyl-tRNA synthetase (human) | P41250 | 69.4 | 77.5 | 5.7 | 5.9 |
IEF387 | 0.15 | 2.41 | T-complex protein 1 (γ-subunit), human/mouse | P80318 | 64.3 | 60.6 | 6.1 | 6.3 |
IEF 506 | 0.17 | 2.34 | T-complex protein 1 (ε-subunit), mouse | P80316 | 60.0 | 59.6 | 5.4 | 5.7 |
NEPHGE 102 | 0.26 | 0.45 | T-complex protein 1 (ζ-subunit), mouse | P80317 | 63.6 | 58.0 | 7.3 | 6.6 |
NEPHGE 326 | 0.09 | 0.17 | Caldesmon, human | Q05682 | 34.5 | — | 8.6 | — |
IEF 895† | 0.11 | 0.69 | Tropomyosin NM4 (TMP-γ) | L24777 | 28.9 | 28.7 | 4.3 | 4.8 |
NEPHGE 269 | 0.15 | 9.04 | Annexin II | Q07936 | 39.9 | 38.5 | 8.0 | 7.5 |
IEF 201 | 0.27 | 2.48 | Calnexin | P35565 | 143.6 | — | 4.1 | — |
IEF 187† | 0.02 | 6.19 | Ischemia-responsive protein 94 kDa (irp94) | AAC27937 | 152.1 | 94.1 | 5.0 | 5.1 |
IEF 267 | 1.69 | 2.12 | HSP90-β (HSP84) | P34058 | 92.5 | 83.2 | 4.8 | 5.1 |
IEF 425† | 0.55 | 2.09 | HSP71c (heat shock cognate 71-kDa protein) | P08109 | 67.2 | 70.9 | 5.2 | 5.4 |
IEF 186 | 0.01 | 22.00 | HSP70KD protein AGP-2 (mouse) | Q61316 | 152.1 | 94.1 | 5.0 | 5.2 |
IEF 507† | 0.16 | 2.72 | Mitochondrial matrix protein p1 (HSP60), mouse | P19226 | 59.9 | 61.0 | 5.3 | 5.9 |
IEF 329/339 | 0.10/0.18 | 2.23/0.16 | Glucose-regulated protein precursor-HSPA5 (GRP78) | P06761 | 72.9/72.6 | 72.3 | 5.3/5.1 | 5.1 |
IEF 344/347 | 0.09/0.46 | 0.44/4.05 | 73.6/77.7 | 5.0/4.6 | ||||
IEF 330/340† | 0.08/0.16 | 2.74/0.32 | Mortalin (GRP75) | P48721 | 68.8/71.3 | 68.8/73.9 | 5.4/5.2 | 5.5/6.0 |
IEF 483/484† | 0.17/0.17 | 1.72/0.27 | Protein disulfide isomerase (PDI) ER60 (GRP58) | P11598 | 31.2/59.2 | 56.6 | 5.7/5.9 | 5.9 |
IEF 614 | 0.93 | 0.27 | Probable protein disulfide isomerase P5 | Q63081 | 52.9 | 47.2 | 4.8 | 5.0 |
IEF 908† | 0.08 | 0.27 | ER protein (ERP31 precursor to ERP29) | P52555 | 25.8 | 28.6 | 6.3 | 6.2 |
IEF 387 | 0.15 | 2.41 | Coatomer (δ-subunit), bovine/human | P53619 | 64.3 | 57.3 | 6.1 | 5.9 |
IEF 194 | 0.02 | 3.31 | Ubiquitin COOH-terminal hydrolase T, mouse | P56399 | 139.3 | 95.8 | 4.7 | 4.9 |
IV. Signal transduction, regulation, differentiation and apoptosis | ||||||||
IEF 941 | 0.29 | 0.08 | Phosphatidylethanolamine-binding protein (P23K) | P31044 | 22.7 | 20.8 | 5.2 | 5.5 |
IEF 655 | 0.04 | 1.85 | Lamin A (split product) | P48679 | 41.4 | 74.3 | 6.2 | 6.5 |
IEF 28† | 0.11 | 2.65 | Lamin B1 | U72353 | 65.8 | 66.6 | 5.1 | 5.2 |
IEF 759 | 0.12 | 2.40 | TGF-β receptor interacting protein 1 (TGFrip) | U36764 | 37.1 | 36.5 | 5.3 | 5.4 |
IEF 949 | 0.17 | 0.21 | 14-3-3 protein-ε-isoform | U53882 | 26.8 | 29.1 | 4.5 | 4.6 |
IEF 387 | 0.15 | 2.41 | Turned on after division (TOAD-64) | P47942 | 64.3 | 62.3 | 6.1 | 6.0 |
IEF 436/441/442 | 0.31/0.16/0.16 | 0.38/0.22/0.16 | Neuroendocrine convertase 2 precurser (NEC2) | P28841 | 66.4/66.8/66.9 | 70.8 | 4.8/4.6/4.5 | 5.9 |
IEF 665 | 0.09 | 0.33 | Metastasis associated protein (MTA-1) | Q62599 | 42.2 | 79.4 | 5.8 | 9.3 |
NEPHGE 298 | 0.70 | 24.45 | Galectin-3, MAC-2 antigen | P08699 | 36.4 | 27.1 | 8.3 | 8.6 |
Gel spot no. . | % IOD . | IOD ratio . | Protein . | Database accession no. . | Mr . | Theoretical Mr . | pI . | Theoretical pI . |
---|---|---|---|---|---|---|---|---|
I. Energy transduction and redox potentials | ||||||||
NEPHGE 171/176† | 0.41/0.91 | 0.21/0.21 | ATP synthase regulatory subunit (A) | P15999 | 52.8/52.6 | 58.8/55.3 | 8.2/8.0 | 9.2/8.3 |
NEPHGE 174 | 0.27 | 0.43 | Rat mitochondrial H+-ATP synthase alpha subunit | J05266 | 53.8 | 58.8 | 7.9 | 9.2 |
IEF 614 | 0.93 | 0.27 | ATP synthase catalytic subunit (B) | P10719 | 52.9 | 56.4/51.7 | 4.8 | 5.2/5.0 |
NEPHGE 306 | 0.29 | 0.09 | Adenylate kinase isoenzyme 2 (mitochondrial) | P29410 | 35.7 | 26.2 | 8.1 | 6.4 |
IEF 265 | 0.05 | 3.31 | Transitional endoplasmic reticulum ATPase | P46462 | 120.0 | 89.3 | 5.0 | 5.1 |
IEF 25 | 0.16 | 1.79 | Vacuolar ATP synthase subunit B, brain isoform, bovine | P31408 | 54.2 | 56.6 | 5.5 | 5.7 |
NEPHGE 102 | 0.26 | 0.45 | 5-aminoimidazole-4-carboxamid ribonucleotide formyl transferase IMP cyclohydrolase | BAA22837 | 63.6 | 64.2 | 7.3 | 6.7 |
NEPHGE 227/231 | 0.34/0.19 | 0.29/0.41 | Acetyl-CoA acetyl transferase (ACAT2) | P17764 | 43.9/44.4 | 44.7/41.4 | 8.4/8.3 | 8.9/8.4 |
NEPHGE 182 | 1.78 | 0.55 | l-3-hydroxyacyl-CoA dehydrogenase | ADD42162 | 54.1 | 34.4 | 7.6 | 9.4 |
NEPHGE 296† | 5.80 | 0.07 | NADH-cytochrome B5 reductase | P20070 | 36.2 | 34.0 | 8.3 | 8.6 |
IEF 329 | 0.10 | 2.23 | NADPH-cytochrome P450 reductase | P00388 | 72.9 | 76.8 | 5.3 | 5.3 |
NEPHGE 231 | 0.19 | 0.41 | Creatin kinase (ubiquitous mitochondrial form) | P25809 | 44.3 | 47.0/43.1 | 8.3 | 8.7/7.8 |
NEPHGE 18/174/182 | 3.57/0.27/1.78 | 0.27/0.43/0.55 | Glutamate dehydrogenase (GDH) | P10860 | 36.4/53.8/54.1 | 61.4/58.8/55.9 | 8.4/7.9/7.6 | 8.1/9.2/6.7 |
NEPHGE 130 | 0.24 | 0.45 | Methylmalonate-semialdehyde dehydrogenase | Q02253 | 55.7 | 57.8 | 8.1 | 8.5 |
II. Glycolytic pathway | ||||||||
NEPHGE 169 | 0.28 | 2.16 | 6-phospho fructo-2-kinase | P07953 | 55.6 | 54.6 | 8.2 | 6.3 |
NEPHGE 668† | DN* | Fructose 1,6-biphosphate aldolase A | P05065 | 42.8 | 39.2 | 8.5 | 8.4 | |
NEPHGE 334 | 0.57 | 0.11 | Triose phosphate isomerase (TPI1) | P48500 | 30.9 | 26.8 | 8.2 | 6.5 |
NEPHGE 17†/269 | 0.40/0.15 | 5.09/9.04 | Glyceraldehyde-3-p-dehydrogenase (GAPDH) | P04797 | 40.0/40.0 | 35.7 | 8.3/8.0 | 8.4 |
NEPHGE 203 | 1.64 | 1.51 | Phospho-glycerate kinase | P16617 | 44.4 | 44.4 | 8.2 | 7.5 |
IEF 908† | 0.08 | 0.27 | Phospho-glycerate mutase (PGAM), brain form | P25113 | 25.8 | 28.5 | 6.3 | 6.2 |
IEF 561 | 0.12 | 2.52 | Enolase α/1 | P04764 | 49.3 | 47.0 | 6.0 | 6.2 |
NEPHGE 102/130 | 0.26/0.24 | 0.45/0.45 | Pyruvate kinase M (processed pseudogene) | M24361 | 63.6/55.7 | 141.4 | 7.3/8.1 | 4.9 |
NEPHGE 1/123/129 | 0.43/0.72/0.62 | 0.17/0.29/0.32 | Pyruvate kinase M2, early fetal tissue | P11981 | 58.8/57.0/57.6 | 57.6 | 8.3/8.2/7.7 | 7.4 |
IEF 85 | 0.004 | 4.21 | Pyruvate carboxylase | P52873 | 164.1 | 127.4 | 6.3 | 6.1 |
III. Protein synthesis, chaperones, and protein folding | ||||||||
IEF 11 | 0.03 | 2.34 | Elongation factor 2 (EF2, polypeptidyl–tRNA translocase) | P05197 | 118.7 | 95.3 | 6.6 | 6.4 |
IEF 28† | 0.11 | 2.65 | Heterogeneous nuclear ribonucleoprotein K/ROK | S41495 | 65.8 | 60.0 | 5.1 | 5.4 |
NEPHGE 156† | 0.01 | 3.22 | Polypyrimidine tract-binding protein (PTB) | Q00438 | 55.6 | 59.3 | 8.7 | 9.2 |
IEF 173 | 0.05 | 2.47 | Major vault protein (RNP protein for nucleocytoplasmic transport) | Q62667 | 136.7 | 98.5 | 5.4 | 5.7 |
IEF 310 | 0.10 | 1.88 | Glycyl-tRNA synthetase (human) | P41250 | 69.4 | 77.5 | 5.7 | 5.9 |
IEF387 | 0.15 | 2.41 | T-complex protein 1 (γ-subunit), human/mouse | P80318 | 64.3 | 60.6 | 6.1 | 6.3 |
IEF 506 | 0.17 | 2.34 | T-complex protein 1 (ε-subunit), mouse | P80316 | 60.0 | 59.6 | 5.4 | 5.7 |
NEPHGE 102 | 0.26 | 0.45 | T-complex protein 1 (ζ-subunit), mouse | P80317 | 63.6 | 58.0 | 7.3 | 6.6 |
NEPHGE 326 | 0.09 | 0.17 | Caldesmon, human | Q05682 | 34.5 | — | 8.6 | — |
IEF 895† | 0.11 | 0.69 | Tropomyosin NM4 (TMP-γ) | L24777 | 28.9 | 28.7 | 4.3 | 4.8 |
NEPHGE 269 | 0.15 | 9.04 | Annexin II | Q07936 | 39.9 | 38.5 | 8.0 | 7.5 |
IEF 201 | 0.27 | 2.48 | Calnexin | P35565 | 143.6 | — | 4.1 | — |
IEF 187† | 0.02 | 6.19 | Ischemia-responsive protein 94 kDa (irp94) | AAC27937 | 152.1 | 94.1 | 5.0 | 5.1 |
IEF 267 | 1.69 | 2.12 | HSP90-β (HSP84) | P34058 | 92.5 | 83.2 | 4.8 | 5.1 |
IEF 425† | 0.55 | 2.09 | HSP71c (heat shock cognate 71-kDa protein) | P08109 | 67.2 | 70.9 | 5.2 | 5.4 |
IEF 186 | 0.01 | 22.00 | HSP70KD protein AGP-2 (mouse) | Q61316 | 152.1 | 94.1 | 5.0 | 5.2 |
IEF 507† | 0.16 | 2.72 | Mitochondrial matrix protein p1 (HSP60), mouse | P19226 | 59.9 | 61.0 | 5.3 | 5.9 |
IEF 329/339 | 0.10/0.18 | 2.23/0.16 | Glucose-regulated protein precursor-HSPA5 (GRP78) | P06761 | 72.9/72.6 | 72.3 | 5.3/5.1 | 5.1 |
IEF 344/347 | 0.09/0.46 | 0.44/4.05 | 73.6/77.7 | 5.0/4.6 | ||||
IEF 330/340† | 0.08/0.16 | 2.74/0.32 | Mortalin (GRP75) | P48721 | 68.8/71.3 | 68.8/73.9 | 5.4/5.2 | 5.5/6.0 |
IEF 483/484† | 0.17/0.17 | 1.72/0.27 | Protein disulfide isomerase (PDI) ER60 (GRP58) | P11598 | 31.2/59.2 | 56.6 | 5.7/5.9 | 5.9 |
IEF 614 | 0.93 | 0.27 | Probable protein disulfide isomerase P5 | Q63081 | 52.9 | 47.2 | 4.8 | 5.0 |
IEF 908† | 0.08 | 0.27 | ER protein (ERP31 precursor to ERP29) | P52555 | 25.8 | 28.6 | 6.3 | 6.2 |
IEF 387 | 0.15 | 2.41 | Coatomer (δ-subunit), bovine/human | P53619 | 64.3 | 57.3 | 6.1 | 5.9 |
IEF 194 | 0.02 | 3.31 | Ubiquitin COOH-terminal hydrolase T, mouse | P56399 | 139.3 | 95.8 | 4.7 | 4.9 |
IV. Signal transduction, regulation, differentiation and apoptosis | ||||||||
IEF 941 | 0.29 | 0.08 | Phosphatidylethanolamine-binding protein (P23K) | P31044 | 22.7 | 20.8 | 5.2 | 5.5 |
IEF 655 | 0.04 | 1.85 | Lamin A (split product) | P48679 | 41.4 | 74.3 | 6.2 | 6.5 |
IEF 28† | 0.11 | 2.65 | Lamin B1 | U72353 | 65.8 | 66.6 | 5.1 | 5.2 |
IEF 759 | 0.12 | 2.40 | TGF-β receptor interacting protein 1 (TGFrip) | U36764 | 37.1 | 36.5 | 5.3 | 5.4 |
IEF 949 | 0.17 | 0.21 | 14-3-3 protein-ε-isoform | U53882 | 26.8 | 29.1 | 4.5 | 4.6 |
IEF 387 | 0.15 | 2.41 | Turned on after division (TOAD-64) | P47942 | 64.3 | 62.3 | 6.1 | 6.0 |
IEF 436/441/442 | 0.31/0.16/0.16 | 0.38/0.22/0.16 | Neuroendocrine convertase 2 precurser (NEC2) | P28841 | 66.4/66.8/66.9 | 70.8 | 4.8/4.6/4.5 | 5.9 |
IEF 665 | 0.09 | 0.33 | Metastasis associated protein (MTA-1) | Q62599 | 42.2 | 79.4 | 5.8 | 9.3 |
NEPHGE 298 | 0.70 | 24.45 | Galectin-3, MAC-2 antigen | P08699 | 36.4 | 27.1 | 8.3 | 8.6 |
The proteins in this table are arranged in the order that they are referred to in discussion. The gel spot numbers refer to numbers in the rat islet protein database, published by Andersen et al. (25). The % IOD represents the average of the % IODs assigned to the same spot in five independent experiments. The IOD ratio expresses the % IOD of a protein spot in 2D gels of IL-1δ exposed rat islets/% IOD of the same spot in 2D gels of rat islets not exposed to IL-1δ. Hence, a % IOD ratio above 1 indicates that the spot is upregulated by IL-1δ (25).
DN, a protein induced to synthesis de novo. Protein spots related by posttranslational modification are given in the same row. Theoretical pIs and molecular weights have been calculated for the mature proteins where applicable.
Proteins previously shown to be altered by IL-1 in an NO-dependent manner (60) are marked with †.
Article Information
This study was in part supported by the Juvenile Diabetes Foundation International (grant no. DK-96-012), the Danish Diabetes Association, the Danish Medical Research Council (grant no. 9502027), and the Danish National Science Research Council (grant no. 9601730).
The skillful technical assistance of Susanne Munch, Rikke Bonne, Lene A. Jakobsen, Andrea Lorentzen, Lotte Christensen, and Viola Mose Larsen is highly appreciated.
REFERENCES
Address correspondence and reprint requests to Dr. Peter Mose Larsen, Center for Proteome Analysis, Forskerparken Fyn, Odense, Denmark. E-mail: [email protected].
Received for publication 6 September 2000 and accepted in revised form 5 February 2001.
BSA, bovine serum albumin; DMEM, Dulbecco’s modified Eagle’s medium; HBSS, Hanks’ balanced salt solution; HSP, heat shock protein; IEF, isoelectric focusing; IL, interleukin; iNOS, inducible nitric oxide synthase; MALDI, matrix-assisted laser desorption/ionization; MS, mass spectrometry; NEPHGE, nonequilibrium pH-gradient electrophoresis; NHS, normal human serum; NO, nitric oxide; pI, isoelectric point; SOD, super oxide dismutase; TCA, trichloroacetic acid.