Cytokine gene expression in the BB rat pancreas: natural course and impact of bacterial vaccines

Diabetologia, Nov 1996

Summary In diabetes prone BB rat pancreas the Th1/Th2 cytokine balance and the expression of inducible nitric oxide synthase (iNOS) was determined by mRNA analysis before and after the onset of insulitis. Specific mRNA was amplified by reverse transcriptase polymerase chain reaction, quantitated with radiolabelled probes by phosphoimaging and calibrated with the amount of co-amplified β -actin mRNA. At 50 days of age, prior to recognizeable insulitis, there was already significantly enhanced expression of both, Th1 and Th2 cytokines, and of iNOS mRNA, when compared to Wistar rat pancreas (p < 0.001). This supports the concept of an inconspicuous early phase of islet infiltration by single immunocytes, called single cell insulitis. At 70 days of age mononuclear infiltration of islets had begun and was associated with upregulation of interferon γ (IFNγ ) and iNOS, but downregulation of interleukin-10 and transforming growth factor β mRNA (p < 0.001). These findings correlate the onset of insulitis with a shift of the Th1/Th2 cytokine balance towards Th1 cell reactivity. Indeed there was a close correlation of the Th1/Th2 cytokine ratio but not of absolute IFNγ mRNA levels with the insulitis score. Vaccination at day 50 with tetanus toxoid did not affect cytokine gene expression while diphtheria toxoid and even more strongly BCG administration induced a shift towards Th2 reactivity (p < 0.001) while iNOS mRNA was decreased (p < 0.01). Oral dosing with immunostimulatory components of Escherichia coli also changed the quality of inflammation. Oral lipopolysaccharide (LPS) from E. coli and OM-89, an endotoxin free extract containing immunostimulatory glycolipopeptides and heat shock protein (hsp) 65, both downregulated IFNγ mRNA while only OM-89 in addition suppressed iNOS mRNA and enhanced Th2 cytokine gene expression (p < 0.001). We conclude that the onset of insulitis is associated with a shift towards Th1 cytokine and iNOS gene expression. Diphtheria toxoid and BCG vaccination stimulates Th2 reactivity but does not downregulate Th1. The latter can be achieved through oral administration of LPS or a glycopeptide fraction (OM-89) from E. coli. [Diabetologia (1996) 39: 1448–1454]

A PDF file should load here. If you do not see its contents the file may be temporarily unavailable at the journal website or you do not have a PDF plug-in installed and enabled in your browser.

Alternatively, you can download the file locally and open with any standalone PDF reader:

https://link.springer.com/content/pdf/10.1007%2Fs001250050597.pdf

Cytokine gene expression in the BB rat pancreas: natural course and impact of bacterial vaccines

Cytokine gene expression in the BB rat pancreas: natural course and impact of bacterial vaccines H. Kolb 1 U. Wo¨rz-Pagenstert 1 R. Kleemann 1 H. Rothe 1 P. Rowsell 0 F. W. Scott 0 0 Nutrition Research Division, Health Canada , Ottawa , Canada 1 Diabetes Research Institute at the Heinrich-Heine-University of Du ̈ sseldorf , Germany BB rat; insulitis; cytokines; inducible NO synthase; BCG; tetanus toxoid; lipopolysaccharide - Summary In diabetes prone BB rat pancreas the Th1/ Th2 cytokine balance and the expression of inducible nitric oxide synthase (iNOS) was determined by mRNA analysis before and after the onset of insulitis. Specific mRNA was amplified by reverse transcriptase polymerase chain reaction, quantitated with radiolabelled probes by phosphoimaging and calibrated with the amount of co-amplified -actin mRNA. At 50 days of age, prior to recognize abble insulitis, there was already significantly enhanced expression of both, Th1 and Th2 cytokines, and of iNOS mRNA, when compared to Wistar rat pancreas (p < 0.001). This supports the concept of an inconspicuous early phase of islet infiltration by single immunocytes, called single cell insulitis. At 70 days of age mononuclear infiltration of islets had begun and was associated with upregulation of interferon (IFN ) and iNOS, but downregulation of interleug kin-1g0 and transforming growth factor mRNA (p < 0.001). These findings correlate the onsb et of insulitis with a shift of the Th1/Th2 cytokine balance towards Th1 cell reactivity. Indeed there was a close correlation of the Th1/Th2 cytokine ratio but not of absolute IFN mRNA levels with the insulitis score. Received: 3 April 1996 and in final revised form: 6 September 1996 Springer-Verlag 1996 Vaccination at day 50 with tetanus toxoid did not affect cytokine gene expression while diphtheria toxoid and even more strongly BCG administration induced a shift towards Th2 reactivity (p < 0.001) while iNOS mRNA was decreased (p < 0.01). Oral dosing with immunostimulatory components of Escherichia coli also changed the quality of inflammation. Oral lipopolysaccharide (LPS) from E. coli and OM-89, an endotoxin free extract containing immunostimulatory glycolipopeptides and heat shock protein (hsp) 65, both downregulated IFN mRNA while only OM-89 in addition suppressed i NgOS mRNA and enhanced Th2 cytokine gene expression (p < 0.001). We conclude that the onset of insulitis is associated with a shift towards Th1 cytokine and iNOS gene expression. Diphtheria toxoid and BCG vaccination stimulates Th2 reactivity but does not downregulate Th1. The latter can be achieved through oral administration of LPS or a glycopeptide fraction (OM-89) from E. coli. [Diabetologia (1996) 39: 1448–1454] Recent studies in the diabetes prone non-obese diabetic (NOD) mouse have led to the concept of two contrasting qualities of islet infiltration. A benign, non-destructive form is characterised by a dominance of Th2 cell reactivities as recognized by the local expression of Th2 cytokines interleukin (IL)-4 and IL10, while there is little expression of the Th1 cytokine interferon (IFN ). In contrast, beta-cell destructive insulitis is ag ssocia tged with enhanced expression of Th1 cytokines and of the inducible nitric oxide (NO) synthase (iNOS) [ 1–4 ]. Studies of the natural or cyclophosphamide accelerated course of insulitis showed that early peri-insular infiltration of islets is dominated by Th2 cells [ 4, 5 ] while the transition to intra-insular insulitis is associated with a rise in Th1 reactivity [ 4, 6, 7 ], most probably initiated by the enhanced production of the Th1 inducing cytokine IL-12 in macrophages [ 8 ]. This shift from benign to destructive insulitis was shown to be suppressed when mice received vaccinations early in life with BCG or complete Freund’s adjuvant [ 9–12 ], treatments that inhibited Th1 cytokine expression in inflamed islets [ 1, 2 ]. Finally, islet-reactive clones of the Th1 but not of Th2 type were shown to adoptively transfer diabetes [ 13 ]. The natural course of insulitis in Bio-Breeding (BB) rats differs strongly from NOD mice in that there is no pronounced and persisting peri-insular infiltration prior to the eventual progression to intra-insulitis. Rather, small foci of peri-ductular or peri-insular infiltrates progress to a mild and transient periinsulitis. Massive infiltration by mononuclear cells is only seen during intra-insulitis [ 14–16 ]. Early infiltration stages are dominated by members of the macrophage/dendritic cell family, followed by T- and subsequently by B-lymphocytes [ 14–21 ]. We therefore compared the kinetics of Th1 and Th2 cytokine gene expression before and soon after the onset of insulitis in BB rats. Since the expression of the iNOS had been found to be linked closely to destructive insulitis in NOD mice, we included an analysis of iNOS mRNA. Similarly, as reported in NOD mice, early treatment with bacterial vaccines has been reported to abrogate or attenuate the disease process in BB rats [ 22–24 ]. The possible impact of such vaccinations on Th1/Th2 cytokine balance and iNOS gene expression was measured as a further endpoint. Materials and methods Experimental animals. Diabetes prone BB rats were kindly provided by Dr. P. Thibert, Animal Resources Division of Health Canada, Ottawa, Canada, where the stock animals were maintained under specific pathogen-free conditions. Experimental animals were kept in Ottawa and in a parallel study at the animal facilities of the Diabetes Research Institute, D u¨sseldorf, Germany, under clean conventional conditions. Routine serological analyses confirmed bacterial and viral specific pathogen-free status. The initial experimental series in Ottawa and repeat experiments in Du¨ sseldorf were without apparent differences in results obtained. Wistar rats were from the Central Animal Laboratory of the University of D u¨sseldorf. Animals received a standard, open formula NIH-07 diet (Ziegler Brothers, Gardner, PA, USA) and tap water ad libitum. Animals were kept according to the respective national guidelines and experiments were approved by the local authorities. Treatment of animals. Rats were randomized into sex and litter matched control and experimental groups. Tetanus toxoid (8 units; Behringwerke, Marburg, Germany), diphtheria toxoid (15 units, Behringwerke) or BCG (1.3 · 107 cells; Connaught, Toronto, Canada) were injected into the rear foot pad at 50 days of age. Lipopolysaccharide (1 mg, from Escherichia coli 0127 : B8; Sigma, Geisenhofen, Germany) or OM-89 (40 mg/kg of E. coli fraction; kindly provided by Dr. J.-C. Farine, Laboratoires OM, Geneva, Switzerland) were administered orally in 1 ml of phosphate buffered saline three times per week starting at 50 days of age. Control rats received vehicle only. The three vaccines and OM-89 contained less than 12 parts per million endotoxin. During the experiment, animals were monitored twice weekly for body weight and glucosuria using test sticks. At 70 days of age animals were killed. Blood glucose was determined by the hexokinase method. The pancreas was removed, split in half longitudinally and immediately snap frozen in pre-chilled isopentane and kept at - 70 C. The two halves were analysed for islet infiltration and mRNA expression. Histological analysis. Cryostat sections were prepared from three different regions of the pancreas, at least 100 m apart. For each animal, 10 islets of 100 m or more in diamm eter were analysed. Sections on coded slidm es were stained with haematoxylin and eosin and were evaluated for mononuclear infiltration after using the following semiquantitative rating system: score 0, no detectable infiltrate; score 1, small mostly focal infiltrate ( 5 mononuclear cells) in the islet periphery; score 2, infiltraKtion by more than 5 mononuclear cells but less than 20 % of the islet area infiltrated; score 3, more than 20 % of the islet area infiltrated with pronounced intra-insulitis and loss of a clear frontier between endocrine and exocrine tissue. All sections were analysed by the same individual. mRNA analysis. Total RNA was isolated by acid guanidinethiocyanate-phenol-chloroform extraction. Analysis of specific mRNA was performed by reverse transcriptase polymerase chain reaction (RT-PCR) followed by quantitation of PCR products obtained. Specific primers for iNOS [ 25, 26 ], for IFN , transforming growth factor (TGF ) (Clontech Laboratorgies, Palo Alto, Calif., USA), I Lb-10 [27]b and -actin (3¢ primer TCA TAG ATG GGC ACA GTG TG, 5b ¢ primer CTA AGG CCA ACC GTG AAA AG) were used in both steps of RT-PCR. Of total RNA 6 g was used for cDNA preparation. PCR was performed with hmot start at 78 C, followed by 5 min of denaturation at 95 C, 1 min of annealing at 60 C (IFN ), 57 C (IL-10), 55 C (iNOS), 62 C (TGF ) or 61 C ( -actgin) and 2 min of elongation phase at 72 C. I nb the followinb g cycles denaturation and elongation was for 1 min. The final cycle had 7 min of elongation. Different mRNAs were amplified in different tubes. After a total of 35 cycles, the products were separated by electrophoresis on a 2 % agarose gel. Quantitation of PCR products was done as described previously [ 3 ] by hybridization to a specific 32P labelled probe and determining 32P stimulated luminescence by phosphoimaging. Counts obtained for iNOS and cytokine mRNA were calibrated to the amount of -actin mRNA which was assigned a value of one. Two types obf control experiments were performed with representative samples. For one, samples were run for only 20, 25 or 30 cycles to check for a linear increase of PCR products. Secondly, samples were run also 1 : 3 diluted to test for reproducibility of differences between mRNA types. Statistical analysis Differences between mean values were evaluated by two-sided Student’s t-test, with a level of significance set as p less than 0.05. Regression analyses were done using Statview 4.01 software package. 1450 IFNγ IL-10 BB (50 days) IFNγ IL-10 BB (70 days) Impact of bacterial vaccines on cytokine and iNOS expression. Three different vaccines were injected into the footpad and analyses of the possible effect on insulitis quality and quantity were performed as described above. Vaccination with tetanus toxoid did not affect Th1/Th2 cytokine gene expression in BB g the rats (p < 0.05). Interestingly, only these four g observed (p < 0.001) while a decrease of IL-10 b Cytokine gene expression before and after onset of insulitis. At 50 days of age about half of the BB rat pancreata analysed had minimal signs of islet infiltration with few mononuclear cells in the periphery of some islets or adjacent ducts. By 70 days of age insulitis had developed in 8 of 15 rats with peripheral or intra-insular infiltration, while the remainder still displayed minimal or no peri-insulitis (Fig. 1). Hence, these two time points described the pancreas prior to and soon after the onset of classic, light microscopydefined mononuclear islet infiltration. These kinetics correlated well with the diabetes incidence observed; i. e. at 70 days of age 18 % of BB rats had progressed to overt diabetes. In pancreas RNA the message for IFN and IL-10 was determined in order to recognize gTh1-like as opposed to Th2-like immune cell activities. There was minimal or no cytokine mRNA detectable in Wistar rat pancreas (Fig. 2 A). Surprisingly, mRNA of IFN and IL-10 was found in significantly elevated levels ig n the pancreas of 50 day old BB compared to Wistar rats (p < 0.001) despite the near lack of recognizable insulitis. Concomitantly with the progression of mononuclear infiltration of islets by 70 days of age a significant increase of IFN mRNA levels was mRNA was seen (p < 0.001) (Fig. 2 B). In a parallel analysis we determined the expression of two genes reflecting the pro- or anti-inflammatory state of immune and also non-immune tissue cells. As shown in Figure 2 B, iNOS mRNA and TGF RNA levels were already elevated in the pancreas ob f 50 day old BB compared to Wistar rats (p < 0.001). With the onset of insulitis by 70 days of age there was a further increase in iNOS gene expression (p < 0.01) while TGF mRNA was downregulated (p < 0.01). An analysis of possible correlations between individual Th1 cytokine mRNA levels and insulitis scores was performed (Fig. 3). Surprisingly, IFN mRNA levels correlated poorly with the insulgitis score (r = 0.40, p > 0.1). However, the ratio of IFN vs IL10 mRNA closely reflected the state of isletg inflammation (r = 0.78, p < 0.001). Moreover, the four animals found to be hyperglycaemic had significantly higher ratios of IFN /IL-10 than the remainder of acutely diabetic animals showed background levels of both Th2 cytokine mRNAs, IL-10 and TGF . b brated to the amount of -actin mRNA (set as 1). Bars indiiNOS TGFβ iNOS TGFβ iNOS TGFβ Wistar BB (50 days) BB (70 days) Fig. 2. A, B Cytokine and iNOS gene expression in BB rat pancreas. Each point represents one animal. mRNA levels for RIFTN- PgC,RIL-f1o0llo(wAe)dabnyd qiNuaOnSti,taTtGioFnb of(Bra)dwioelraebedlebteyrmphinoespdhboyr stimulated luminesence (PSL). Shown are mRNA levels calicate means plus SD 4 )L 3 S P ( y itt an 2 u q A N R m 1 Wistar BB (50 days) BB (70 days) Fig. 1. Onset of insulitis in BB rats. Each point represents one animal. The average score of 10 islets is shown for each animal. Bars indicate means plus SD IFNγ IL-10 Wistar e 2 r o c s s liit u s In 1 0 A B 0 2.0 )L1.5 S P ( y it tan1.0 u q A N R m0.5 0 p rr= 0=.784, 0.784, p= 0.0=5 0.005 3.5 0 1 2 3 4 5 6 IFNγ (ng / ml) Fig. 3. Regression analysis of cytokine gene expression vs insulitis score. Each symbol represents an individual BB rat at 70 days of age rat pancreas (Fig. 4 A) nor did it alter the levels of iNOS and TGF mRNA (Fig. 4 B). The injection of diphtheria toxoibd did not alter IFN mRNA levels but significantly upregulated thg ose of IL-10 (p < 0.001) resulting in a decrease of the IFN /IL-10 ratio (Fig. 4 A). In parallel iNOS gene ex pgression was downregulated (p < 0.05) and levels of counterregulatory TGF increased (p < 0.001) (Fig. 4 B). The strongestb effects on insulitis quality were observed after BCG vaccination, with a similar gene expression pattern as seen after diphtheria toxoid. There was no change of IFN mRNA but a major upregulation of IL-10 mRN Ag (p < 0.001) (Fig. 4 A), downregulation of iNOS mRNA (p < 0.01) and TGF gene expression was increased (Fig. 4 B, p < 0.b001). In total, tetanus toxoid vaccination had no effect on pancreatic cytokine patterns while both diphtheria toxoid and BCG administration enhanced the gene expression of Th2 anti-inflammatory cytokines (IL-10, TGF ) and downregulated iNOS. In parallel, a sembi-quantitative analysis of insulitis was performed. As shown in Figure 5 A there were no significant differences between the experimental and control groups for mean insulitis scores, indicating no effect on the degree of islet inflammation. Also, no effect of vaccination on body growth was detected. 6.0 )L4.5 S P ( y itt an3.0 u q A N R m1.5 0 2.0 )L1.5 S P ( y itt an1.0 u q A N R m0.5 0 A B IFNγ IL-10 Control IFNγ IL-10 Tetanus IFNγ IL-10 DDiipphhttheeriraia IFNγ IL-10 BCG iNOS TGFβ iNOS TGFβ iNOS TGFβ iNOS TGFβ Control Tetanus DDiipphhttheeriraia BCG Fig. 4. A, B Impact of bacterial vaccines on cytokine and iNOS gene expression. Each point represents one animal. Shown are bmrRatNedAtloevtehles afomroIuFnNtg o, fIL--1a0ct(iAn)maRnNdAiNO(sSe,t TaGsF1)b a(sBd)ectaelri-mined by RT-PCR. Bars i nbdicate means plus SD Impact of orally administered E. coli fractions on cytokine and iNOS gene expression. Lipopolysaccharide or the endotoxin free glycoprotein preparation OM-89 from E. coli were given three times per week. Oral administration of the two immunostimulatory compounds had significant effects on cytokine and iNOS gene expression patterns. As shown in Figure 6 A IFN mRNA was significantly downregulated by bot hg compounds (p < 0.001). OM-89 also had a stronger impact on the two other parameters, suppressing iNOS mRNA (p < 0.001) and strongly inducing TGF gene expression (p < 0.05) (Fig. 6 B). As seen befob re for the injected bacterial vaccines there was no suppression of islet infiltration in BB rats receiving oral immunostimulation (Fig. 5 B). Control Tetanus DDiipphhttehreiaria BCG LPS OM-89 Fig. 5. A, B Impact of bacterial vaccines and E. coli fractions on the onset of insulitis. Each point represents one animal of groups injected with saline or bacterial vaccines (A) or orally treated with E. coli fractions (B). The average score of 10 islets is shown per animal. Bars indicate means plus SD A e 2 r o c s s ilit u s In 1 0 6.0 )L4.5 S P ( y it tan3.0 u q A N R m1.5 0 2.0 )1.5 L S P ( y it tna1.0 u q A N R m0.5 0 This study focused on the expression of key Th1 and Th2 cytokines and of iNOS mRNA in the BB rat pancreas before and after the onset of insulitis. IL-10 was given preference over IL-4 in our analysis since in NOD mice IL-10 but not IL-4 mRNA was found to correlate with Th2-type benign insulitis [ 4 ]. Some experiments were performed in parallel with RNA from freshly isolated islets. However, the latter approach yielded in our hands highly variable results while the analysis of total pancreatic RNA provided data correlating well with islet histology and disease state of the animal. An earlier study in NOD mice had also shown close correlation of mRNA detected in total pancreas with the immunohistochemically determined presence of cytokines or iNOS in islet sections [ 3 ]. Therefore, although cytokine and iNOS mRNA from whole pancreas also included message from non-endocrine tissue, the mRNA profile was in close accord with the histological determined state of islet inflammation. Several points render the use of RNA from isolated islets difficult. For one, the number of islets obtained from inflamed NOD mouse or BB rat pancreas is 25–50 % lower than from tissue prior to insulitis onset. This suggests loss of damaged islets and enrichment for non-infiltrated or poorly infiltrated islets. Secondly, the isolation procedure includes several centrifugation steps which will favour the loss of peri-insular and in particular peri-ductular infiltrates. Finally, mRNA levels can be rapidly up- or downregulated, and such changes may occur during the in situ digestion period with collagenase at 37 C. The time points of the study were chosen to have one sample with minimal or no insulitis (at 50 days of age 4 pancreases appeared normal, in 6 cases there was minimal peri-insular infiltration (grade 1) in one or a few islets. The next time point chosen was at 70 days of age. By this time 8 of 15 rats had progressed to grade 2 or 3 insulitis in some or all of the islets, with 4 animals being acutely diabetic. RNA samples were not pooled before analysis but determined individually so that correlations between cytokines, iNOS and insulitis grade could be tested. The data obtained reveal several characteristic features of the disease process. At 50 days of age, the elevated cytokine and iNOS mRNA levels demonstrate the presence of an inflammatory process in the absence of classic, light microscopy-defined insulitis. These data are in accord with previous reports of single cell insulitis, an inconspicuous form of islet infiltration, recognizable only with the aid of an electron microscope [ 28, 29 ], and preceding “classical” insulitis. However, constitutively elevated cytokine and iNOS mRNA levels cannot be ruled out. A further argument in support of early inflammatory changes in the observation of IFN expression in beta cells of young BB rats which alsao indicates some inflammatory changes prior to insulitis [ 30 ]. In addition, a decrease in beta-cell volume has been noted prior to the onset of classical insulitis [ 31 ]. Focal infiltrates in the exocrine pancreas have been seen in the BB rat and in man [ 14, 29, 32, 33 ] and they may contribute. It is of interest that this early stage is characterized by the expression of proinflammatory (IFN , iNOS) as well as counter-regulatory mediators (gIL-10, TGF ). With the progression towards significant peri- andb intra-insular infiltration, this balance shifts towards proinflammatory gene expression, while Th2 cytokine mRNA is decreased. To our knowledge, this is the first demonstration of a close individual correlation between the degree of insulitis and the Th1/Th2 balance. An earlier study demonstrated the expression of IL-1, IFN , TNF and IL-6 mRNA in pancreata of acutely diagbetic BBa rats by in situ hybridization [ 34 ]. There was an excess of tumour necrosis factor (TNF) positive over IL-6 positive cells during intrainsulitia s, which is compatible with a dominance of Th1 reactivity [ 35 ]. A very recent study [ 35 ] demonstrated age-dependent increases of mRNA levels of IFN and IL-12 p40 in islets of diabetes-prone BB rats,g whereas IL-10 mRNA levels in islets did not change, except for a transient increase at 7 weeks of age in diabetes-prone BB rats. Although the latter went uncommented by Zipris et al. [ 36 ], it supports our conclusion of an early contribution of Th2 reactivity followed by a dominance of Th1 as insulitis progresses. Analysis of mRNA levels of single pancreata underscored the role of the Th1/Th2 balance. There was a close correlation between the individual insulitis score and the Th1/Th2 balance found. Surprisingly, the absolute level of IFN mRNA showed only poor correlation with insuliti sg scores; i. e. animals with high IFN mRNA levels only had an advanced insulitis score ig f the level of IL-10 mRNA was low. We conclude that the immune status of an inflammatory infiltrate can only be assessed by determining the ratio of counter-regulatory cytokines or mediators. A recent study of peripheral mononuclear cells in man has also found the ratio of Th1 compared to Th2 cytokines to be the best reflection of the immunomodulatory status [ 37 ]. In addition, the ratio of IFN /IL-10 was recognized as distinguishing patients withg type I (insulin-dependent) diabetes from healthy individuals or patients with atopic dermatitis [ 38, 39 ]. The second part of our study dealt with the impact of treatment with bacterial vaccination on the cytokine balance in pancreatic lesions. Several laboratories have reported that the administration of highly immunogenic bacterial preparations or adjuvants suppressed diabetes development in BB rats or NOD mice but did not necessarily inhibit islet infiltration [ 9–12, 22–24 ]. In the NOD mice, these treatments were found to freeze the Th2 state of peripheral islet infiltration and inhibit the progression towards destructive Th1 insulitis [ 1, 4, 12 ]. The data presented here show that strongly immunogenic vaccines significantly inhibit the shift towards Th1 reactivity although the development of insulitis was not suppressed. Hence, despite major differences in the extent and kinetics of islet infiltration BB rats and NOD mice respond quite similarly to bacterial immunostimulants. Our study included five different bacterial preparations, and the comparison of results obtained offers some further insight into the mechanisms involved. Tetanus toxoid was the only compound lacking general immunostimulatory properties and also did not affect mRNA levels or ratios. Hence, it is not the immune response to a foreign (bacterial) protein which accounts for the immunoregulatory effect of the bacterial preparations. Secondly, a single subcutaneous injection of BCG or diphtheria toxoid did not “normalize” the Th1/Th2 balance by inhibiting the production of IFN mRNA but upregulated the counter-regulatory IL-g10 and TGF . As a consequence, there was a downregulation ofb inflammatory activity, as apparent from the significantly decreased iNOS mRNA levels, despite the persistence of elevated IFN mRNA levels. Inhibition of IFN gene expressiong was only reached after repeated og ral administration of lipopolysaccharide or OM-89. The E. coli extract OM-89 contains mainly acidic glycolipoproteins of 10–300 kDa molecular weight and mostly unknown nature, heat shock protein (hsp) 65 and minimal amounts of lipopolysaccharide ( < 12 ppm) [ 24, 40, 41 ]. Hsp 65 is also an immunodominant antigen of BCG. Whether the hsp 65 content of BCG and OM-89 is similar, has not been reported. OM-89 administration turned out to be more strongly anti-inflammatory than lipopolysaccharide in that there was not only downregulation of IFN mRNA but also upregulation of both Th2 cytoking e mRNAs. With respect to an eventual application in man it is important to learn that orally administered bacterial preparations may have an impact on the quality of islet inflammation. Interestingly, OM-89 was shown to ameliorate clinical measures of rheumatoid arthritis in placebo-controlled trials [ 40 ]. We conclude that immunostimulatory bacterial preparations exert strong effects on the Th1/Th2 ratio of insulitis. Whether changes in the Th1/Th2 ratio will delay or prevent diabetes development depend on the extent and duration of such effects. Dosing and timing of bacterial vaccinations will be important. Further studies are warranted to elucidate the mechanism of action and therapeutic effects of bacterial immunostimulants on autoimmune disease. The ratio of Th1/Th2 cytokine gene expression, as described here, may become a useful tool for monitoring the immunoregulatory effect of these compounds. Acknowledgements. We thank Mrs. G. Bosse for typing the manuscript, Mrs. M. Rinker for technical assistance and Dr. B. Kallmann for help with statistical analyses. We acknowledge the assistance of Mrs. J. Souligny in carrying out the animal studies in Ottawa. This work was supported by the Bundesminister f u¨r Gesundheit, the Bundesminister f u¨r Bildung, Wissenschaft und Forschung und Technologie and by the Minister f u¨r Wissenschaft und Forschung des Landes Nordrhein-Westfalen. 1. Shehadeh NN , La Rosa F , Lafferty KJ ( 1993 ) Altered cytokine activity in adjuvant inhibition of autoimmune diabetes . J Autoimmunity 6 : 291 - 300 2. Rabinovitch A ( 1994 ) Immunoregulatory and cytokine imbalances in the pathogenesis of IDDM . Diabetes 43 : 613 - 621 3. Rothe H , Faust A , Schade U et al. ( 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 4. Rabinovitch A , Sorensen O , Suarez-Pinzon WL , Power RF , Rajotte RV , Bleackley RC ( 1994 ) Analysis of cytokine mRNA expression in syngenic islet grafts of NOD mice: interleukin-2 and interferon-gamma mRNA expression correlate with graft rejection and interleukin-10 with graft survival . Diabetologia 37 : 833 - 845 5. Anderson JT , Cornelius JG , Jarpe AJ , Winter WE , Peck AB ( 1993 ) Insulin-dependent diabetes in the NOD mouse model II cell destruction in the autoimmune diabetes is a Th2 and not a Tb h1 mediated event . Autoimmunity 15 : 113 - 122 6. Campbell IL , Kay TWH , Oxbrow L , Harrison LC ( 1991 ) Essential role for interferon-gamma and interleukin-6 in autoimmune insulin-dependent diabetes in NOD/Wehi mice . J Clin Invest 87 : 739 - 742 7. Trembleau S , Penna G , Bosi E , Mortara A , Gately MK , Adorini L ( 1995 ) Interleukin-12 administration induces T helper cells and accelerates autoimmune diabetes in NOD mice . J Exp Med 181 : 817 - 821 8. Rothe H , Burkart V , Faust A , Kolb H ( 1996 ) Interleukin-12 gene expression is associated with rapid development of diabetes mellitus in non-obese diabetic mice . Diabetologia 39 : 119 - 122 9. Harada M , Kishimoto Y , Makino S ( 1990 ) Prevention of overt diabetes and insulitis in NOD mice by a single BCG vaccination . Diabetes Res Clin Prac 8 : 85 - 89 10. Sadelain MWJ , Qin HY , Lauzon J , Singh B ( 1990 ) Prevention of type I diabetes in NOD mice by adjuvent immunotherapy . Diabetes 39 : 583 - 589 11. McInerney MF , Pek SB , Thomas DW ( 1991 ) Prevention of insulitis and diabetes onset by treatment with complete Freund's adjuvant in NOD mice . Diabetes 40 : 715 - 725 12. Uleato D , Lacy PE , Kipnis DM , Kanagawa O , Unanue ER ( 1992 ) A T-cell dormant state in the autoimmune process of nonobese diabetic mice treated with complete Freund's adjuvant . Proc Natl Acad Sci 89 : 3927 - 3931 13. Katz JD , Benoist C , Mathis D ( 1995 ) T helper subsets in insulindependent diabetes . Sciences 268 : 1185 - 1188 14. Seemayer TA , Tannenbaum GS , Goldmann H , Colle E ( 1982 ) Dynamic time course studies of the spontaneously diabetic BB Wistar rat . Am J Pathol 106 : 237 - 249 15. Logothetopoulos J , Valiquette N , Madura E , Cvet D ( 1984 ) The onset and progression of pancreatic insulitis in the overt, spontaneously diabetic young adult BB rat studied by pancreatic biopsy . Diabetes 33 : 33 - 36 16. Hanenberg H , Kolb-Bachofen V , Kantwerk-Funke G , Kolb H ( 1989 ) Macrophage infiltration precedes and is a prerequisite for lymphocytic insulitis in pancreatic lesions of prediabetic BB rats . Diabetologia 32 : 126 - 134 17. Walker R , Bone AJ , Cooke A , Baird JD ( 1988 ) Distinct macrophage subpopulations in pancreas of prediabetic BB/E rats . Diabetes 37 : 1301 - 1304 18. Ku L , Kim MK , Amano K , Pak CY , Jaworski MA , Mehta JG , Yoon J-W ( 1989 ) Preferential infiltration of macrophages during early stages of insulitis in diabetes-prone BB rats . Diabetes 37 : 1053 - 1058 19. Voorbij HAM , Jeucken PHM , Kabel PJ , de Haan M , Drexhage HA ( 1989 ) Dendritic cells and scavenger macrophages in pancreatic islets of prediabetic BB rats . Diabetes 38 : 1623 - 1629 20. Bone AJ , Walker R , Varey A-M , Cooke A , Baird JD ( 1990 ) Effect of cyclosporin on pancreatic events and development of diabetes in BB/Edinburgh Rats . Diabetes 39 : 508 - 514 21. Ziegler A-G , Erhard J , Lampeter EF , Nagelkerken LM , Standl E ( 1992 ) Involvement of dendritic cells in early insulitis of BB rats . J Autoimmunity 5 : 571 - 579 22. Satoh J , Shintani S , Oya K , Tanaka S-I , Nobunaga T , Toyota T , Goto Y ( 1988 ) Treatment with streptococcal preparation (OK432) suppresses anti-islet autoimmunity and prevents diabetes in BB rats . Diabetes 37 : 1188 - 1194 23. Sadelain MWJ , Qin H-Y , Sumoski W , Parfrey N , Singh B , Rabinovitch A ( 1990 ) Prevention of diabetes in the BB rat by early immunotherapy using Freund's adjuvant . J Autoimmun 3 : 671 - 680 24. Kiesel U , Farine J-C , Kolb H ( 1994 ) Prevention of diabetes in BB rats by an E. coli extract (OM-89) . Int J Immunopathol Pharmacol 7 : 193 - 404 25. Kleemann R , Rothe H , Kolb-Bachofen V et al. ( 1993 ) Transcription and translation of inducible nitric oxide synthase in the pancreas of prediabetic BB rats . FEBS Lett 328 : 9 - 12 26. Suschek C , Fehsel K , Kr o¨ncke K-D , Sommer A , Kolb-Bachofen V ( 1994 ) Primary cultures of rat islet capillary endothelial cells . Am J Pathol 145 : 685 - 695 27. Siegling A , Lehmann M , Platzer C , Emmrich F , Volk HD ( 1994 ) A novel multispecific competitor fragment for quantitative PCR analysis of cytokine gene expression in rats . J Immunol Methods 177 : 23 - 28 28. Kolb-Bachofen V , Epstein S , Kiesel U , Kolb H ( 1988 ) Low-dose streptozotocin-induced diabetes in mice . Diabetes 37 : 21 - 27 29. Kolb-Bachofen V , Schraermeyer U , Hoppe T , Hanenberg H , Kolb H ( 1992 ) Diabetes manifestation in BB rats is preceded by panpancreatic presence of activated inflammatory macrophages . Pancreas 7 : 578 - 584 30. Huang X , Hultgren B , Dybdai N , Stewart TA ( 1994 ) Islet expression of interferon- precedes diabetes in both the BB rat and streptozotocin-treataed mice . Immunity 1 : 469 - 478 31. L o ¨hr M , Markholst H , Dyrberg T , Klo¨ ppel G , Oberholzer M , Lernmark A ( 1989 ) Insulitis and diabetes are preceded by a decrease in -cell volume in diabetes-prone BB rats . Pancreas 4 : 95 - 100 32. bGepts W , LeCompte M ( 1985 ) The pathology of type I (juvenile) diabetes . In: Volk BW , Arquilla EA (eds). The diabetic pancreas . Plenum Publishing Corporation , New York pp. 337 - 365 33. Somoza N , Vargas F , Roura-Mir C et al. ( 1994 ) Pancreas in recent onset insulin-dependent diabetes mellitus. Changes in HLA, adhesion molecules and autoantigens, restricted T cell receptor V usage, and cytokine profile . J Immunol 153 : 1360 - 1365 34. Jiang Z , Woda BA ( 1991 ) Cytokine gene expression in the islets of the diabetic Biobreeding/Worcester rat . J Immunol 146 : 2990 - 2994 35. Powrie F , Coffman RL ( 1993 ) Cytokine regulation of T-cell function: potential for therapeutic intervention . Immunol Today 14 : 270 - 274 36. Zipris D , Greiner DL , Malkani S , Whalen B , Mordes JP , Rossini AA ( 1996 ) Cytokine gene expression in islets and thyroids of bb rats . J Immunol 156 : 1315 - 1321 37. Katsikis PD , Cohen SBA , Londei M , Feldmann M ( 1995 ) Are CD4(+) T(h)1 cells pro-inflammatory or antiinflammatory? The ratio of interleukin-10 to interferon-gamma or interleukin-2 determines their function . Int Immunol 7 : 1287 - 1294 38. Kallmann B , H u¨ther M, Bertrams J , Gries FA , Lampeter EF , Kolb H ( 1995 ) A bias towards Th1 activity in the peripheral blood of recent onset type I diabetic patients . Autoimmunity 21 : 52 (Abstract) 39. Kallmann BA , Kolb H , H u¨ther M, Martin S , Hellermann M , Lampeter EF ( 1996 ) Interleukin-10 is a predominant cytokine in atopic dermatitis . Arch Dermatol , in press 40. Peters DH , Goa KL ( 1994 ) OM-8980 An initial review of its pharmacology and therapeutic potential in rheumatoid arthritis . Clin Immunother 2 : 65 - 67 41. Polla BS , Baladi S , Fuller K , Rook G ( 1995 ) Presence of hsp65 in bacterial extracts (OM-89): a possible mediator of orally-induced tolerance? Experientia 51 : 775 - 779


This is a preview of a remote PDF: https://link.springer.com/content/pdf/10.1007%2Fs001250050597.pdf

H. Kolb, U. Wörz-Pagenstert, R. Kleemann, H. Rothe, P. Rowsell, F. W. Scott. Cytokine gene expression in the BB rat pancreas: natural course and impact of bacterial vaccines, Diabetologia, 1996, 1448-1454, DOI: 10.1007/s001250050597