Interleukin-1β (IL-1) Does Not Reduce the Diabetes Incidence in Diabetes-Prone Bb Rats

Autoimmunity, 1994, Vol. 17, pp. 105-1 18 Reprints available directly from the publisher Photocopying permitted by license only

0 1994 Harwood Academic Publishers GmbH Printed in Singapore

INTERLEUKIN-1P (IL-1) DOES NOT REDUCE THE DIABETES INCIDENCE IN DIABETES-PRONE BB RATS

JESPER I. REIMERS, LISE M 0 R C H , HELLE MARKHOLST, LISE D. WOGENSEN, HENRIK U. ANDERSEN, THOMAS MANDRUP-POULSEN, AND J 0 R N NERUP

Steno Diabetes Center and Hagedorn Research Institute, DK-2820 Gentofte, Denmark.

(Received July 26, 1993; in final form November 23, 1993)

The cytokine interleukin Ip (IL-I) has been implicated as a pathogenetic factor in the initial events leading to insulin-dependent diabetes mellitus. Previous studies investigating the impact of IL-I on diabetes incidence in spontaneously diabetic rodent models have been conflicting. IL-1 induces anorexia and previous studies are hampered by the lack of pair-fed controls to the IL-1 treated animals. We report that daily injections of 4.0 pg/kg/day of recombinant human IL-1 (rhIL-1) for 13 weeks from 25-30 days of age did not alter the incidence of diabetes in the diabetes-prone (DP) BB rats (75%) when compared to pair-fed, vehicle treated controls (55%, p = O.I8), or to unhandled DP BB rats (80%, p = 0.71). However, IL-1 induced significantly higher blood glucose concentrations in the prediabetic period (p < 0.00005) and at diabetes onset (p < 0.00005) in the DP BB rats and caused episodes of blood glucose concentrations > 11 mmol/l in the prediabetic period in 11/20 DP BB rats compared to 4/27 diabetes-resistant (DR) BB rats and 4/28 Wistar Furth (WF) rats (both p < 0.004), compared to D P BB). Further, rhIL-1 induced fever in 1 1 weeks in the DP BB rats compared to 3 weeks in the DR BB and 6 weeks in the WF rats. Using high performance size exclusion chromatography specific anti-rhILl-antibodies were demonstrated in DR BB and WF, but not in DP BB rats. These antibodies neutralized the inhibitory effect of rhIL-l on insulin secretion from isolated islets of Langerhans in vitro. The reduced pyrogenic and endocrine effect of rhIL-1 in the DR BB and WF rats compared to the DP BB rats could be explained by the impaired ability of the DP BB rats to produce anti-rhIL-I-antibodies. In conclusion, administration of rhIL-1 modulated the prediabetic period, and produced higher blood glucose concentrations at diagnosis, but did not change the diabetes incidence in DP BB rats. The results are not in conflict with the hypothesis that IL-1 is a pathogenetic factor in IDDM, caused by high local concentrations of rat IL-1 in the islets during early insulitis. The results also show the necessity of pair-feeding of the control group to the rhIL-l group when interpreting data from experiments investigating rhIL-1 effects on diabetes development in animal models.

KEY WORDS: Diabetes incidence, food intake, temperature, histology, leukocytes, RT6 + D P BB rats

INTRODUCTION

At the onset of overt insulin-dependent diabetes (IDDM), the islets of Langerhans are infiltrated by mononuclear cells and beta-cells are destroyed (insulitis) (1-3). Interleukin l p (IL-l), a pro-inflammatory cytokine and a major mediator of the acute phase response, has been implicated alone or in combination with other cytokines as an effector molecule in the initial steps of the pathogenesis of IDDM (4). IL-1 markedly affects beta-cell function in vitro in a bipha- sic fashion, dependent upon the glucose concentration in the media and the time of exposure to and concen- tation of IL-1 (5-7). In vitro IL-1 is preferentially beta-cell cytotoxic (8). The effect of IL-1 on beta-cell function is potentiated by other cytokines e.g. tumor

Correspondence to: Professor J0rn Nerup, DMSc, Steno Dia- betes Center, 2 Niels Steensensvej, DK-2820 Gentofte, Den- mark, Telephone: + 45 31 68 08 00. Fax: + 45 3 1 68 23 22

necrosis factor alfa (TNFa), and interferon gamma (IFNy) (9-10). Daily injections of IL-1 for 5 days induce transient diabetes in normal rats characterized by hyperglycemia and hypoinsulinemia (1 1).

Mononuclear cells expressing mRNA for IL 1, TNFa, and IFNy are present in the islets before and at onset of diabetes in the animal models for IDDM, the diabetes prone BB rat (DP BB) and the NOD mouse (1 2,13). Long-term administration of high doses of I L 1 to DP BB rats from 30 to 150 days of age accelerates the onset of diabetes (1 4,15), but reduces the incidence of diabetes in NOD mice (1 6 ) .

It is unknown whether the effect of IL-1 on diabetes incidence in animal models of spontaneous IDDM is caused by IL-1 effects on the immune-system, the beta-cells, via effects on other cell types or a combination of the three. We therefore investigated food- intake, body weight, temperature, blood glucose concentration, and diabetes incidence in DP BB, diabetes resistant BB (DR BB), and Wistar Furth (WF) rats

105

Aut

oim

mun

ity D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y Q

UT

Que

ensl

and

Uni

vers

ity o

f T

ech

on 1

1/02

/14

For

pers

onal

use

onl

y.

106 JESPER I. REIMERS ETAL.

during daily injections of recombinant human IL- 1 p (rhIL-1) from 30 days of age for 13 weeks or until diabetes onset. At the end of the study we investigated peripheral leukocyte counts, expression of the T-lymphocyte differentiation marker RT6, plasma levels of insulin, glucagon, corticosterone, rhIL- 1 binding proteins, and pancreatic histology.

We report that rhIL-1 did not significantly alter diabetes incidence compared to pairfed, vehicle treated controls, but rhIL-1 modulated the natural history of the prediabetic phase in the DP BB rats. The DP BB rats had an increased response in temperature, blood glucose and plasma corticosterone to rhIL- 1 compared to the WF and DR BB rats. This was associated with low levels of neutralizing anti-rhIL- 1 - antibodies in the DP BB rats compared to the other rat strains investigated.

MATERIALS AND METHODS

Animals. Male diabetes-prone (DP) (n = 40) and diabetes-resistant (DR) (n = 55) BB rats (Hagedorn Research Institute, Gentofte, Denmark) and Wistar Furth rats (WF) (n = 55) (Iffa Credo, France) were investigated. The DR and DP BB rats were obtained from a colony free of SDA virus, H1 virus, Kilham virus, pneumonia virus of mice, Reo-3, rat corona virus, Sendai virus, Hantaan virus, and mycoplasma. Rats (23-26 days old) were housed under controlled conditions of light (light on from 06:OO a.m. to 06:OO p.m.), humidity (60-70%) and temperature (2 1-22°C) for 3-4 days before experiments were initiated. Ani- mals were offered a standard chow (Altromin, Chr. Petersen A / S , Ringsted, Denmark) as detailed below and millipore water ad libitum. Two or three rats were placed in each cage at 25-30 days of age and randomized to subcutaneous injections in the back of the neck of either 4.0 pg/kg body weight (BW) of rhIL-1 or an identical volume of vehicle (see below). Injections were performed daily for 13 weeks at 09:OO a.m. The BW was measured twice weekly and the average food- intake per 100 g BW in each cage in the rhIL- 1 groups was determined by weighing residual food. Rats in the control groups were offered an amount of food equal to that consumed by the rhILl treated rats and were therefore truly pair-fed with the rhIL- 1 treated rats. Body temperature was measured weekly 2 hours after the injection of rhIL-1 or vehicle by means of a rectal thermometer (Medical Precision Thermometer Dh 852, Ellab, Copenhagen, Denmark). Rats were tested for glucosuria (Tes-Tape, Liily, Indianapolis, Indiana, USA) 2 h after injections twice weekly. In glucosuric rats, blood glucose (BG) was analysed by means of a Hypocount MX-B (Boehringer Mannheim, Germany) immediately after urine glucose analysis. Rats were

considered diabetic if BG levels exceeded 14 mmol/l on two consecutive days. The day after urine glucose analysis and 5 h after injection of rhIL-1 or vehicle and withdrawal of food, blood was sampled for glucose measurement by a automatic analyser. After 13 weeks of treatment or at diabetes onset, tail-vein blood was sampled for counting and flow cytometric analysis of leukocytes. Rats were then decapitated, the blood from the neck (trunk blood) was sampled and tissues were collected for histological examination. Rats did not receive injections at the day of decapitation (2 24 h after last injection). Twenty DP BB rats (6 female, 14 male) receiving no treatment and with free access to food, and 16 DP BB (1 6 female) receiving no treatment but being pair-fed to the rhIL-1 treated DP BB rats, were also included, weighed, and monitored for diabetes twice weekly. The animal facilities and care were approved by the Danish National Health Service.

Preparation of rhIL-I: Recombinant human I L l p (Batch C16, Novo Nordisk A / S , Bagsvierd, Denmark) was expressed in E. coli. It was identical to the authentic natural IL-1 p, exhibiting the correct N-terminal amino-acid (alanine) (1 7). With minor modifications rhIL/ 1 was purified as described elsewhere ( 1 7). The protein concentration was determined using E = 0.59 (277 nm, 0.1%) (1 7). The specific biological activity of this rhIL-1 preparation was 350 WHO Uhg, being identical to the specific biological activity of natural human IL-lP as estimated by the mouse co- stimulatory thymocyte (LAF) assay (18). rhIL-1 was diluted in sterile, endotoxin-free (i.e. I 0.8 pg/ml) 0.9% NaCl containing 0.1 % strain-identical rat serum. The volume injected was dependent upon the final concentrations of the diluted rhIL-1 (n = 5, 1.9 ? 0.2 pg/ml) as determined by an enzyme-linked immunosorbent assay technique (ELISA) ( I 8). Endot- oxin was not detectable in the diluted rhILl preparation or the vehicle by the Limulus amoebocyte test ( 19).

Blood glucose and plasma hormone analysis. Ten pl blood for gluocse (BG) analysis were sampled from a tail vein and analysed by a Kobas MIRA automatic analyzer using the glucose dehydrogenase method. Trunk blood for hormone analysis was collected in tubes containing 125 pl aprotinin 20.000 KIE/ml (Bayer AG, Leverkusen, Germany) and 50 p1 of EDTA (30%, pH 8.0) and kept on ice until centrifuged for 10 min at 4.000 rpm and 4°C. Plasma was kept frozen at - 20°C until hormone analysis. The plasma insulin concentrat ion was measured by radioimmunoassay (RIA) (Novo Nordisk A/S, Bagsvierd, Denmark) using rat standards and anti-insulin serum M8 170. The detection limit was 10 pmol/l. Inter- and intraassay coefficients of variation were 13.1% and 11.4Y0,

Aut

oim

mun

ity D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y Q

UT

Que

ensl

and

Uni

vers

ity o

f T

ech

on 1

1/02

/14

For

pers

onal

use

onl

y.

EFFECTS OF IL-1 IN BB AND WISTAR RATS 107

respectively. Glucagon was extracted by ethanol precipitation, redissolved in buffer and analysed by RIA (Novo Nordisk). The detection limit of the assay was 50 ng/l and inter- and intraassay coefficients of variation were 11.7?40 and 17.0%, respectively. The detection limit of the corticosterone RIA assay (ICN Bio- medicals Inc., Costa Mesa, California, USA) was 250 ng/l and inter- and intrassay coefficients of variation were 11.5% and 7.9%, respectively.

Histology: The splenic part of the pancreas, the kidney, adrenal, gastric ventricle, proximal part of duodenum, liver, thymus, lung, and skin from the injection area were sampled immediately after decapitation of 4 randomly chosen rats from each group. The tissues were fixed in 4% formalin buffer, embedded in paraffin and sectioned in 4 pm thickness. Deparaffinized sections were stained with hematoxylin/eosin and van Gieson-Hansen. Coded sections were screened for mononuclear cell infiltration. The mononuclear cell infiltration in the islets of Langerhans was evaluated by two independent investigators in a blinded fashion including positive and negative calibrators and were scored according to the following graduation: 0) no infiltration, 1) periislet infiltration, 2) < 25% of the islet infiltrated, 3) 25-50% of the islet infiltrated, 4) 50 to 100% of the islet infiltrated. Ten randomly selected islets from each animal were scored by the investigators.

rhIL-1 binding proteins in plasma. 490pl of plasma sampled as detailed above was incubated for 16-22 h at 4°C with 10 pl of 250 ng/ml of iodine- labeled rhIL- 1 (' 251-rhIL- 1) ( 1 1 .O x 1 O6 cpm/pg). The iodination of rhlL- 1 were performed as previously described (20). The plasma was filtered through a 0.22 pm Millex-GV filter (Millipore, Molsheim, France), radioactivity in 2 x 10 pl serum was counted, and 100 pl was analysed by high performance size exclusion chromatography (HPSEC) as described previously (20). A linear MW standard curve (Fig. 6) was obtained using the following standards: human serum albumin 11 (MW: 136 kDa), transferrin (MW: 76kDa), human serum albumin I (MW: 68kDa), ovalbumin (MW: 43 kDa), myoglobin I1 (MW: 35 kDa), P-lactoglobin (MW: 35 kDa), human growth hormone (MW: 22 kDa), myoglobin I (MW: 18 kDa) and insulin (MW: 6 kDa). '251-rhIL-1 incubated in rat serum or eluent buffer eluted in fractions 40-45 and 36-44, respectively (peak in fractions 42 and 39-40, respectively) with a recovery of 97% and 1039'0, respectively. I2'I incubated in rat serum or eluent buffer eluted in fractions 50-52 (peak in fraction 51) with a recovery of 98% and 94%, respectively.

To investigate the nature of the rhIL-1 binding protein, 490 of pl plasma from rhIL-1 or vehicle

treated WF and DR BB rats (diluted 1:lOO in eluent buffer) were incubated with 10 pl of '2SI-rhIL-l (250 ng/ml) for 16 h. One hundred p1 of protein G sepharose@ (4 fast flow, Lot#OE 11887, Pharmacia), was washed four times in IMP-buffer, added to the samples and incubated for 2 h at 4°C. After incubation the samples were centrifuged for 5 min at 3000 rpm and 4"C, and the supernatant was filtered through a 0.22 l m Millex-GV filter. 100 pl of the supernatant was counted, and 100 p1 was applied on a HPSEC column, The binding specificity was investigated by a 16 h coincubation of 480 p1 of rat serum (diluted 1:lOO in eluent buffer) and 10 pl '"I-rhILl (250 ng/ ml) and 10 pl of unlabelled rhIL- 1 in 0.1, 1, 10, 100, 1 .OOO, and 10.000-fold molar excesses, or recombinant human IL- 1 a (British Bio-technology Limited, Oxford, U.K.) or recombinant human TNFa (Gen- zyme Corporation, Cambridge, MA) in 1, 10, 100, and 1 .OOO-fold molar excesses to 12'I-rhIL-l. To investigate whether the circulating rhIL- 1 binding proteins neutralized the effect of rhIL- 1 on beta-cells in vitro, 4- to 6-day-old outbred Wistar rats (Mdlegaard) were killed by decapitation. The pancreatic islets were isolated by collagenase digestion as previously described (21,22). Islets were precultured for 5-7 days in humid- ified atmospheric air at 37°C in RPMl 1640 (Flow Laboratories, Irvine, Scotland) with 11 mM glucose, 20 mM Hepes buffer, pH 7.3, 100,000 IU/l penicillin, 100 mg/l streptomycin and 10% newborn calf serum (NCS). For the 24 h experimental culture 10% NCS was replaced by 1% rat serum. The plasma sampled after 13 weeks of treatment was pooled from 5 of the rhIL-1 treated or 5 of the vehicle treated, pair fed rats from the WF strain. Five p1 of the plasma pools were incubated with rhIL- 1 (1 50 pg/ml) in 495 p1 RPMI for 16-24 h at 5°C. Thirty islets were then incubated with the 500 pl RPMI + serum for 24 h at 37°C. The supernatant was sampled in duplicate from 6 independent experiments for analysis of accumulated insulin.

Counting and f low cytometric analysis of leukocytes: For leukocyte counting 25 pl of tail vein blood was collected in tubes containing 475 pl of methylvio- let 6B. Stained cells were counted manually in a Burker-Turk counting chamber. The average of two counts of cells per sample was recorded. For lymphocyte immuno-staining blood from a tail vein was sampled in EDTA, and 50 pl were incubated for 30 min at 4°C with 50 pl of an monoclonal antibody to RT6.1 (DS4.23) (BB DP and BB DR rats) or RT6.2 (6A5) (WF rats) (23) diluted 1:lOO in PBS+ 1.0% BSA. The two monoclonal antibodies served as each others controls. A control sample was incubated in buffer alone. The samples were washed in 3 ml PBS + 1.0% BSA, followed by incubation with 50 pl of an F(ab'), fragment of a FITC-conjugated goat anti-rat IgG in a final dilution of 1:lO (23). For each sample the control

Aut

oim

mun

ity D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y Q

UT

Que

ensl

and

Uni

vers

ity o

f T

ech

on 1

1/02

/14

For

pers

onal

use

onl

y.

108 JESPER I. REIMERS ET AL.

sample incubated with second antibody alone was used to define the channel number corresponding to 95Yo of cells being negative. This channel then determined the lower limit of positivity in the sample incubated with both antibodies. The percentage positive cells per lo5 nucleated cells was measured using a fluorescence activated cell sorter (FACScan, Beckton- Dickenson, Glostrup, Denmark). The data was analysed using the LYSIS I1 software (Beckton- Dickenson).

Statistical analysis: Results are expressed as means k SD except where noted. The two-tailed Mann-Whitney rank sum test and the Wilcoxon-Pratt test were used to test differences between sets of unpaired data and paired data, respectively. The Kruskal-Wallis test was used for comparing more than two sets of unpaired data. The chi-square test were used to test differences in frequencies of IDDM between the groups. The level of significance chosen was

100

80

60

40

20

I

RESULTS

Diabetes incidence: Overt diabetes only developed in the DP BB rats with an incidence of 80% at 120 days of age in unhandled rats. rhIL-1 treatment did not statistically significantly alter the diabetes incidence at 120 days of age (75%) compared to the relevant vehicle treated, pair-fed control group (55%) (chi-square = 1.76, p = 0.18) or to the non-treated, ad libitum fed rats (80%) (chi-square = 0.14, p = 0.71) (Fig. 1). Pair-feeding with or without vehicle injections tended to reduce, albeit not significantly, the diabetes incidence when compared to the non-treated, ad libitum fed rats (chi-square=2.85, p=O.O9, and chi- square = 2.36, p = 0.12, respectively).

Body weight: To ensure that pair-feeding was effec- tive BW was measured in DP BB, DR BB, and WF rats (Fig. 2). On the first day of treatment the body weights of the rats randomized to rhIL-1 treatment were similar to those randomized to vehicle and pair-

4 t-----

Figure 1. The effect of recombinant human IL-lP on diabetes incidence in diabetes-prone BB rats. From 25-30 days of age, the rats received daily injections of rhIL-1 (4.0 pg/kg) (---) (n = 20), or vehicle (-A-) (n = 20) for 13 weeks. Vehicle treated rats were pair-fed to the rhIL1 treated. As control groups not receiving injections, one group was pair-fed to the rhIL-1 treated rats (. + . .) (n = 16), and one group fed ad libitum (--0--) (n = 20).

Aut

oim

mun

ity D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y Q

UT

Que

ensl

and

Uni

vers

ity o

f T

ech

on 1

1/02

/14

For

pers

onal

use

onl

y.


500

400

300

200

h

0 W

r' 0 .-

I x U 0 m

- Wistar Fufh Food-intake: The food-intake in the three rat strains decreased during rhIL-1 treatment (Fig. 3). The food - intake in the DR BB and the WF rats was similar in the 13th week of treatment, but the DP BB rats had a significantly lower food intake compared to the WF and the DR BB rats (89.2% and 82.7%, respectively, both p < 0.025). Since the vehicle treated rats were

p-rf-rrfP pPPPf-f-p

- -

400

300

Figure 2. Effect of human recombinant IL-I p on body weight of Wistar Furth (n = 5 5 ) , diabetes-resistant BB (DR BB) (n = 5 9 , and diabetes-prone BB (DP BB) rats (n=40 at week 0, and n = 15 at week 13) Rats received daily injections of rhIL-1 (4.0 @kg) (-C) (n = 28 (WF), 27 (DR BB), and 20 to 6 (DP BB)) or vehicle (--O--) (n = 27 (WF), 28 (DR BB), and 20 to 9 (DP BB)) starting at 25-30 days of age for 13 weeks or until onset of IDDM. Vehicle treated rats were pair-fed to the rhIL1 treated rats (see methods). Means + SD.

Blood glucose and plasma insulin, glucagon and corticosterone concentrations: Although rhIL1 did

- -

feeding in the three rat strains. After 13 weeks of treatment the rhIL-1 treated DP BB rats had identical BW to the vehicle treated, pair-fed rats. However, the rhILl treated WF and DR BB rats had slightly (7.4% and 6.9%, respectively), but significantly higher body weights when compared to the vehicle treated, pair-fed rats (p < 0.0003 and p < 0.008, respectively).

ber of DP BB (1 1/20) compared to DR BB (4/27) and WF (4/28) (both p < 0.004). None of the vehicle injected, pair-fed rats had BG levels > 7.3 mmoYl during the same observation period (p < 0.0001, < 0.04 and < 0.04, compared to the respective rhIL-1 injected rats). From 6 weeks to 1 week before onset of diabetes the average BG levels were significantly higher in the rhIL-1 treated (6.9+ 3.6 mmol/l) compared to the vehicle treated DP BB rats (4.7 f 0.5 mmol/l, p < 0.00005) (Fig. 5). However, 3 to 4 days before diabetes onset the BG levels were similar in the two groups, because the BG in the vehicle treated rats had started to increase. At onset of diabetes the rhIL 1 treated rats had significantly higher BG levels compared to the vehicle injected. The unhandled DP BB rats with or without pair-feding had significantly lower BG concentrations (20.0 f 2.1 mmol/l, and 18.7 k 2.6 mmol/l, respectively) compared to the rhIL-1 treated group (23.6 k 2.2 mmol/l,

Aut

oim

mun

ity D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y Q

UT

Que

ensl

and

Uni

vers

ity o

f T

ech

on 1

1/02

/14

For

pers

onal

use

onl

y.

110

A x a -0

2 m 0

0 0 r \ 0

0 Y a t I

-0 0 0 LL

v

-P

. -

18

16

14

12

I 0

8

6

4

JESPER I. REIMERS ET AL.

0 1 2 3 4 5 6 7 8 9 I 0 1 1 1 2 1 3 Weeks of treatment

Figure 3. Food intake (g/lOOg body weight (BW)/day) per cage during 13 weeks of treatment with human recombinant IL-lp in Wistar Furth (---) (number of cages = lo), diabetes-resistant BB (---) (number of cages = lo), and diabetes-prone BB (.....) (number of cages - 8 at week 0, and n = 5 at week 13) rats. Rats received daily injections of rhIL-1 (4.0 pgikg) from 25-30 days of age. There were 1 to 3 rats per cage. Means * SD.

p < 0.004 and p c 0.001, respectively) at the day of onset of diabetes.

The plasma insulin levels were not altered by rhIL- 1 treatment compared to the vehicle injected controls (Table 1). As expected, the diabetic DP BB rats had significantly lower levels of insulin compared to the non-diabetic DP BB (p < 0.001), and the non-diabetic DP BB had levels of insulin similar to WF and DR BB rats. rhIL- 1 treatment increased the plasma concentrations of glucagon in the WF rats, but not in the other strains. The rhIL1 and vehicle treated diabetic DP BB rats had significantly higher levels of glucagon compared to the non-diabetic DP BB (p < 0.03). In the rhIL-1 treated groups, the non-diabetic DP BB had levels of glucagon similar to the two other strains. In the vehicle treated groups the non-diabetic DP BB had levels of glucagon similar to the DR BB, and glucagon was significantly higher in both groups when compared to the WF rats (both p < 0.012). The level of corticosterone was significantly decreased in rhIL- 1 treated diabetic DP BB rats compared to the diabetic vehicle treated (p < 0.04). The onset of diabetes did not alter the level of corticosterone in the DP BB rats. The rhIL-1 treated DR BB rats had a significantly higher level of corticosterone compared to the rhIL- 1 treated WF, diabetic and non-diabetic DP BB rats (all

p < 0.01). However, no differences in levels of corticosterone were observed between the vehicle treated groups using the Kruskal-Wallis test.

Histology: As expected, the mononuclear cell infiltration in the islets of Langerhans was more severe in the diabetic (mean score = 2.4 and 2.6) compared to the non-diabetic (mean score = 0.9 and 1.0) vehicle treated, pair-fed controls and rhIL-1 treated DP BB rats, respectively. In the DP BB rats, rhILl treatment did not induce any histological changes in the exocrine pancreas, the adrenal glands, kidney, gastric ventricle, duodenum, liver, lung, skin, or thymus. There were no histological changes observed in any of the examined tissues comparing rhIL1 treated WF and DR BB rats to the respective vehicle treated, pair-fed controls.

rhIL-1 binding proteins in plasma. The fact that the pyrogenic effect of rhIL-1 disappeared after a few weeks of treatment in the WF and the DR BB rats prompted us to investigate if the rats had generated neutralizing anti-rhIL- 1 -antibodies. HPSEC of plasma from vehicle treated rats demonstrated radioactivity eluting in three major peaks in fractions 40-45 corresponding to intact '2SI-rhIL-l (MW = 17.5 kDa), in fractions 50-52 corresponding to '251-rhIL-l degrada-

Aut

oim

mun

ity D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y Q

UT

Que

ensl

and

Uni

vers

ity o

f T

ech

on 1

1/02

/14

For

pers

onal

use

onl

y.


40.01 T Wistar Futh

DR BB

40-01

DP BB

40.01 T

* * * * * * * * * * * . " . . . . ' " ' ' '

0 1 2 3 4 5 6 7 0 910111213

Weeks of treatment Fig. 4. Rectal temperatures during 1 3 weeks of treatment with recombinant human IL lp (4.0 kg/kg) (-O-) or vehicle and pair-feeding (--0--) from 25-30 days of age in Wistar Furth (n - 27/28), diabetes-resistant BB (DR BB) (n - 28/27), and diabetes-prone BB (DP BB) rats (n - 20/20). Means * SD. *p < 0.05.

tion products (MW < 1 kDa) or free '"I, and in fractions 24-27 corresponding to protein-bound tracer (MW = 200 kDa) (Fig 6). The rhIL1 treated WF and DR BB rats had a 6-9-fold higher amount of eluted tracer in fractions 24-27, and a 139-181-fold lower amount in fractions 40-45 compared to the respective vehicle injected strain controls (Table 2). However, tracer incubated with plasma from rh IL l and vehicle treated DP BB rats eluted similarly, indicating that rhILl treatment induced formation of rhIL-1-binding proteins in WF and DR BB rats, but only weakly so in

DP BB rats. Precipitation with protein G reduced the counts eluted in fractions 24-27 from the rhIL-1 treated rats to the level of the vehicle treated (Table 3), indicating that the rhIL-1-binding protein was IgG. Unlabelled rhIL-1 in 100- to 10.000-fold molar excesses displaced bound '251-rhIL-1 from IgG (Fig. 7). However, neither TNFa nor IL-la in a 1000-fold molar excess displaced binding of 12%rhIL-1 to the IgG.

The inhibitory effect of 150pg/ml of rhIL-1 on accumulated insulin release from isolated rat islets of Langerhans was blocked by adding 1Yo plasma from rhIL-1 treated WF rats (Table 4) indicating that the anti-rhIL- 1-antibody was neutralizing.

Counting and j low cytometric analysis of leukocytes: To investigate if rhIL-1 affected peripheral blood leukocyte counts and the percentage of RT6' T-cells we counted these cells (Table 5 ) . rhIL-1 treatment significantly increased the leukocyte counts in the diabetic DP BB and the WF rats, but not in the DR BB rats. The number of rats in the rhIL-1 treated, non-diabetic DP BB group were to low for statistical analysis. The leukocyte counts in the vehicle treated, non-diabetic DP BB rats were significantly higher compared to the vehicle treated, diabetic DP BB rats (p < 0.03), but similar to the vehicle treated WF and DR BB rats. However, the vehicle treated, diabetic DP BB rats were similar to the vehicle treated DR BB rats, but significantly lower compared to the vehicle treated WF rats (p -= 0.004). The percentage of RT6 + T-cells in blood was not altered by repetitive rh1L-1 injections in any of the strains compared to the vehicle injected strain controls. The percentage of RT6 + T-cells in the blood from vehicle treated DP BB rats with or without IDDM were similar to the vehicle treated DR BB rats. However, the percentage of RT6 + T-cells in the blood from vehicle treated WF rats was significantly higher compared to the vehicle treated DP BB with IDDM (p c 0.03) and the DR BB rats.

DISCUSSION

We have shown that daily injections of rhIL-1 modu- late the natural course of the prediabetic phase in DP BB rats. The rhIL-1 treated DP BB rats had significantly higher BG concentrations in the prediabetic phase and at onset of IDDM compared to the relevant vehicle treated, pair-fed control rats. rhIL 1 induced hyperglycemic episodes in the WF, DR BB and the DP BB rats, but with higher number of hyperglycemic episodes in the DP BB compared to the DR BB and the WF rats. The pyrogenic effect of rhIL-1 declined over time in the DR BB and the WF but not in the DP BB rats. These differences in sensitivity to rhIL-1 were related to the finding that rhIL-1 treated DR BB and

Aut

oim

mun

ity D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y Q

UT

Que

ensl

and

Uni

vers

ity o

f T

ech

on 1

1/02

/14

For

pers

onal

use

onl

y.


28

24

20

16

12

8

4 * * * * * * * * * * * 6 5 4 3 2 1 0

Time before diabetes onset (weeks)

0 I I 1 I I I I I I I I

Figure 5. The blood glucose concentration in diabetes-prone BB rats in the last 6 weeks before diabetes onset. The rats were treated with recombinant human IL-Ip (4.0 lg/kg) (-0-) (n = 15), or vehicle (--0--) (n = 11) from 25-30 days of age until onset of diabetes.

The blood glucose concentration was measured at onset of diabetes in untreated rats (V) (n = 16) or untreated, pair-fed rats (v) (n - 9) Means k SD. *p < 0.05.

Table 1. Plasma insulin, glucagon and corticosterone concentrations in Wistar Furth, DR BB and DP BB rats after daily injections of rhIL-1 (4.0 lg/kg) or vehicle for 13 weeks.

rhIL-1 Insulin Glucagon Corticosterone (4 (pmol/l) (ng/ml) (Pdml) Rat strain

- 252 i 88 98 k 32 1 2 1 i 6 9

(27) Wistar Furth

+ 307 f 115 129 ~t 44* 1 2 0 f 7 1

277 f 103 135+38 2 0 8 f 144

(28)

(28) +

DR BB +

(27) 310k 125 144k40 217+ 125

- 93k65 317k 182 204+ 122

( 1 1)

(15)

Diabetic DP BB i 85 i 34 220 i 89 107+68 A

- 228 k 55 152k60 135k74

(9)

( 5 )

Non-diabetic DP BB + 231 k 8 7 1 4 2 t 2 9 65A 17

All values are means t SD. A p < 0.04 and *p = 0.005 comparing rhIL-l treated wilh vehicle treated rats.

Aut

oim

mun

ity D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y Q

UT

Que

ensl

and

Uni

vers

ity o

f T

ech

on 1

1/02

/14

For

pers

onal

use

onl

y.

1000

800

600

400

200

EFFECTS OF IL- 1 IN BB AND WISTAR RATS 113

MW (kDa)

0 ’ I ,

20 25 30 35 40 45 50 55

100

10

1 60

fraction # Figure 6. Size exclusion chromatography standard curve (. . . X . . .) and one representative eluation curve of plasma (-0-) sampled from a Wistar Furth rat after daily injections of endotoxin free saline for 13 weeks to which ‘251-rhIL-1 in a final concentration of 5 ng/ml was added. One hundred p1 were applied on a high performance size exclusion chromatography column (see Materials and Methods for details) and the elutriated radioactivity was counted in a gamma counter.

WF rats but not DP BB rats generated neutralizing anti-rhIL- 1 -antibodies.

We could not confirm previous studies showing that rhIL 1 significantly changed diabetes incidence in the DP BB rats, i.e. that low doses decrease the incidence and high doses accelerate the onset of IDDM (1 4,15). There may be several explanations for this. First, the preparations of I L 1 are probably different, i.e. we use authentic rhIL-lp with a specific activity of 350 U/ng. The specific activity of the rhIL-1 used by Wilson et al. (1 4) and Vertrees et al. (1 5 ) were not reported. Second, the strains of DP BB rats were different. This has been shown to be of importance with regard to the impact of IL-2 on IDDM in DP BB rats (24). Third, and probably most importantly, we included controls being pair-fed to the rhIL- 1 treated rats. Previous studies investigating the impact of rhIL-1 and other immunos- timulating substances on the incidence of IDDM in both the DP BB rat and the NOD mouse, did not include pair-fed controls (14-1 6,24-27). Our data in-

dicated that pair-feeding per se reduced the incidence of IDDM compared to rats offered food ad libitum, underlining the necessity of pair-feeding the control group to the treated group, when performing experiments investigating the effect of substances, that affect food intake, on diabetes incidence in animal models.

The body weight in the WF and DR BB rats was higher after injection of rhIL-1 compared to vehicle after 13 weeks of treatment. Since the vehicle treated rats were thoroughly pair-fed to the rh IL l treated rats, the higher BW in the rhIL-1 treated rats may be explained by reduced locomotor activity in these rats (28). The DP BB rats had a lower food intake compared to the two other strains at the end of the study. Whether this is another example of differences in sensitivity of the strains to rhIL-1 cannot be answered by this study.

rhIL-1 injections induced episodes of BG > 11 mmol/l in all the rat strains 5 h after the injection and significantly increased BG concentrations in the

Aut

oim

mun

ity D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y Q

UT

Que

ensl

and

Uni

vers

ity o

f T

ech

on 1

1/02

/14

For

pers

onal

use

onl

y.

114 JESPER I. REIMERS ET AL.

Table 2. High performance size exclusion chromatography analysis of rat plasma after 13 weeks of rhIL-1 (4.0 pgfkg) or vehicle treatment.

cpm eluted (% of applied) cpm applied on the fractions fractions fractions

Incubated with rhIL-1 column 24-27 40-45 50-52

Rat strain

- 2767 f 678 11.0+ 1.0 54.2 f 4.1 13.0f2.6

+ 2523 f 497 70.6+ 6.5* 0.3 f 0.4* 11.7k0.9

- 2444 f 545 9.0+ 2.1 55.7f4.1 13.0k1.5 DR BB

+ 2337 f 480 77.4f 1.8* 0.4 f 0.2* 12.0f0.9

Diabetic - 2366 f 612 11.5k1.4 56.3f4.8 13k1.8

Wistar Furth

DP BB + 2161 f 5 2 6 13.6f 1.4 57.8 f 5.1 1 4 f 1.3

Non-diabetic - 2223f410 10.5f 1.4 53.9 f 3.9 12.4f0.7

DP BB + 2250 f 449 17.8 k 11.9 45.2 f 17.7 12.7+ 1.7 490 pI of plasma were incubated for 16 h with 10 pl 'Z'l-rhILI (250 ng/ml, 11.0 x lo6 cpm/pg). The plasma was then filtered (22 pm) and 100 pl were applied on the

column. Fractions 21 to 65 were counted. Fractions 24-27 correspond to protein bound tracer (MW - 200 kDa), fractions 40-45 to intact tracer (MW - 17.5 kDa), and fractions 50-52 to tracer degradation products (MW < I kDa) or free I2'I. n = 5 in all groups. Values are means+SD.

p < 0.008 comparing rhIL-1 treated to vehicle treated rats.

DP BB rats before and at onset of IDDM. It is inter- esting to note that not only the DP BB rats but also the non-diabetes prone rat strains presented episodes of diabetic blood glucose concentrations. This is consistent with our recent data showing that rhIL-1 induce hyperglycemia in normal rats due to inhibition of beta-cell function and induction of peripheral insulin resistance (1 1). These episodes were seen in a significantly higher number of DP BB rats compared to rats from the other strains probably related to the absence of neutralizing anti-rhIL- 1-antibodies.

Thirteen weeks of rhIL-1 treatment did not change plasma insulin concentrations in the rats. However, rhIL- 1 treatment induced an unanticipated increase in glucagon in the WF rats, since these rats had developed neutralizing anti-rhIG 1 -antibodies. The decrease in corticosterone 24 h after last injection in the diabetic DP-BB rats compared to the vehicle treated animals is in accordance with previous observations

DP BB rats has been described to be leukopenic at (1 1).

young age (29,30). However, at 100 days of age the leukocyte counts are increased to DR BB rats levels (30). This is consistent with our findings, showing that vehicle treated DP BB rats were not leukopenic at diabetes onset or at 120 days of age. Injections of IGI has been described to increase leukocyte counts (14), as was the case in both the DP BB rats at onset of diabetes and in the WF rats. However, similar to glucagon it was not anticipated that the leucocyte counts was increased in the WF rats, since these rats developed neutralizing anti-rhIG 1-antibodies. It might be explained by a higher non-IL-1 mediated stress responce in the WF rats.

The DP BB/Worcester rats express low levels of the differentiation alloantigen RT6 on T-cells (3 I) , and the low levels of RT6' T-cells is of importance for the development of diabetes in the DP BBIWorcester rats (32). Our analysis of RT6 + T-cells demonstrated that the DP BB rats of our colony have the same percentage of circulating RT6 + T-cells as the DR BB and the WF rats. This is consistent with the finding that up to 60%

Table 3. Identification of rat-anti-rhIL1-antibody in plasma by protein G precipitation.

cpm eluted in fractions # (% of cpm before precipitation) Protein G cpm applied on

Treatment precipitation the column 24-27 40-45 50-52

128 (9.0) 769 (54.2) 223 (1 5.7)

after 1.325 116 (8.2) 718 (50.6) 200(14.1)

before 1.333 1.058 (79.4) 21 (1.6) 202 ( 1 5.2)

after 387 133 (10.0) 4 (0.3) 1 5 5 ( 1 1.6) Plasma was sampled from Wistar Furth rats aRer 13 weeks of rhIL-1 or vehicle treatment. 5 kl plasma was incubated with 485 Nu2 HPOd and 10 PI '%rhILlp (250 nglml)

for 16 h. The samples were precipitated with protein G repharose'. One hundred p1 of the supernatant were counted and applied on a high performance size exclusion chromatography column. Fractions 2 I to 65 were counted. Fractions 24-27 correspond to protein bound tracer (MW2200 m a ) , fractions 40-45 correspond to intact tracer (MW - I7.S ma), and fractions SO-52 correspond to tracer degradation products (MW < 1 kDa) or free '211. The data shown are from I representative of 2 experiments.

before 1.419 Vehicle

rhIL 1

Aut

oim

mun

ity D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y Q

UT

Que

ensl

and

Uni

vers

ity o

f T

ech

on 1

1/02

/14

For

pers

onal

use

onl

y.

EFFECTS OF IL- 1 IN BB AND WISTAR RATS

60

50

40

30

20

10

1 I 5

- *- - - - - - --•

- . 4’ , .‘ . .

- p’ I

I I

I I - I

I I

I I

I I - I

I I I

I I

I - I I

I I

I 0

Figure 7. Binding specificity of rat-anti-rhIL-lP-antibodies. Plasma was sampled from one Wistar Furth rat after 13 weeks of treatment with recombinant human IL-1P (4.0 pg/kg). Plasma, in a final dilution of 1: 100 in Na2HP0,, pH 6.5, was coincubated for 16 h with 1251-rhIL-lp (-0-) and “cold” rhIL-lP in 0.1, 1, 10, 100, 1.000, and 10.000 molar excess, or recombinant human IL-la (A) or recombinant human TNFa (V) in 1, 10, 100, and 1.000 fold molar excess to 1251-rhIL1P. Data are presented as the cpm eluted in fractions 40-45 (corresponds to intact 1251-rhIL-lP) in percent of the applied cpm.

of the DP BB/Ottawa rats have a substantial number of RT6+ peripheral blood lymphocytes (33), but its precence appears to confer protection against diabetes. Kilham’s rat virus has been shown to induce diabetes in RT6 + DR BB rats (34). Since our BB rat colony are free of a long list of viruses including Kilham’s rat virus, we exlude the possibility that our DP BB are DR rats with viral induced diabetes. Since we are using the same anti-RT6-antibody for detection of RT6 + T-cells as the Worcester group (23,31,32,35), the differences in the levels of RT6 + T-cells in the DP BB rats may be related to the strain of DP BB rats or less likely the different origin of cells for the analysis (lymph nodes

and spleen cells vs peripheral blood). The higher sensitivity of the DP BB rats to the

pyrogenic, hyperglycemic, anorectic and corticosterone-lowering effects of IL-1 led us to investigate if these findings could be explained by differences in production of neutralizing anti-rhIL- 1-antibodies, as reported by others (15). The antibodies diluted 100 fold neutralized the effect of rhIL-1 (1 50 pg/ml) on beta-cells in vitro, suggesting that the antibodies in undiluted plasma can neutralize rhIL-1 in a concentration of at least 15 ng/ml. Kinetic studies has shown that the plasma concentrations of rhIL-1 after S.C. injection of 4.0 pg/kg do not exceed 1 ng/ml (20).

Table 4. The neutralizing effect of anti-rhIL-1-antibodies in rat plasma.

Culture medium with

0.5% human serum + 63*13* 1% plasma from 5 rhILl treated WF rats + 106k37 1% plasma from 5 vehicle treated WF rats + 66*13*

rhIL-1 Accumulated insulin (in percent of control) ( 1 50 pg/ml)

Plasma was collected aAer 13 weeks o f daily injections of rhILl (4.0 p 1 4 ) (n - 5) or vehicle (n - 5) starting from 25-30 days o f age. Isolated rat islets of Langerhans were incubated for 24 h in 5OOpl RPMI medium with 1% rat plasma or control medium (0.5% human serum)ihuman rhIL-1 (150 pglml). Accumulated insulin release to the medium was measured. Data are means f SEM of duplicate determinations from six individual experiments.

* p c 0.05 compared to control.

Aut

oim

mun

ity D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y Q

UT

Que

ensl

and

Uni

vers

ity o

f T

ech

on 1

1/02

/14

For

pers

onal

use

onl

y.


Table 5. Leukocyte counts and percentage of RT6 + T-cells in peripheral blood from Wistar Furth, DP BB and DR BB rats after daily injections of rhIL-1 (4.0 pg/kg) or vehicle for 13 weeks.

Rat strain rhlL-1 Leukocytes ( 1 0 9 ~ )

RT6' T-cells (%)

14.2 f 2.2 8O.Ok 3.7 (10) ( 8 )

20.6 k 7.2A 75.1 f 8.5 (10) ( 8 )

11.4k 3.0 69.5 k 9.4

- Wistar Furth

+

10.3k5.1 68.9k 10.6 (1 1) (7)

17.9 * 3.8' 63.72 7.5 (14) (10)

15.5k5.5 70.1 k 12.2 ( 8 ) (7)

(3) (3)

- Diabetic DP BB

+

- Non-diabetic DP BB

21.9k11.0 65.1 k 13.8 + All values are means+SD. The number of rats in each group is indicated in brackets. A p < 0.05 and *p - 0.0007 comparing rhIL-1 treated with vehicle treated rats.

Hence, the neutralizing capacity of the antibody in the WF and the DR BB rats exceed the expected plasma levels after the dose of rhIL-1 injected even if the exogeneous rhIL 1 induce IL- 1 -production in the recipient animal. The inability of the DP BB rats to produce neutralizing anti-rhIC 1 -antibodies, is not likely to be due to lymphopenia per se as previously proposed (1 5). First, the DP BB rats used in our study were not lymphopenic at 120 days of age, second, DP BB rats are capable of producing antibodies against autoantigens (36-39) and viruses (40), and third, they have serum IgG levels comparable to DR BB rats (41). Naturally occurring autoantibodies against I L 1 a and IL-lP have been described in humans (42). A T-cell-dependent defect in the capability of B-lymphocytes to produce I L 1 autoantibodies in the DP BB rats could be one explanation. Whether this defect plays a role in the genetic susceptibility of this strain to develop IDDM needs further investigation.

The fact that we did not find rhIL1 induced changes in the diabetes incidence or in the severity of the insulitis process in the DP BB rats when compared to the pair-fed, vehicle treated controls does not exclude that cytokines, expressed in the infiltrating cells in the early insulitis process, may be important as effector molecules in the process leading to beta cell destruction and eventually IDDM (1 2,13). Bolus injections of

ACKNOWLEDGEMENTS

The skilful technical assistance by Ms. A.M. Flarup, B. Andreasen, and B. Born at Hagedorn Research Insti- tute (Gentofte, Denmark) and by Mr T. Momberg- Jerrgensen and his staff at the Biological Department of Novo Nordisk A / S (Gentofte, Denmark), and by Mrs J. Nowak and her staff at DPD Dev. Pharmacology Growth of Novo Nordisk A / S is greatly appreciated. The authors are grateful to prof. A.A. Rossini (Univer- sity of Massachusetts Medical School, Worcester, MA 01655, USA) for the monoclonal antibodies DS4.23 and 6A5 and for the F(ab')2 fragment of a FITC- conjugated goat anti-rat IgG and for the constructive comments to the manuscript. J.I. Reimers is the recipient of a fellowship award from the Foundation of 17- 12- 198 1, Copenhagen, Denmark, H. Markholst is recipient of a Career Development Award, L.D. Wo- gensen af a Research Fellowship Award and J. Nerup of a Research Grant (#192 1 125) from the Juvenile Diabetes Foundation International. The study was supported by the Danish Diabetes Association, Bern- hardt Kleins Foundation, Garde Foundation and Novo Nordisk A / S , Gentofte, Denmark.

REFERENCES IL-1 are unable to mimic high local concentrations of IL-1 in the inflamed islet. To investigate if IL-1 can induce IDDM in non-diabetes Prone rat strains, rat-IL-1 should probably be used.

1. Gepts. W. Pathologic anatomy of the pancreas in juvenile diabetes mellitus. Diabetes 1965; 14: 61 9-33

2. Voorbij HAM, Jeucken PHM, Kabel PJ, de Haan M, Drex- hage HA. Dendritic cells and scavenger macrophages in

Aut

oim

mun

ity D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y Q

UT

Que

ensl

and

Uni

vers

ity o

f T

ech

on 1

1/02

/14

For

pers

onal

use

onl

y.


pancreatic islets of prediabetic BB rats. Diabetes 1989; 38:

3. Signora A, Pozzilli P, Gale EAM, Andreani D, Beverley PCL. The natural history of lymphocyte subsets infiltrating the pancreas of NOD mice. Diabetologia 1989; 32: 282-9

4. Mandrup-Poulsen T, Helqvist S, Melvig J, Wogensen LD, Nerup J. Cytokines as immune effector molecules in autoimmune endocrine diseases with special reference to insulin-dependent diabetes mellitus. Autoimmunity 1989;

5. Mandrup-Poulsen T, Bendtzen K, Nerup J, Dinarello CA, Svenson M, Nielsen JH. Affinity purified human Interleukin 1 is cytotoxic to isolated islets of Langerhans. Diabetologia

6. Sandler S, Anderson A, Hellerstrom C. Inhibitory effects of interleukin 1 on insulin secretion, insulin biosynthesis, and oxidative metabolism of isolated rat pancreatic islets. Endo- crinology 1987; 121: 1424-31

7. Spinas GA, Palmer JP, Mandrup-Poulsen T, Andersen H, Nielsen JH, Nerup J. The bimodal effect of interleukin 1 on rat pancreatic beta-cells - stimulation followed by inhibition - depends upon dose, duration of exposure, and ambient glucose concentration. Acta Endocinol (Copenh)

8. Mandrup-Poulsen T, Egeberg J, Nerup J, Bendtzen K, Nielsen JH, Dinarello CA. Ultra-structural studies of time- course and cellular specificity of Interleukin- 1 mediated islet cytotoxicity. Acta Path Microbiol Immunol Scand (C)

9. Mandrup-Poulsen T, Bendtzen K, Dinarello CA, Nerup J. Human tumor necrosis factor potentiates human interleukin I-mediated rat pancreatic P-cell cytotoxicity. J Immunol

10. Puke1 C, Baquerizo H, Rabinovitch A. Destruction of rat islet cell monolayers by cytokines. Synergistic interactions of Interferony, Tumor Necrosis Factor, Lymphotoxin, and Interleukin I . Diabetes 1988; 37: 133-6

11. Wogensen L, Reimers J, Nerup J, Kolb-Bachofen V, Kroncke K-D, Almdal T, Mandrup-Poulsen T. Repetitive in vivo treatment with human recombinant interleukin 1 p modify Beta-cell function in normal rats. Diabetologia

12. Jiang Z , Woda BA. Cytokine gene expression in the islets of the diabetic BiobreedindWorcester rat. J Immunol 199 I ;

13. Held W, MacDonald HR, Weissman IL, Hess MW, Mueller C. Genes encoding tumor necrosis factor a and granzyme A are expressed during development of autoimmune diabetes. Proc Natl Acad Sci USA 1990; 87: 2239-43

14. Wilson CA, Jacobs C, Baker P, Baskin DG, Dower S, Lernmark A, Toivola B, Vertrees S, Wilson D. IL-Ip mod- ulation of spontaneous autoimmune diabetes and thyroidi- tis in the BB rat. .I Immunol 1990; 144: 3784-8

15. Vertrees S, Wilson CA, Ubungen R, Wilson D, Baskin DG, Toivola B, Jacobs C, Boiani N, Baker P, Lernmark A. Interleukin-1 p regulation of islet and thyroid autoimmunity in the BB rat. JAutoimmun 1991; 4 717-32

16. Jacob CO, Aiso S, Michie SA, McDevitt HO, Acha-Orbea H. Prevention of diabetes in non-obese diabetic mice by tumor necrosis factor (TNF): Similarities between TNF-a and interleukin 1. Proc Natl Acad Sci USA 1990; 87:

17. Dalbege H, Bayne S, Christensen T, Hejnres K. Cloning and expression of an interleukin-I P precursor and it conversion to interleukin-lp. FEBS Lett 1989; 246: 89-93

18. Melvig J, Hansen BS, Worsaae H, Hejnies KR, Helle M, Dalbege H, et al. Comparison of biological and immunolog- ical activities of human monocyte-derived interleukin 1 p

1623-9

4: 191-218

1986: 29: 63-7

1988; 119: 307-1 1

1987; 95: 55-63

1987; 139: 4077-82

1992; 35: 331-9

146: 2990-4

968-72

and human recombinant interleukin Ip. Scand J Immunol

19. Biek L. New, sensitive rocket immunoelectrophoresis assay for measurement of the reaction between endotoxin and Limulus amoebocyte lysate. J Clin Microlbiol 1983; 17:

20. Reimers J, Wogensen LD, Welinder B, Hejnres KR, Poulsen SS, Nilsson P, Nerup J. The pharmacokinetics, distribution, and degradation of human recombinant interleukin Ip in normal rats. Scand J Immunol 1991; 34: 597-610

21. Brunstedt J, Nielsen JH, Lernmark A, Hagedorn Study Group. Isolation of islets from mice and rats. In: Lamer J, Pohl SL, ed. Methods in diabetes research laboratory methods. Vol I . New York: C. Wiley, 1984; 254-8

22. Spinas GA, Hansen BS, Linde S, Kastern W, Melvig J, Mandrup-Poulsen T, Dinarello CA, Nielsen JH, Nerup J. Interleukin-1 dose-dependently affects the biosynthesis of (pro)insulin in isolated rat islets of Langerhans. Diabetolo- gia 1987; 30: 474-80

23. Greiner DL, Handler ES, Nakano K, Mordes JP, Rossini AA. Absence of the RT-6 Tcell subset in diabetes-prone BB/W rats. J Immunol 1986; 136: 148-51

24. Zielasek J, Burkart V, Naylor P, Goldstein A, Kiesel U, Kolb H. Interleukin-2-dependent control of disease development in spontaneously diabetic BB rats. Immunology

25. Kida K, lnoue T, Kaino Y, Goto Y, Ikeuchi M, Ito T, Matsuda H, Elliot RB. An immunopotentiator of P-1,6; 1,3 D-glucan prevents diabetes and insulitis in BB rats. Diab Res Clin Pruct 1992; 17: 75-9

26. Formby B, Jacobs C, Dubuc P, Shao T. Exogenous administration of IL-la inhibits active and adoptive transfer autoimmune diabetes in NOD mice. Autoimmunity 1992;

27. Seino H, Takahashi K, Zhu XP, Sagara M, Masuda T, Nobunaga T, Funahashi I, Kajikawa T, Toyota T. Preven- tion of autoimmune diabetes with lymphotoxin in NOD mice. Diabetes 1993; 42: 398-404

28. Otterness IG, Seymour PA, Golden HW, Reynolds JA, Daumy GO. The effects of continuous administration of murine interleukin-la in the rat. Physiol Behav 1988; 43:

29. Like AA, Guberski DL, Butler L. Diabetic BioBreedind Worcester (BBIWor) rats need not to be lymphopenic. J Immunol 1986; 136: 3254-8

30. Eastman S, Markholst H, Wilson D, Lernmark A. Leukocy- tosis at the onset of diabetes in crosses of inbred BB rats. Diabetes Res Clin Pract 199 1 ; 12: 1 1 3-23

3 1. Crisa L, Sarkar P, Waite DJ, Haag F, Koch-Nolte F, Rajan TV, Mordes JP, Handler ES, Thiele H-G, Rossini AA, Greiner DL. An RT6a gene is transscribed and translated in lymphopenic diabetes-prone BB rats. Diabetes 1993; 42:

32. Burstein D, Mordes JP, Greiner DL, Stein D, Nakamura N, Handler ES, Rossini AA. Prevention of diabetes in BBlWor rats by single transfusion of spleen cells. Diabetes 1989; 38:

33. Marliss EB, Metroz-Dayer M-D. Montambault M, Faucher M, Grose M. Effects of single and repeated blood withdraw- als on circulating mononuclear cells in BB rats. Diabetes

34. Guberski DL. Thomson VA. Shek WR, Like AA, Handler ES, Rossini AA, Wallace JE, Welsh RM. Induction of type 1 diabetes by Kilham’s rat virus in diabetes-resistant BB/Wor rats. Science 1991; 254: 1010-3

35. Angelillo M, Greiner DL, Mordes JP, Handler ES, Naka- mura N, Mckeever U, Rossini A. Absence of RT6 + T-cells in diabetes-prone BiobreedindWorcester rats due to genetic

1990; 31: 225-35

1013-20

1990 69: 209-14

12: 21-7

797-804

688-95

24-30

1990; 39: 1099-105

Aut

oim

mun

ity D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y Q

UT

Que

ensl

and

Uni

vers

ity o

f T

ech

on 1

1/02

/14

For

pers

onal

use

onl

y.

118 JESPER 1. REIMERS ETAL.

and cell developmental defects. J. Irnrnunol 1988; 141:

36. Baekkeskov S, Dyrberg T, Lernmark A. Autoantibodies to a 64-kilodalton islet cell protein precede the onset of spontaneous diabetes in the BB rat. Sicence 1984; 224: 1348-50.

37. Dyrberg T, Nakhooda AF, Baekkeskov S, Lernmark A, Poussier P, Marliss EB. Islet cell surfase antibodies and lymphocyte antibodies in the spontaneously diabetic BB Wistar rat. Diabetes 1982; 31: 278-8 1

38. Like AA, Appel MC, Rossini AA. Autoantibodies in the BBlW rat. Diabetes 1982; 31: 816-20

39. Elder M, MacLaren N, Riley W, McConnell T. Gastric parietal cell and other autoantibodies in the BB rat. Diabe- tes 1982; 31: 3 13-8

4146-51 40. Oldstone MBA, Tishon A, Schwimmbeck PL, Shyp S,

Lewicki H, Dyrberg T. Cytotoxic T lymphocytes do not control lymphocytic choriomeningitis virus infection of BB diabetes-prone rats. J. Gen Virol 1990; 71: 785-91

41. Contreas G, Dyrberg T, Madsen OD, Markholst H, Lern- mark A. Increased levels of cirkulating immune complexes are not associated with diabetes in BB rats. Diabetes Res

42. Galley P, Mach J-P, Carrel S. Characterization and detection of naturally occurring antibodies against IL-la and I L 1 p in normal human plasma. Eur Cytokine Netw 199 1; 2:

1989; 10: 109-13

329-38

Aut

oim

mun

ity D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y Q

UT

Que

ensl

and

Uni

vers

ity o

f T

ech

on 1

1/02

/14

For

pers

onal

use

onl

y.

Interleukin-1β (IL-1) Does Not Reduce the Diabetes Incidence in Diabetes-Prone Bb Rats

Documents

Transcript of Interleukin-1β (IL-1) Does Not Reduce the Diabetes Incidence in Diabetes-Prone Bb Rats