DIABETES RESEARCH CLINICALAND EXPERIMENTAL DIABETES AND METABOLISM An International journal ISSN 0265-5985

Editor

W. J . Irvine (Edinburgh, U K )

Editorial Board

U . Di Mario (Rome, Italy) A . MacCuish (Glasgow, U K )

J . Hors (Paris, France) N . MacLaren (Gainesville, USA)

L. Jarett (Philadelphia, USA) H . Rifkin (New York, USA)

Y . Kanazawa (Tokyo, Japan) J . Rotter (Torrance, USA)

W. Knowler (Phoenix, USA) J . Turtle (Sydney, Australia)

A. Kurtz (London, U K ) C. Wollheim (Geneva, Switzerland) o

A . Lernmark (Gentofte, Denmark) P . Zimmet (Caulfield South, Australia)

V O L U M E 10 Pages 1-200

TEVIOT-KIMPTON PUBLICATIONS

EDINBURGH and LONDON

CONTENTS

Volume 10, Number 1—January 1989 O R I G I N A L P A P E R S

The Rela t ionship of Insulin Antibodies , Platelet-fixing Immune Complexes and Platelet-associated to I g G to i n vitro Platelet Aggregation and Thromboxane B 2 Synthesis in C h i l d h o o d Type 1 (Insulin-dependent) Diabetes Me l l i t u s .

G . T R I O L O , G . D A V I , E . G I A R D I N A , F . C A R D E L L A , F . M E L I , A . S T R A N O and G . D . B O M P I A N I (Palermo and Rome, Italy) 1

Platelet F u n c t i o n in Patients with Type 2 Diabetes Mel l i tus : the Effect of Glycaemic C o n t r o l . G . D A V I , M . A V E R N A , I. C A T A L A N O , C . M . B A R B A G A L L O , E . G I O V E N C O , A . C A R R O C C I O , A . N O T A R B A R T O L O and A .

S T R A N O (Palermo and Rome, Italy) 7

G l y c a e m i c C o n t r o l Affects Cel lu lar Sod ium Metabo i i sm. D . S I M M O N S , L . L . N G , M . H A R K E R and T . D . R . H O C K A D A Y (Oxford, U K ) 13

A L o n g i t u d i n a l Study of M u l t i m o d a l E v o k e d Potentials in Diabetes Mel l i tus . G . P O Z Z E S S E R E , P . A . R I Z Z O , E . V A L L E , M . A . M O L L I C A , L . S A N A R E L L I , S. M O R A N O , P . P I E T R A V A L L E , U . D I M A R I O and C .

M O R O C U T T I (Rome, Italy) 17

Trace Elements in B l o o d Cells of Diabe t ic Subjects.

i . R A Z and E . H A V I V I (Jerusalem, Israel) 21

Twenty-four H o u r Excret ion of U r i n a r y C-peptide in Gesta t ional Diabet ic W o m e n before and after Treatment wi th Die t or Die t and Insulin. E . N O R D L Ä N D E R , u . H A N S O N , M . S T A N G E N B E R G and B . P E R S S O N (Stockholm, Sweden) 25

Successful Transplantat ion of Canine Islets of Langerhans after 24 H o u r s Cold-storage.

D . C A S A N O V A , j . w . H E I S E , M . J . F I E L D , s. R . M U N N and D . E . R . S U T H E R L A N D (Minneapol is , U S A ) 31

C o m p a r i s o n of Sites for Transplantat ion of Can ine Pancreatic Microfragments.

M . G . E V A N S , G . L . W A R N O C K and R . v. R A J O T T E (Edmonton , Canada) 35

Improved Intestinal F o r m and F u n c t i o n in Diabe t ic Rats F e d Long- te rm wi th a Polyunsaturated Fa t ty A c i d Diet .

M . K E E L A N , A . B . R . T H O M S O N , M . T . C L A N D I N I N , B . S I N G H , R . R A J O T T E and M . G A R G (Edmonton , Canada) 43

G A S E R E P O R T Erythrocyte Insul in B ind ing in a F a m i l y with Heredi tary M o t o r Sensory Neuropa thy (Charco t -Mar ie -Tooth) with

Superimposed Insulin-dependent Type I Diabetes Mel l i tus . s. A . I V A R S S O N and i . B J E R R E ( M a l m ö , Sweden) 49

Volume 10, Number 2—February 1989 O R I G I N A L P A P E R S

Supravi ta l Di th izone Staining in the Isolation of H u m a n and Rat Pancreatic Islets. W . A . H A N S E N , M . R . C H R I S T I E , R . K A H N , A . N O R G A A R D , I . A B E L , A . M . P E T E R S E N , D . W . J O R G E N S E N , S. B A E K K E S K O V , J . H .

N I E L S E N , A . L E R N M A R K , J . E G E B E R G , H . R I C H T E R - O L E S E N , T . G R A I N G E R , J . K . K R I S T E N S E N , S. B R Y N I T Z and T . B I L D E

(Gentofte, Copenhagen, H v i d o v r e and Glos t rup , Denmark) 53

Diazoxide- induced Long-term Hyperglycemia . I L Slackening of P ro insu l in Convers ion .

i . V A L V E R D E , c. A L A R C O N , A . R O V I R A , F . MALAissE-LAGAE and w . J . M A L A I S S E ( M a d r i d , Spain/Brussels , Belgium) 59

E?etection of Ant i -phosphol ip id (Card io l ip in , Phosphatidylserine) Ant ibodies in the Serum of Patients with N o n Insulin-dependent (Type 2) Diabetes Mel l i tus and Macroangiopa thy . Coexistence of Anti-platelet Reactivity. G . T R I O L O , E . G I A R D I N A , G . S C A R A N T I N O , G . S E D D I O and G . B O M P I A N I (Palermo, Italy) 63

Glycaemic Effects of Tradi t ional European P lan t Treatments for Diabetes. Studies in N o r m a l and Streptozotocin Diabet ic M i c e . s. K . S W A N S T O N - F L A T T , c. D A Y , p. R . F L A T T , B . J . G O U L D and c. J . B A I L E Y (Gui ldford and B i r m i n g h a m , U K ) 69

Effect of L o n g T e r m Glycemic C o n t r o l on the Onset of Ret inopathy in I D D M Subjects. A L o n g i t u d i n a l and Retrospective Study.

M . L . A L V A R S S O N and v. E . G R I L L (Stockholm, Sweden) 75

Effect of Alpha-glucosidase Inhibi t ion on Glucose Profiles in Insulin Dependent Diabetes. R . J . HiLLMAN, M . S C O T T and R . s. G R A Y (Broxburn , U K ) 81

C O N T E N T S V

Effects of Enzyme Induc t ion The rapy on Glucose and D r u g Me tabo l i sm in Obese M i c e M o d e l of N o n - i n s u l i n Dependent Diabetes M e l l i t u s . I. K A R V O N E N , J . H . S T E N G Ä R D , R . H U U P P O N E N , F . G . S T E N B Ä C K and E . A . S O T A N I E M I (Oulu , F in land) 85

M E T H O D S

M i c r o - E L I S A for the Q u a n t i t a t i o n of H u m a n U r i n a r y and Serum Ret inol -b inding Prote in . T . J E N S E N , M . D E C K E R T , A . D A W N A Y and B . F E L D T - R A S M U S S E N (Gentofte, D e n m a r k / L o n d o n , U K ) 93

Determina t ion of Islet C e l l An t ibod ies using an E L I S A System with a Prepara t ion of Rat Insul inoma ( R I N A2) Cells . w . A . S C H E R B A U M , J . SEißLER, u . H E D D E R I C H , B . o. B O E H M , M . S P E C K E R and E . F . P F E I F F E R ( U l m and Frankfurt , F R G ) 97

Volume 10, Number 3—March 1989 O R I G I N A L P A P E R S

C o u l d Bovine Serum A l b u m i n be the Ini t ia t ing Ant igen Ul t imate ly Responsible for the Development of Insul in Dependent Diabetes Me l l i t u s?

M . G L E R U M , B . H . R O B I N S O N and J . M . M A R T I N (Toronto, Canada) 103

Increased Levels of C i r c u l a t i n g Immune Complexes are not Associated wi th Diabetes in B B Rats.

G . C O N T R E A S , T . D Y R B E R G , o. D . M A D S E N , H . M A R K H O L S T and Ä . L E R N M A R K (Gentofte, Denmark ) 109

Access of Ma te rna l Insul in to the R a t Conceptus pr ior to Al l an to i c Placentat ion. T . G . U N T E R M A N , T . A . B U C H A N A N and N . F R E I N K E L (Chicago, U S A ) 115

G l u c a g o n Storage, Release and Degrada t i on by T u m o r a l Islet Cel ls ( R I N m 5 F Line). M . B A R R E T O , i . V A L V E R D E and w . J . M A L A I S S E ( M a d r i d , Spain/Brussels, Belgium) 121

Analys i s of the P r o - o p i o m e l a n o c o r t i n Gene in N o n - i n s u l i n Dependent Diabe t ic Famil ies .

s. O ' R A H I L L Y , P . P A T E L , J . s. W A I N S C O A T and R . C . T U R N E R (Oxford, U K ) 125

R e n a l Effects of P ros tag land in E ! i n Type 2 (Non-insulin-dependent) Diabet ic Patients with Subcl in ica l Nephropathy . Y. NINOMIYA and M . A R A K A W A 129

N e u t r o p h i l A c t i v a t i o n Detected by Increased Neu t roph i l Elastase Ac t iv i ty i n Type 1 (Insulin-dependent) Diabetes Mel l i tus . A . C O L L I E R , M . J A C K S O N , D . B E L L , A . W . P A T R I C K , D . M . M A T T H E W S , R . J . Y O U N G , B . F . C L A R K E and J . D A W E S

(Edinburgh, U K ) 135

Reversal by Vanadate of the Effect of Diabetes on Intestinal G r o w t h and Transport .

J . - J . H A J J A R , M . p. DOBiSH and T . K . TOMicic (Newington and Farming ton , U S A ) 139

Ac t iv i t y of Hepat ic G l u c o s e Phosphory la t ing and N A D P H Generat ing Enzymes in Zucke r Rats.

R . H U U P P O N E N , i . K A R V O N E N and E . S O T A N I E M I (Tu rku and O u l u , F in land) 143

Fa i lu re of Glucose Tolerance Fac tor -conta in ing Brewer's Yeast to Amel iora te Spontaneous Diabetes in C 5 7 B L / K s J D B / D B M i c e . P . R . F L A T T , L . J U N T T I - B E R G G R E N , P . - O . B E R G G R E N , B . J . G O U L D and S. K . S W A N S T O N - F L A T T (Gui ld ford , U K / U p p s a l a and S t o c k h o l m , Sweden) 147

Volume 10, Number 4—April 1989 O R I G I N A L P A P E R S

U r i n a r y IgG4: A n A d d i t i o n a l Parameter in Character iz ing Patients wi th Incipient Diabe t ic Nephropa thy . S. G A M B A R D E L L A , S. M O R A N O , A . C A N C E L L I , P . P I E T R A V A L L E , S. F R O N T O N I , S. B A C C I , U . D I M A R I O and D . A N D R E A N I

(Rome, Italy) 153 E a r l y Alterat ions of Rena l F u n c t i o n in Insulin-dependent Diabe t ic Pregnancies and their Importance in Predic t ing

Pre-eclamptic T o x a e m i a .

p. H . W I N O C O U R and R . J . T A Y L O R (Salford, U K ) 159

Effect of H u m a n P r o i n s u l i n o n B l o o d Glucose C o n t r o l and Metabol ie Instabili ty in Type 1 Diabetes Mel l i tus .

E . B O C K and H . - J . Q U A B B E (Ber l in , F R G ) 165

Transglutaminase-dependent L y m p h o c y t e Transformat ion in Type 2 Diabetes Mel l i tus . K . B E R N T O R P , K . - F . E R I K S S O N and E . B E R N T O R P ( M a l m ö , Sweden) 171

vi C O N T E N T S

Pattern of Obesi ty and Insul in, G l u c a g o n , Sex H o r m o n e B i n d i n g G l o b u l i n and L i p i d s i n Obese A r a b W o m e n . M . K . E M A R A , A . S A A D A H , M . H A S S A N , M . M O U S S A and H . H O U R A N I (Kuwai t ) 175

Metabo l i e Effects of H i g h Glucose Concentrat ions: Inhib i t ion of Hepat ic Pyruvate K i n a s e . F . B E L F I O R E , S. I A N N E L L O , R . C A M P I O N E and G . V O L P I C E L L I (Catania , Italy) 183

Card iac M y o s i n Isoenzyme Shifts in N o n - i n s u l i n Treated Spontaneously Diabe t i c Rats .

v. c. B Ä U M , w . A . C L A R K and D . A . P E L L I G R I N O (Chicago, U S A ) 187 The Pup i l l a ry L igh t Reflex in Diabetes Mel l i tus : Eva lua t ion of a N e w l y Deve loped Infrared L i g h t Reflection

M e t h o d . A . D E V O S , J . T . M A R C U S , J . P . H . R E U L E N , H . F . M . P E T E R S , J . J . H E I M A N S and E . A . V A N D E R V E E N (Amsterdam, The Netherlands) 191

H o w C o m m o n is Diabet ic Nephropa thy?

G . V . G I L L (Wir ra i , U K ) 197

I N D E X O F ÄUTHORS 199

C O N T E N T S v i i

Diabetes Research (1989) 10,97-102 DIABETES RESEARGH © T E V I O T - K I M P T O N 1989

M E T H O D S

DETERMINATION OF ISLET C E L L ANTIBODIES USING AN ELISA SYSTEM WITH A PREPARATION OF RAT INSULINOMA (RIN A2) CELLS

W . A . S C H E R B A U M , J . S E I B L E R , U . H E D D E R I C H , B . O . B O E H M , 1 M . S P E C K E R and E . F . P F E I F F E R

Department of Internal Medicine I, Medizinische Klinik und Poliklinik University of U l m and 'Department of Endocrinology, University Hospital Frankfurt (Received 25 July 1988)

SUMMARY An enzyme-linked immunosorbent assay (ELISA) was established for the detection of islet cell antibodies in human sera. The antigen was prepared from rat insulinoma (RIN A2) cells. Cells were dissociated in lysis buflfer and the lysate was centrifuged at 100,000 x g. The supernatant was used to coat microtiter ELISA plates (lOjig protein/ml in PBS pH 7-2). Non-specific binding sites on the plates were blocked with 2% PBS-BSA. Human test sera were preabsorbed on separate plates using 2% PBS-BSA and incubated on precoated plates at an optimal dilution of 1/10 in 60 mM PBS for 60 min at 37°C. Phosphatase-labeled anti-human IgG serum and Phos­phatase Substrate were applied and the reaction was stopped by adding 3 M NaOH. Out of 90 sera from type I diabetic patients,

47 (52-2%) reacted in the new ELISA whereas none of 15 type II diabetics, 50 sera containing non-islet specific antibodies or 100 normal controls were positive. In the same group of patients, ICA were positive in 63-3%. When both, the ELISA and con-ventional ICA testing were applied, the number of positives was increased to 83%. The ICA-ELISA with the above described antigen preparation provides a well standardized and re-producible test method which is highly specific for type I diabetes. It may therefore be useful for large Screening procedures.

Key words: Islet cell antibodies, ELISA, RIN cell preparation

I N T R O D U C T I O N

T H E detection of islet cell antibodies ( I C A ) in the serum is a hal lmark of insulitis and type I (insulin-dependent) diabetes mellitus (1,2). U s i n g the indirect immunofluor-escence ( I F L ) test on unfixed cryostat sections of human pancreas, 60-90% of patients are positive at the onset of disease, whereas i n normal individuals the antibodies occur in less than 1% of the cases (3-5). The presence in the serum of I C A points to an autoimmune form of diabetes even when the disease Starts atypically at an older age or when the diabetes can be init ially control led by ora l antidiabetic drugs (6-10). The detection of I C A

C o r r e s p o n d e n c e : Dr. W. Scherbaum, Abteilung Innere Medizin I, Medizinische Klinik und Poliklinik der Universität, Robert-Koch-Strasse 1, D 7900 Ulm, West Germany.

may therefore be relevant to the initial aetiological Classification of diabetes (11).

Prospective family studies as well as the systematic investigation of non-diabetic, H L A identical twins and triplets of type I diabetics have shown that I C A may be present in the serum years before the cl inical manifes-tation of diabetes and well before a deficiency of insul in secretion can be established by the sensitive i.v. glucose tolerance test (12,13). Therefore, the determination of I C A has become increasingly important to identify i n ­dividuals at risk to develop type I diabetes among pre-disposed individuals (14-19) or i n popula t ion S c r e e n i n g programs (20-22). Unfortunately, the standardization of the convent ional immunofluorescence test for the detection of I C A has caused mul t ip le problems and the results vary considerably when different donor organs are used as a Substrate. W e now describe a reproducible

0265 5985/89/0200 0097 S06.50 © 1989 T E V I O T - K I M P T O N PUBL1CATIONS 97

detection System for I C A using an enzyme-l inked immunosorbent assay ( E L I S A ) with a preparation of the R I N A 2 cell line.

M A T E R I A L S A N D M E T H O D S

Sera The sera from 90 type I diabetics (32 males/58 females, mean age

19 yrs, ränge 17-45 yrs) were investigated. The duration of diabetes varied between onset of disease and a 17 yr course. Thirty sera could be gained within three months of diagnosis.

Fifteen sera from type II diabetics were tested (10 males/5 females, mean age 56 yrs, ränge 39-74 yrs). Only patients with more than 2 yrs of diabetes were studied. A l l patients were under treatment with diet and antidiabetic drugs. Cases of secondary diabetes were excluded.

To test the specificity of results, 50 sera from non-diabetic individuals were also investigated. Twelve of them had autoantibodies (Ab) to exocrine pancreatic tissue, five thyroid microsomal Ab, two adreno-cortical Ab, six gastric parietal cell Ab, five d s D N A Ab, 10 other antinuclear Ab, five smooth muscle Ab, five anti-mitochondrial Ab.

One hundred sera from apparently healthy blood donors were also investigated. None of these normals had a family history of type I diabetes. The mean age of the control group was 24-5 yrs, ränge: 10-30 yrs.

A l l sera were stored at - 2 0 ° C . Before testing in the cell assay, the sera were decomplemented by a 30 min incubation in a water bath at 56°C.

R I N A I C e l l s This insulin-producing rat insulinoma cell line R I N A2 derived

from the subclone R I N m5F was kindly gained from Prof. Ammon, Tübingen. The cells were cultured in R P M I 1640 medium supplemented with 10% v/v heat-inactivated fetal calf serum, 5% glutamin, penicillin lOOU/ml and streptomycin lOOjig/ml at 37°C in an atmosphere containing 95% air and 5% C 0 2 .

The cells were fed every second day and passaged every four days. The cells were dissociated from the culture surface by incubation with t rypsin-EDTA Solution (0-5 mg/ml and 0-2 mg/ml respectively in calcium/magnesium-free Hank's balanced salt Solution) for 4-8 min at 37°C. Cells were then washed twice in their original medium and resuspended. At the final harvesting, the cells were washed three times in R P M I 1640 without fetal calf serum.

P r e p a r a t i o n of P a n c r e a t i c I s l e t C e l l s f r o m Rat P a n c r e a s Wistar rats with a body weight of about 150g were used throughout.

Islets were prepared after the method of Lasy and Kostaniovsky (22). After nembutal anesthesia, the pancreas was ballooned by an intra-aortic injection of a Solution of neutral red. The pancreas was removed, mechanically minced with scissors and digested with collagenase (Sigma, type V, 7*5 mg/ml tissue) at 37°C. After microscopic control the digestion was stopped with ice cold Hank's Solution. Exocrine and endocrine pancreas were separated using the discontinuous Ficoll density gradient centrifugation. The islets were selectively picked with a pipett under a stereo microscope. Picked islets were incubated in E G T A Solution for 30 min at room temperature. The yield of islets was about 300/rat, and from these, about 200,000-300,000 islet cells were gained.

A n t i g e n P r e p a r a t i o n R I N A2 cells were washed three times in phosphate buffered saline for

10 min each, and centrifuged at 200 x g in between. Lysis of cells was obtained by incubation with lysis buffer (Triton X-100 lO ml /L , Trasylol 50ml/L, P M S F 0-17g/L, N a C l 8-8 g/L, Hepes l l - 9 g / L , p H 7-6) for 90 min at 4°C under constant stirring. The lysate was subsequently centrifuged at 100,000 x g for 90 min at 4°C.

The supernatant was designated as fraction A. For the isolation of the soluted membrane glycoproteins (23), the supernatant was further processed by affinity chromatography using Lentil-Lectin

Sepharose 4B (Pharmacia): the supernatant was mixed with the gel and incubated overnight in the roller at 4 ° C It was then transmitted to the ice cooled chromatography column. The unbound fractions were washed with P B S + 1 M N a C l and eluted with PBS + 0-5 M methyl-alpha-D-mannopyranosid. This preparation was designated as fraction B. The protein concentration was determined according to the method ofBradford (24).

C o a t i n g of E L I S A P l a t e s To determine the antibody binding of patients' sera to surface

membranes of intact islet cells, 100 ul of Suspension from freshly prepared viable rat islet cells and R I N A2 cells were submitted to microtiter plates ((a) Immulon M 129B, (b) Nunc-Immuno plate) at a density of 1 x 10 5cells/ml in R P M I 1640 medium. In different experiments, cells were seeded with or without preincubation of plates with poly-L-lysin (Sigma), respectively. The cell suspensions were primarily attached to the microtiter plates by centrifugation at 100 x g for 5 min. Suspended cells were cultured for 24 h and they were finally washed twice by gentle incubation with PBS p H 7-4. For their use in E L I S A tests, the cells on coated plates either remained unfixed, or they were fixed with 4% paraformaldehyde or 0-25% glutaraldehyde in different experiments.

The coating of plates with islet cell preparations was performed with six different buffers. (a) 50 m M Carbonate buffer, p H 9-6. (b) 50 m M Carbonate buffer, p H 7-2. (c) 0-2 M Carbonate buffer, p H 8-3. (d) 0-2 M Carbonate buffer. p H 7-2. (e) 150mM PBS, p H 7-2. (0 150mM PBS, p H 5-5.

E L I S A Test P r o c e d u r e Coated plates were washed three times with PBS p H 7-4. Non-specific

binding sites were blocked by incubation with 2% bovine serum albumin and subsequent washing with PBS. Human sera were diluted in 60 m M PBS buffer and applied in doubles. After incubation with sera for 60 min the plates were washed three times with Tris-Tween 01%, and incubated for 30 min with phosphatase-conjugated anti-human IgG (Tago) diluted 1/2,000 in PBS. After washing four times, 100 1̂ of Substrate Solution were added to each well. The reaction was stopped after visual control by adding 3 M N a O H . The coloured reaction product was measured at 410nm with a Dynatech microELISA autoreader M R 600.

D e t e c t i o n ofBSA A n t i b o d i e s Because all E L I S A plates had been blocked with 2% BSA and the

R I N cells had been originally cultured with B S A , the positive reaction of a serum could be due to the presence of albumin antibodies. There­fore, all positive sera were preabsorbed on E L I S A plates coated with 2% BSA and they were then retested using the above E L I S A System.

I n d i r e c t I m m u n o fluorescence Test For the detection of conventional I C A the indirect I F L test was

applied using unfixed cryostat sections of human pancreas from a donor with blood group O (25).

R E S U L T S

Adhesion of Cells to the E L I S A M i c r o t i t e r Plates The adhesion of viable R I N A 2 cells and of freshly

isolated rat islet cells to the microt i ter plates was in-sufficient. It was only marginal ly improved by pre­incubat ion of plates wi th P o l y - L - L y s i n . Even after extremely cautious washing and incubat ion procedures there were large gaps i n the or ig inal ly dense cell layer (Figure l a and lb ) . The attachment of cells was not improved after fixation wi th 4% paraformaldehyde, but fixation wi th 0-25% glutaraldehyde resulted in a good adhesion throughout the E L I S A procedure.

98 S C H E R B A U M et a l

Figure l(a). Phase contrast photography of an E L I S A microtitre plate coated with unfixed R I N A2 cells before further testing.

Figure l(b). At the end of the E L I S A procedure many R I N A2 cells were washed off from the bottom both with native plates as well as after preincubation with Poly-L-Lysin . These holes caused an unacceptable non-specific binding of sera.

E L I S A - I C A Test Results with V i a b l e Islet Cells The large gaps in the coat ing cell layer observed wi th

unfixed or formalin-fixed islets resulted in a significant nonspecific b ind ing of serum components to the plates so that no reproducible results cou ld be obtained. W h e n glutaraldehyde-fixed R I N A 2 cells were used, the ind iv idua l E L I S A test results were not consistent and they d id not a l low to dist inguish type I diabetics from normal individuals . W h e n freshly isolated glutar­aldehyde-fixed rat islet cells were used, the mean b inding of type I diabetic sera was higher i n each test than the one obtained wi th cont ro l sera (Figure 2). However , the i n ­div idual reproducibi l i ty of positive and negative test results was insufficient. The in t ra assay variance was 94%, the inter assay variance was as high as 31%. This system was therefore abandoned.

E L I S A with F r a c t i o n A f r o m R I N A 2 Cells E v a l u a t i o n of the optimal coating buffer F igure 3 shows the E L I S A test results in relat ion to the sort and p H of

Figure 2. Results of an E L I S A using freshly isolated rat islet cells fixed with 0-25% glutaraldehyde. The sera from type I diabetics gave an increased overall binding as compared to controls. However, the individual test results were very variable and not clearly reproducible. (#), Type I diabetics; (X), normal controls.

coat ing buffer used for the d i lu t ion of the antigen. The antigen (50 jil/well) was incubated overnight at 4°C. The further test condit ions were as described below (Figure 6). The highest b inding of diabetic sera and the best d iscr imi-nat ion between diabetics and no rma l controls was achieved with a phosphate buffer p H 7-2.

O p t i m a l antigen concentration for coating The R I N A 2 cell fraction A was used at di lut ions 1/50-1/10,000. It is evident from Figure 4 that the best d iscr iminat ion between diabetic and normal human sera was obtained at a d i lu t ion of 1/200 which contained 10 jxg protein/ml .

O p t i m a l serum d i l u t i o n The highest ant ibody b inding was obtained with a serum d i lu t ion of 1/5 (Figure 5). The absorpt ion values of both patients and controls

ph9.6 ph7.2 carbonate buffer 1

ph8.3 ph7.2 carbonate buffer 2

coating buffer

RS

ph7.2 ph5.5 phosphate buffer

Figure 3. Influence of various coating buffers upon the I C A - E L I S A with the R I N A 2 cell preparation (fraction A). The mean values of results obtained with 10 ICA-positive type I diabetic patients and with 10 controls are given in the figure. ( • ) , Type I diabetics; ( ü ) , normal controls.

D E T E R M I N A T I O N O F I S L E T C E L L A N T I B O D I E S 99

0.7--

0.1--

50 100 200 300 500 1000 5000 10000 x 10̂

dilution of fraktion A (log scale)

Figure 4. Influence of the concentration of coating antigen upon I C A -E L I S A test results. Fraction A from R I N A2 cells was diluted in PBS p H 7-2 as a coating buffer and 100 ul/well were used. The figure gives the mean values of results obtained with 10 sera from type I diabetics and from 10 normal controls. The best discrimination between the two groups was obtained at a serum dilution of 1/200 (= 10 ug protein/ml). (#), Type I diabetics; (X), normal controls.

decreased wi th increasing serum di lu t ions used i n the assay. The best d i sc r imina t ion between patients and controls was obtained wi th sera appl ied 1/10. A t this d i lu t ion the mean value of patients versus mean value of controls was 2*72.

E L I S A with F r a c t i o n B f r o m R I N A 2 Cells W h e n the membrane g lycopro te in fraction was used

for coating, diabetic sera d i d not produce an increased

1.0--

0 . 8 -

5 10 20 1*0 80 160 320 640 x W

dilution of test serum

Figure 5. Influence of the concentration of test serum upon the specific binding in the I C A - E L I S A . Fraction A from R I N A 2 cells was used at 1/200. PBS p H 7-2 as a coating buffer. The best discrimination between type I diabetic patients and controls was achieved at a serum concentration of 1/10. (#), Type I diabetics; (X), normal controls.

Preabsorption of sera on BSA-coated plates lh at 37oc

• coating of microtiter plates with RIN A2 c e l l f r a c t i o n A

(100 u l / 10 ug/ml in PBS pH 7.2) 18 h at 4<>C

• wash 3x with PBS

block non-specific binding s i t e s with 2% PBS-BSA 60 min at 37 °C

remove BSA by skipping of plates

incubation of serum (100 u l , 1/10 i n PBS) 60 min at 37 <>C

wash 3x with TRIS buffer

phosphatase-labeled anti-human IgG (100 u l , 1/5/000 i n PBS) 30 min at 37 <>C

wash 3x with TRIS buffer

•

Phosphatase Substrate (100 u l / 1 mg/ml Substrate)

stop reaction with 100 ul 3M NaOH read absorption at 410 nm

Figure 6. Schematic description of the optimal I C A - E L I S A using fraction A from the R I N A2 cell preparation.

bind ing wi th any of the procedures appl ied for fraction A . Therefore, fraction B was not used for further testing.

O n the basis of the results above described the E L I S A for I C A testing was applied as indicated in F igure 6.

Specificity o f the islet cell a n t i b o d y — E L I S A and its cor-r e l a t i o n with conventional I C A testing Forty-seven of the 90 sera (52-2%) from type I diabetic patients were positive in the new E L I S A . Fifty-seven (63-3%) were I C A -positive as measured by the indirect immunofluorescence test. N o n e of the sera from type II diabetics or from patients with non-islet cell specific autoantibodies or controls was positive in the E L I S A . The results obtained with the E L I S A and with the I F L test on 30 sera from type I diabetics gained wi th in three months of diagnosis are given in Table 1. I C A were positive in 20 (67%),

Table 1 Results of the I C A - E L I S A using fraction A from R I N A2 cells compared with the results obtained with conventional ICA deter-mination using the indirect immunofluorescence test on unfixed cryostat sections of human pancreas. The 30 sera were obtained within the first three months of diagnosis of type I diabetes

I C A ( I F L - T e s t )

p o s i t i v e n e g a t i v e

E L I S A positive 14 5 E L I S A negative 6 5

100 S C H E R B A U M et a l

E L I S A - I C A reacted i n 19 (63%) of the cases. W i t h a combina t ion of the two methods the number of positives was increased to 25 (83%).

Antibodies to bovine serum a l b u m i n ( B S A ) Eleven out of 80 sera (19%) from type I diabetics contained B S A anti­bodies. After preabsorpt ion wi th B S A , eight of these sera remained posit ive i n the I C A - E L I S A indicat ing that they contained bo th B S A antibodies and islet cell antibodies. N o n e of the sera from type II diabetics, four of the sera from patients wi th associated autoantibodies and three out of 100 n o r m a l h u m a n sera were positive for B S A antibodies.

D I S C U S S I O N

The feasibility of I C A determinat ion i n large popula t ion Screening programs depends o n an easily reproducible and objective test System. F o l l o w i n g a Standard p ro tocol with the convent ional I F L test on unfixed cryostat sections of h u m a n pancreas even specialized laboratories recorded a wide var iabi l i ty of titres ranging from negative to 128. W i t h l ow titre sera the interlaboratory concordance for a posit ive result was only 52 -79% (20). This is main ly due to the subjective reading of the I F L test and the variance of human pancreases used as an antigen.

The new E L I S A using a rat insu l inoma ( R I N A2) cell preparat ion for the detection of islet cell antibodies provides the advantage of easy s tandardizat ion and objective measurement of I C A . N o t only antibodies to membrane proteins but also reactivities directed to soluble islet cell proteins can be measured in the assay. This may expla in part of the positive E L I S A test results in ICA-negat ive sera from type I diabetics. In contrast to the immunofluorescence test, the E L I S A may also detect low affinity antibodies. The negative E L I S A - I C A results wi th sera conta in ing I C A as determined by indirect im­munofluorescence may be explained by the l imi ted cross-reactivity of h u m a n sera wi th the R I N cell antigens.

The disease specificity of E L I S A - I C A is shown by the fact that patients and n o r m a l individuals as well as patients wi th non-diabetes-specific antibodies were constantly negative in the new assay. L i k e convent ional I C A the E L I S A - I C A can therefore be considered as a marker of au to immune insulit is . W i t h the E L I S A System here described the lower l imi t of detection of islet cell specific autoantibodies i n the serum of type I diabetics was clearly elevated. The two tests may be used comple-mentary since b io logica l ly false positive results are very rare in the I F L (3-5) and they were not detected in our E L I S A .

Some authors have tried to perform an E L I S A for the detection of islet cell surface antibodies ( I C S A ) by coating microtitre plates wi th fixed pr imary rat islet cells (26) or with R I N m 5 F cells (27). The promis ing first reports,

however, were not reproduced by other authors. In our o w n experiments, the at tachment of native cells to the precoated plates was insufficient to resist the E L I S A procedure. E v e n wi th gentle washing cell-free holes appeared which p roduced an unacceptable increase of non-specific b ind ing to the plates. Sufficient attachment of cells was achieved by fixation wi th 0-25% glutar­aldehyde, but this resulted in an increase of false negative and false posit ive results. T h e change of islet cell antigens by fixatives has also been documented in previous reports (28,29).

T h e presence of B S A ant ibodies in human sera is well explained by the ingest ion of c o w mi lk or beef (30), but it may disturb the Interpretation of a positive I C A - E L I S A result so that a preabsorp t ion of sera wi th B S A is recom-mended. B S A antibodies have also been detected by other authors (31,32). U s i n g an immunoprecip i ta t ion method, C o l m a n et al. (32) were able to discriminate the B S A reactivity f rom I C A . In contrast to the above ment ioned report, we found an increased prevalence of B S A antibodies i n the sera from type I diabetic. A l t h o u g h this finding may be due to a selection of our patients it is wor th while to study B S A antibodies i n a larger group of type I diabetic patients.

T h e I C A - E L I S A here described is well standardized and highly specific for type I diabetes. It may therefore be considered as a valuable t o o l for islet cell ant ibody testing i n large Screening investigations. Fur ther research should be a imed at the i so la t ion of the islet cell antigen which w i l l a l low to establish the op t ima l test for I C A deter­mina t ion .

A C K N O W L E D G E M E N T S

We thank Mrs . Trischler for excellent help with the R I N cell cultures and Mrs . Ketzler-Sasse for technical assistance with the preparation of antigen. This work was supported by the Deutsche Forschungsgemein­schaft (Sehe 225/2-2 and Sehe 225/3-1) as well as the Riese and Scheidel Foundation (B.O.B.).

R E F E R E N C E S

1. Bottazzo, G . F., Pujol-Borrell, R. and Gale, E. A . M . (1987). Auto-immunity and type I diabetes: bringing the Störy up to date. In The D i a b e t e s A n n u a l , 3. K . G . M . M . Alberti and L . P. Kra l l (Hrsg), S. 15-38. Elsevier, Amsterdam.

2. Vardi , P., Dibella, E . E. , Pasquarello, T. J. and Sirkanta, S. (1987). Islet cell autoantibodies: pathophysiology and clinical applications. D i a b e t e s C a r e , 10, 645-654.

3. Lendrum, R., Walker, G . , Cudworth, A . G. , Theophanides, C , Pyke, D . A. , Bloom, A . and Gamble, D . R. (1976). Islet-cell antibodies in diabetes mellitus. L a n c e t , ii, 1273-1276.

4. Irvine, W. J , M c C a l l u m , C . J., Gray, R. S., Campbell, C. J., Duncan, L . J . P., Farquhar, J . W , Vaughan, H . and Morris , P . J . (1977). Pancreatic islet-cell antibodies in diabetes mellitus correlated with the duration and type of diabetes coexistent autoimmune disease, and H L A type. D i a b e t e s , 26, 138-147.

5. Del Prete, G . F., Betterle, C , Padovan, D., Erle, G . , Toffolo, A . and Bersahi, G . (1977). Incidence and significance of islet-cell auto-

D E T E R M I N A T I O N O F I S L E T C E L L A N T I B O D I E S 101

antibodies in different types of diabetes mellitus, Diabetes, 26, 909-915.

6. Haies, A. H.-J. (1967). Plasma levels of glucose, non-esterfied fatty acid, glycerol and insulin four years before the onset of diabetic ketoacidosis. Lancet, ü, 389-390.

7. Johansen, K. (1981). Mild carbohydrate intolerance developing into classic juvenile diabetes, Acta M e d . Scand,, 189, 337-339.

8. Orchard, T. 1 and Rosenbloom, A. L. (1985). The development of insulin-dependent diabetes mellitus among relatives, Diabetes C a r e , 8(SuppL 1), 45-50.

9. Rosenbloom, A. L., Hunt, S, S., Rosenbloom, E. K. and MacLaren, N, K. (1982). Ten-year prognis of impaired glucose tolerance in siblings of patients with insulin-dependent diabetes. Diabetes, 31, 385-387.

10. Tarn, A. C , Smith, C. P., Spencer* K. M . , Bottazzo, G. F. and Gale, E. A. M , (1987), Type I (insulin dependent) diabetes: a disease of slow clinical onset? B r i t , M e d . J . , 294,342-345.

11. Wilson, R. M . , Van der Minne, P. and Deverill, I. (1985), Insulin dependence: problems with the Classification of 100 consecutive patients. D i a b e t i c M e d . , 1,167-172,

12. Gorsuch, A. N,, Spencer* K. M. , Lister, J., McNally, J. M . , Dean, B. M . , Bottazzo, G . F. and Cudworth, A. G. (1981). The natural history of Type I (insulin dependent) diabetes mellitus: evidence for a long prediabetic period. Lancet, ii, 1363-1365.

13. Sirkanta, S., Ganda, O. M . P., Rabizadeh, A., Soeldner, J. S. and Eisenbarth, G . S. (1985). First degree relatives of patients with type I diabetes mellitus: islet-cell antibodies and abnormal insulin secretion. N . E n g l . J. M e d . , 313,461-464.

14. Barnet, A. H. , Eff, C , Leslie, R. D. G . and Pyke, D. A. (1980). Diabetes in identical twins: a study of 200 pairs, D i a b e t o l o g i a , 20, 87-93.

15. Gottlieb, M . S. and Root, H . F. (1968). Diabetes mellitus in twins. Diabetes, I i , 693-704.

16. Lorini, R., Larizza, D., Livieri, C., Cammereni, V., Martini, A., Plebani, A., Zanaboni, D. and Sevefi, F. (1986). Autoimmunity in children with diabetes mellitus and their relative. E u r o . J. Pediatr., 145,182-184.

17. Betterle, C , Presetto, F., Pedini, B., Moro, L., Slack, R. S., Zanette, F. and Zanchetta, R. (1987). Islet cell and insulin autoantibodies in organ-specific autoimmune patients. Their behaviour and predictive value for the development of Type 1 (insulin dependent) diabetes mellitus. A 10-year follow-up study. D i a b e t o l o g i a , 30, 292-297.

18. Schulz, B., Witt, S., Hildmann, W., Rathmann, K. P., Strese, J, and Keilacker, H . (1984). Islet-cell antibodies in individuals at increased risk for I D D M . E x p . C l i n . E n d o c r i n o l . , 83,192-198.

19. Tarn, A. C , Thomas, J. M. , Dean, B. M . , Ingram, D., Schwarz, G ,

Bottazzo, G. F. and Gale, E. A. M . (1988). Predicting insulin-dependent diabetes. Lancet, i, 845-850.

20. Bergua, M. , Sole, J., Perez, M . C , Recasens, A., Fernandez, J., Casamitjana, R. and Gomis, R. (1987). Prevalence of islet cell antibodies, insulin antibodies and hyperglycaemia in 2291 school-children, D i a b e t o l o g i a , 30,724-726.

21. Vissing, H., Papadopoulos, G. and Lernmark, A. (1986). Monoclonal antibodies against pancreatic islet-cell surface antigen selected by flow cytoflourometry. Scand. J. I m m u n o L , 23, 425-433,

22. Lacy, P. E. and Kostaniovsky, M . (1967). Method for the isolation of intact islet of Langerhans from the rat pancreas. Diabetes, 16, 35-39,

23. Pearlstein, E. (1977). Isolation and partial characterisation of plasma membrane glycoproteins from normal and transformed mammalian cells employing plant lectin affinity chromatography. E x p . C e l l Res., 109,95-103.

24. Bradford, M . M . (1976). A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. A n a l Biochem., 72,248-254.

25. Scherbaum, W. A., Mirakian, R., Pujol-Borrell, R., Dean, B. M . and Bottazzo, G. F. (1986). Immunocytochemistry in the study and diagnosis of organ-specific autoimmune disease. In I m m u n o -chemistry. M o d e r n Methods and Applications (edited by J. M . Polak and S, Van Noorden), pp. 456-476. Wright, Bistrol.

26. Sai, P. and Maurel, C. (1983). Enzyme-linked immunosorbent assay for quantitation of islet cell antibodies. J. I m m u n o L M e t h . , 64, 189-198.

27. McEvoy, R., Franklin, B, H . and Ginsberg-Fellner, F. (1986). Bio-logical differences in a rat insulinoma cell line obtained from different laboratories do not affect binding of human anti-islet immunoglobulins. Diabetes, 3, 339-344.

28. Eisenbafth, G. S., Linnenbach, A., Jackson, R., Scearce, R. and Croce, C. M . (1982). Human hybridomas secreting anti-islet auto­antibodies. N a t u r e , 300,264-267.

29. Bottazzo, G. F., Scherbaum, W. A. and Hanafusa, T. (1983). Recent advances in the study of autoimmune endocrine diseases by the use of immunofluorescence. A n n . N . Y . Sei. U S A , 420,275-280.

30. Bock, S. A. (1980). Food sensitivity: a critical review and practical approach. A m , J. D i s . C h i l d . , 134,973-982.

31. Gregor, J., Iberg, N , Berger, W. and Flückinger, R. (1986). Albumin-directed antibodies in diabetes: demonstration of human serum albumin-directed IgM autoantibodies. D i a b e t o l o g i a , 29, 481-484.

32. Colman, P. G., Campell, I. L., Kay, T. W. K. and Harrison, L. C. (1987). 64.000-Mr autoantigen in type I diabetes. Evidence against its surface location on human islets. Diabetesr 36,1432-1440.

102 S C H E R B A U M et al.