Europ. J. Pediat. 122, 217--222 (1976) 9 by Springer-Verlag 1976
Insulin Antibodies in 104 Children with Diabetes mellitus P. W. Nars, G. Herz, and J. Girard Department of Endocrinology (Head: P.D. Dr. J. Girard), University Children's Hospital, Basel (Head: Prof. Dr. G. Stalder) l~eeeived March 1, 1976 Abstract. Binding capacity for porcine, bovine, and human insulin was estimated in 104 diabetic children using homologous systems of iodinated and noniodinated insulin. All patient sera bound porcine and bovine insulin whilst 11.5~o did not bind human insulin. There was no clear correlation between duration of insulin treatment and binding capacity. However, some patients with only short duration of treatment had high binding capacity for porcine insulin. The binding capacity for human insulin was low in most patients. High insulin requirement was on average combined with high binding for bovine insulin. Cases of diabetes, which are difficult to control, treated with individual mixtures of porcine and bovine insulin, often showed high binding for porcine and also for human insulin. The value of estimations of insulin-binding capacity in diabetic children is debatable. However, the high incidence of antibodies seems to justify the use of low immunogenie (monocomponent) insulin preparations. Key words: Diabetes mellitus - - Insulin antibodies. P o l y p e p t i d e h o r m o n e s for t h e r a p e u t i c a l use are g e n e r a l l y p r o d u c e d b y e x t r a c t i o n a n d s u b s e q u e n t purification from a n i m a l or h u m a n glands. I f t h e s t r u c t u r e of p e p t i d e s from o t h e r species differs from t h e h u m a n endogenous hormone, such p r e p a r a t i o n s can be a n t i g e n i c if i n j e c t e d into p a t i e n t s . I n 1956, Berson a n d Y a l o w [1] used insulin l a b e l e d w i t h r a d i o i o d i n e to d e m o n s t r a t e t h e presence o f insulin a n t i b o d i e s in d i a b e t i c s t r e a t e d w i t h insulin. Since this basic w o r k t h e r e h a v e been several r e p o r t s on insulin a n t i b o d i e s in d i a b e t i c a d u l t s b u t o n l y few studies in d i a b e t i c children [8, 10, 11]. J u v e n i l e d i a b e t e s is g e n e r a l l y t r e a t e d w i t h insulin injections once or twice daily, w h e r e b y porcine, bovine, or a c o m b i n a t i o n o f porcine a n d b o v i n e insulin is used. T h e aim o f t h e p r e s e n t s t u d y was t h e a s s e s s m e n t o f b i n d i n g c a p a c i t y for porcine, bovine, a n d h u m a n insulin in s e r u m of 104 d i a b e t i c children on insulin treatment. M a t e r i a l and Methods One hundred and four juvenile diabetic patients aged 7--20 years (mean 13 years) were studied. Their diabetes mellitus had been known and treated for 6 months--17 years (mean 5 years). All patients injected insulin once or twice daily, and in all children the diabetic state was well controlled. Thirty two patients used an insulin preparation of bovine origin, 72 injected either a commercial mixture of porcine and bovine origin or prepared their mixture themselves according to varying insulin requirements. Blood was withdrawn from an anteeubital vein before insulin injection, generally in the early morning. The blood samples were allowed to clot at room temperature, the serum was separated by centrifugation, and stored a t - - 2 0 ~ until analysis. The sera were investigated for insulin antibodies using a modification of the method described by Dunlop and Court [4]. Porcine (Aetrapid Novo), bovine (Insulin Lente Novo),
218
P . W . Nars et al.
Table 1. Method used for the estimation of binding capacity for porcine, bovine, and human insulin in sera of patients with diabetes mellitus on insulin treatment 0.1 ml Incubation medium containing 0.2 ~U of porcine, bovine, or human I125-insulin 0.1 ml Incubation medium containing 0---1000 t~U of corresponding standard insulin 0.1 ml Patient serum diluted in 5.9% human serum albumin Incubation for 16 h at H-4~ Charcoal separation (0.5 ml 0,5% chareoal--0.05~o dextran 80).
and human (batch 12569 mC, a gift from Dr. L. Heding, Novo A/S) insulin were labeled with iodine 125 according to the method of Greenwood et al. [6]. The calculated specific activity varied between 90 and 110 tLCi/ttg. All incubation studies were done overnight at +4~ in a phosphate buffer containing human serum albumin. Free and antibody-bound hormone were separated by adsorption to dextran-coated charcoal [7]. Initially serum dilutions were incubated with 0.2 ~tU of the tracer preparations. For the assessment of binding capacity, patient sera were diluted to yield a 30% binding of the tracer. The amount of noniodinated hormone required to give a significant displacement of the radioactive tracer (defined as a 5% drop in binding percents) was determined by the addition of varying concentrations of nonlabeled insulin. All assays were done in duplicate. The method is summarized in Table 1. Statistics were done using Student's t test and the Z~ test.
Results All p a t i e n t sera w i t h o u t e x c e p t i o n were f o u n d t o b i n d a significant a m o u n t of 0.2 ~ U o f i o d i n a t e d porcine a n d b o v i n e insulin. Twelve o u t of 104 sera (11.5~ d i d n o t b i n d i o d i n a t e d h u m a n insulin. T h e r e was a wide s c a t t e r in t h e a m o u n t o f u n l a b e l e d insulin r e q u i r e d for a significant d i s p l a c e m e n t of t h e l a b e l e d h o r m o n e (range t e s t e d : 1 - - 1 0 0 0 ~U). T h e b i n d i n g capacities for porcine, bovine, a n d h u m a n insulin in t h e 104 sera are s u m m a r i z e d in T a b l e 2. A l t h o u g h m o s t sera b o u n d i o d i n a t e d h u m a n insulin, o n l y 18 (17.8%) h a d a b i n d i n g c a p a c i t y of 40 txU/ml or more. I n m o s t p a t i e n t s t h e b i n d i n g capacities for porcine a n d b o v i n e insulin were of t h e s a m e m a g n i t u d e ; 18 sera showed m u c h h i g h e r b i n d i n g o f porcine insulin, in 15 t h e r e l a t i o n was reversed. T h e c o r r e l a t i o n b e t w e e n d u r a t i o n o f insulin t r e a t m e n t a n d a n t i b o d y - b i n d i n g c a p a c i t y is shown in F i g u r e 1. Seven o u t of 18 p a t i e n t s w i t h o n l y 1 y e a r of t r e a t -
Table 2. Binding capacity for porcine, bovine, and human insulin in sera of 104 diabetic children on insulin treatment Insulin-binding Number of patients capacity ([zU/ml) < 40 40 < 100 100< 1000 > 1000
35 36 19 i4
: : : =
34% 35% 18% i3%
porcine
37 38 19 10
: = = =
36% 37% 18% i0%
bovine
86 = 83% 16 = 15% 2 : 2% 0 human
Insulin Antibodies in 104 Children with Diabetes mellitus
105
porcine 9,
219
human
bovine
9
104 Q
Q*
Z)
:= 103
B
9 1 7 6 1 4 9 99
~
,~176149 ~
I
D
I Q
9
9
~149
t02 40
0
0
Q
9
.~149
9
~ ~ 1 4 9 1 4 99 1 4 9
:*%,~149
9
000
~ 9
,
9
9
9
~
9176
;9
. . . .
F'""
~ 1 7 6 1 4 9 1 4 9o1 4~9
~
'
10
,
15y
i
1'o
15y
lb 'T'sy
Fig. 1. Individual values of binding capacity for porcine, bovine, and human insulin compared to duration of insulin treatment in 104 diabetic children. Note logarithmic scale ment had a high binding ( > 1000 FU/ml) for porcine insulin. Compared to those with longer duration of treatment this difference is statistically significant (P 0.01). From the second year of treatment on, there is a fall in mean binding of porcine insulin. The mean binding of bovine insulin seems independent of duration of treatment. Because of the high percentage of sera binding less than 40 FU/ml of human insulin, no mean values could be calculated. The relative number of sera with measurable binding was independent of duration of treatment. The daily insulin requirement in the patients studied varied greatly, although a positive correlation between insulin dosage in U/m 2 body surface and duration of treatment could be calculated (P 0.01). For a correlation between insulin requirement and binding capacity the patients were divided into three groups: (1) insulin requirement less than 18 U/mS/day, (2) 18--29 U/mS/day, and (3) more than 29 U/m2/day. Only the third group of patients requires insulin dosages clearly above normal endogenous production [8]. These results are summarized in Table 3. The patients in group 1 had a significant later onset of their diabetes and also of duration of treatment, whilst the mean chronologic age of the three groups showed Table 3. Comparison between insulin requirement and porcine, bovine, and human insulin binding capacity in serum. Mean and standard deviation Insulin requirement (U/m2/day) 1000 ~U/ml (mean binding capacity ~U/ml)
10.1•
3.1~2.9
6=22% (6270) 5 = 10% (1740) 3 = 12% (990) porcine
18 30 (n = 25)
7.8 -4- 4.0
6.0 :L 4.0
1---- 4% (140) 6 = 12% (1010) 3 = 12% (570) bovine
Patients withbinding capacity _->40 FU/ml 6=22% 10 = 19% 2= human
8%
220
P.W. Nars et al.
Table 4. Comparison between type of insulin used for treatment of diabetes mellitus and porcine, bovine, and human insulin-binding capacity in serum Insulin therapy
Mean dosage (U/m~/d)
Duration (years)
Patients with binding Patients with capacity > 1000 ~U/ml binding capacity (mean binding capacity __>40 f~U/ml ~U/ml)
Bovine (n~--32)
22.8
6.0
3 = 9% (730)
2 = 6% (320)
2=
20.9
3.8 4.7
3 ~- 10% (12so) 5 ~ 12% (560) bovine
3 = 10%
25.0
2 = 7% (900) 9 ~ 21% (4630) porcine
Bovine Jr porcine: Commercial mixture (~ = 29) Mixed individually (n -~ 43)
6%
12 ~ 28~o human
no difference. With greater insulin requirement there was on average a rise in binding capacity for bovine insulin, for porcine insulin the results were the reverse. The relative number of patients with a binding capacity of more than 1000 ~U/ml shows the same picture. Patients with high insulin requirement had relatively less frequently a measurable binding capacity for h u m a n insulin. The data on type of insulin used for treatment and binding capacity are summarized in Table 4. The number of patients with high binding capacities for porcine insulin was elevated in those using an individual mixture of bovine and porcine insulin, as was the mean binding capacity. Patients on bovine insulin alone showed somewhat lower binding capacities for bovine insulin as those on bovine plus porcine insulin. The same was the case as to binding of h u m a n insulin. All these figures failed to reach statistical significance.
Discussion As in the study of Murty et al. [10] there was a significant binding of iodinated porcine and bovine insulin in all sera of juvenile diabetics on insulin treatment for at least 6 months. I n contrast, Hiirter and Kiihnau [8] and Ortved Andersen [11] found no binding in 9% of their patients. Incorporation of iodine into the insulin molecule might lower immunoreactivity. Berson and Yalow [2] found no impairment with an incorporation of 10 ~0 ~Ci per ~g of hormone. The porcine, bovine, and h u m a n tracer preparations used in this study had a specific activity of 90--110 ~Ci/~g. High binding to guinea pig antiinsulin sera indicates intact immunoreaetivity of all tracer hormones. The physical characteristics of the h u m a n tracer preparation used were equal to the animal ones. Still 11.5% of the patients showed no significant binding of h u m a n insulin. Most of the sera binding iodinated h u m a n insulin had a binding capacity of less than 40 ~U/ml. Berson and Yalow [2] demonstrated t h a t sera from diabetics bind h u m a n insulin much less strongly t h a n porcine or bovine insulin. According to these authors the difference in binding characteristics is due to energetically different reactions with the same antibody combining sites and
Insulin Antibodies in 104 Children with Diabetes mellitus
221
not to various populations of antibodies. In contrast l~osselin et al. [12] studying adult diabetics and using human insulin as tracer found binding capacities for human insulin similar to those for porcine and bovine hormone. From our results it seems that antibodies induced in diabetic children do not strongly bind human insulin. Hiirter and Kiihnau [8] found comparable binding for porcine and bovine insulin in one third of their patients. In 53% bovine insulin capacity was higher, in 14% porcine. Meyer [9] and Murty et al. [10] also found more patients with higher binding for bovine than porcine insulin than the opposite. In the present study, only 17% of the patients had much higher (at least twice as high) binding of bovine insulin, in 14% the relation was reversed. Murty et al. [10] found no correlation between duration of insulin treatment and level of antibody capacity. In contrast Hiirter and Kiihnau [8] showed a significant rise in porcine and bovine insulin binding capacity after more than 4 years of treatment. The findings of Ortved Andersen also show a slow rise over 10 years of insulin application. In the present study some patients with only short duration of treatment had very high binding capacity of porcine insulin whilst binding of bovine insulin seems rather independent of duration of treatment. Ortved Andersen [11] demonstrated a maximal binding for porcine insulin in the first 2 years of treatment followed by a subsequent slow decrease. Several authors [3, 8, 10, 11] found a positive correlation between insulin requirement and insulin-binding capacity. Murty et al. [10] were unable to confirm this finding. In our patients there was no significant correlation between insulin dosage and binding of porcine, bovine, or human insulin. However, high insulin requirement was more often correlated with high binding of bovine insulin. As in the report of ttiirter and Kiihnau [8] there is a correlation between duration of treatment and insulin requirement. On the other hand, some patients with low insulin requirement had a very high binding of porcine insulin. Ditschuneit and Federlin [3] presumed a relatively high endogenous insulin production in patients with low insulin requirement not influenced by high binding of exogenous hormone. In contrast some of our patients with low insulin dosage tended to have measurable binding capacities for human insulin. Ditschuneit and Federlin [3] found no correlation between insulin requirement and calculated association constants which would indicate that the binding capacities measured reflect only antibody concentration and not biological function. The patients in the study of Hfirter and Kiihnau [8] had all received a mixture of bovine and porcine insulin. Waldh/~usl et al. [14] demonstrated lower insulinbinding capacity in patients on bovine insulin alone as compared to a mixture of bovine and porcine insulin. Our patients could be divided into three groups according to insulin preparations used. Patients on bovine insulin alone or on a commercial mixture of porcine and bovine insulin showed lower binding of bovine, porcine, and human insulin than those on individual mixture of porcine and bovine insulin. The second group probably contains patients who have a diabetes which is difficult to control. However, their daily insulin requirements are not significantly higher than in the other groups. The binding of porcine insulin was especially high in this group, the binding of bovine insulin somewhat lower than in the patients on a commercial mixture. The patients with "difficult-to-control"
222
P.W. Nars et al.
diabetes showed a binding of human insulin of 40 ~U/ml or more in 28% as compared to only 6--10~ in the patients on commercial mixtures or on bovine insulin alone. One explanation for the instability os the carbohydrate metabolism in these patients could be high binding of endogenous insulin. The present study and most data published show little and sometimes controversial correlations between antibody capacity and insulin requirements in diabetic patients. The presence of these antibodies seems to influence the action of exogenous and probably also os endogenous insulin. Circulating antibodies can block hormonal activity and lead to elevated requirements. They can also protect the insulin molecule from being destroyed and so prolong its action. Furthermore, the presence of circulating antibodies has been shown to play a role in the pathogenesis of diabetic complications as retinopathy and arteriosclerosis. Recent studies with monocomponent insulin preparations [5, 13] show only little or even no induction of antibodies. Although the biological role of such antibodies is not well understood and few estimations will be of clinical significance for the individual patient the use os such insulin preparations seems indicated.
References 1. Berson, S. A., Yalow, R. S. : Quantitative aspects of the reaction between insulin and insulin-binding antibody. J. clin. Invest. 88, 1996--2016 (1959) 2. Berson, S. A., Yalow, R. S. : Species-specificity of human anti-beef, pork insulin serum. J. clin. Invest. 88, 2017--2025 (1959) 3. Ditschuneit, H., Federlin, K.: Beitrag zur Pathogenese der Insulinresistenz. Dtsch. med. Wschr. 91, 853--859 (1966) 4. Dunlop, M., Court, J. M. : The measurement of total plasma insulin and insulin antibody levels in diabetic children. Clin. chim. Acta 52, 353--359 (1974) 5. Fankhauser, S., Michl, J.: 2. Internationales Donausymposium fiber Diabetes mellitus, Budapest 1971. Separatum (eds. I. Maggyar, A. Beringer). Wien: Verlag der Wiener Medizinischen Akademie 1971 6. Greenwood, F. C., Hunter, W. M., Glover, J. S. : The preparation of IiSi-labelled human growth hormone of high specific activity. Bioehem. J. 89, 114--123 (1963) 7. Herbert, V., Lau, K.-S., Gottlieb, C.W., Bleicher, S. J. : Coated charcoal immunoassay of insulin. J. clin. Endocr. 25, 1375--1381 (1965) 8. Hiirter, P., Kiihnau, J., Jr.: Die Aktivitit zirkulierender InsulinantikSrper bei kindlichen Diabetikern. Helv. paediat. Acta 25, 154---164 (1970) 9. Meyer, H.-W. : ~ber die Bindung yon Rinder- und Schweineinsulin durch zirkulierende AntikSrper im Serum yon Diabetikern. ivied. Welt 19, 1758--1763 (1968) 10. Murthy, D. Y. N., Guthrie, R.A., Womack, W.N., Jackson, R.L.: Insulin binding in children with diabetes mellitus. Pediatrics 48, 558--566 (1969) 11. Ortved Andersen, 0. : Insulin antibody formation. I. The influence of age, sex, infections, insulin dosage and regulation of diabetes. Acta endocr. 71, 126--140 (1972) 12. Rosselin, G., Tehobroutsky, G., Assan, R., Lelloueh, J., Dolais, J., Derot, M.: Etude quantitative d'anticorps humains anti-insulines animales par la m~thode radio-immunologique de Berson et u Diabetologia 1, 33--38 (1965) 13. Schlichtkrull, J., Brange, J., Christiansen, Aa. H., Hallund, 0., Heding, L. G., Jorgensen, K. H.: Clinical aspects of insulinantigenieity. Diabetes 21, Suppl. 2, 649--656 (1972) 14. Waldhiusl, W. K., Frisch, H., Haydl, H. : Vorkommen mid Bedeutung yon Insulinantik6rpern bei Diabetikern und ihre Beeinflussung durch Monocomponent-Insulin. Paper presented on the 8th annual meeting of the European Association for the Study of Diabetes, Madrid, September 6--8, 1972 Dr. P. W. I~lars Kinderspital Basel RSmergasse 8 CH-4005 Basel, Schweiz
Insulin Antibodies in 104 Children with Diabetes mellitus P. W. Nars, G. Herz, and J. Girard Department of Endocrinology (Head: P.D. Dr. J. Girard), University Children's Hospital, Basel (Head: Prof. Dr. G. Stalder) l~eeeived March 1, 1976 Abstract. Binding capacity for porcine, bovine, and human insulin was estimated in 104 diabetic children using homologous systems of iodinated and noniodinated insulin. All patient sera bound porcine and bovine insulin whilst 11.5~o did not bind human insulin. There was no clear correlation between duration of insulin treatment and binding capacity. However, some patients with only short duration of treatment had high binding capacity for porcine insulin. The binding capacity for human insulin was low in most patients. High insulin requirement was on average combined with high binding for bovine insulin. Cases of diabetes, which are difficult to control, treated with individual mixtures of porcine and bovine insulin, often showed high binding for porcine and also for human insulin. The value of estimations of insulin-binding capacity in diabetic children is debatable. However, the high incidence of antibodies seems to justify the use of low immunogenie (monocomponent) insulin preparations. Key words: Diabetes mellitus - - Insulin antibodies. P o l y p e p t i d e h o r m o n e s for t h e r a p e u t i c a l use are g e n e r a l l y p r o d u c e d b y e x t r a c t i o n a n d s u b s e q u e n t purification from a n i m a l or h u m a n glands. I f t h e s t r u c t u r e of p e p t i d e s from o t h e r species differs from t h e h u m a n endogenous hormone, such p r e p a r a t i o n s can be a n t i g e n i c if i n j e c t e d into p a t i e n t s . I n 1956, Berson a n d Y a l o w [1] used insulin l a b e l e d w i t h r a d i o i o d i n e to d e m o n s t r a t e t h e presence o f insulin a n t i b o d i e s in d i a b e t i c s t r e a t e d w i t h insulin. Since this basic w o r k t h e r e h a v e been several r e p o r t s on insulin a n t i b o d i e s in d i a b e t i c a d u l t s b u t o n l y few studies in d i a b e t i c children [8, 10, 11]. J u v e n i l e d i a b e t e s is g e n e r a l l y t r e a t e d w i t h insulin injections once or twice daily, w h e r e b y porcine, bovine, or a c o m b i n a t i o n o f porcine a n d b o v i n e insulin is used. T h e aim o f t h e p r e s e n t s t u d y was t h e a s s e s s m e n t o f b i n d i n g c a p a c i t y for porcine, bovine, a n d h u m a n insulin in s e r u m of 104 d i a b e t i c children on insulin treatment. M a t e r i a l and Methods One hundred and four juvenile diabetic patients aged 7--20 years (mean 13 years) were studied. Their diabetes mellitus had been known and treated for 6 months--17 years (mean 5 years). All patients injected insulin once or twice daily, and in all children the diabetic state was well controlled. Thirty two patients used an insulin preparation of bovine origin, 72 injected either a commercial mixture of porcine and bovine origin or prepared their mixture themselves according to varying insulin requirements. Blood was withdrawn from an anteeubital vein before insulin injection, generally in the early morning. The blood samples were allowed to clot at room temperature, the serum was separated by centrifugation, and stored a t - - 2 0 ~ until analysis. The sera were investigated for insulin antibodies using a modification of the method described by Dunlop and Court [4]. Porcine (Aetrapid Novo), bovine (Insulin Lente Novo),
218
P . W . Nars et al.
Table 1. Method used for the estimation of binding capacity for porcine, bovine, and human insulin in sera of patients with diabetes mellitus on insulin treatment 0.1 ml Incubation medium containing 0.2 ~U of porcine, bovine, or human I125-insulin 0.1 ml Incubation medium containing 0---1000 t~U of corresponding standard insulin 0.1 ml Patient serum diluted in 5.9% human serum albumin Incubation for 16 h at H-4~ Charcoal separation (0.5 ml 0,5% chareoal--0.05~o dextran 80).
and human (batch 12569 mC, a gift from Dr. L. Heding, Novo A/S) insulin were labeled with iodine 125 according to the method of Greenwood et al. [6]. The calculated specific activity varied between 90 and 110 tLCi/ttg. All incubation studies were done overnight at +4~ in a phosphate buffer containing human serum albumin. Free and antibody-bound hormone were separated by adsorption to dextran-coated charcoal [7]. Initially serum dilutions were incubated with 0.2 ~tU of the tracer preparations. For the assessment of binding capacity, patient sera were diluted to yield a 30% binding of the tracer. The amount of noniodinated hormone required to give a significant displacement of the radioactive tracer (defined as a 5% drop in binding percents) was determined by the addition of varying concentrations of nonlabeled insulin. All assays were done in duplicate. The method is summarized in Table 1. Statistics were done using Student's t test and the Z~ test.
Results All p a t i e n t sera w i t h o u t e x c e p t i o n were f o u n d t o b i n d a significant a m o u n t of 0.2 ~ U o f i o d i n a t e d porcine a n d b o v i n e insulin. Twelve o u t of 104 sera (11.5~ d i d n o t b i n d i o d i n a t e d h u m a n insulin. T h e r e was a wide s c a t t e r in t h e a m o u n t o f u n l a b e l e d insulin r e q u i r e d for a significant d i s p l a c e m e n t of t h e l a b e l e d h o r m o n e (range t e s t e d : 1 - - 1 0 0 0 ~U). T h e b i n d i n g capacities for porcine, bovine, a n d h u m a n insulin in t h e 104 sera are s u m m a r i z e d in T a b l e 2. A l t h o u g h m o s t sera b o u n d i o d i n a t e d h u m a n insulin, o n l y 18 (17.8%) h a d a b i n d i n g c a p a c i t y of 40 txU/ml or more. I n m o s t p a t i e n t s t h e b i n d i n g capacities for porcine a n d b o v i n e insulin were of t h e s a m e m a g n i t u d e ; 18 sera showed m u c h h i g h e r b i n d i n g o f porcine insulin, in 15 t h e r e l a t i o n was reversed. T h e c o r r e l a t i o n b e t w e e n d u r a t i o n o f insulin t r e a t m e n t a n d a n t i b o d y - b i n d i n g c a p a c i t y is shown in F i g u r e 1. Seven o u t of 18 p a t i e n t s w i t h o n l y 1 y e a r of t r e a t -
Table 2. Binding capacity for porcine, bovine, and human insulin in sera of 104 diabetic children on insulin treatment Insulin-binding Number of patients capacity ([zU/ml) < 40 40 < 100 100< 1000 > 1000
35 36 19 i4
: : : =
34% 35% 18% i3%
porcine
37 38 19 10
: = = =
36% 37% 18% i0%
bovine
86 = 83% 16 = 15% 2 : 2% 0 human
Insulin Antibodies in 104 Children with Diabetes mellitus
105
porcine 9,
219
human
bovine
9
104 Q
Q*
Z)
:= 103
B
9 1 7 6 1 4 9 99
~
,~176149 ~
I
D
I Q
9
9
~149
t02 40
0
0
Q
9
.~149
9
~ ~ 1 4 9 1 4 99 1 4 9
:*%,~149
9
000
~ 9
,
9
9
9
~
9176
;9
. . . .
F'""
~ 1 7 6 1 4 9 1 4 9o1 4~9
~
'
10
,
15y
i
1'o
15y
lb 'T'sy
Fig. 1. Individual values of binding capacity for porcine, bovine, and human insulin compared to duration of insulin treatment in 104 diabetic children. Note logarithmic scale ment had a high binding ( > 1000 FU/ml) for porcine insulin. Compared to those with longer duration of treatment this difference is statistically significant (P 0.01). From the second year of treatment on, there is a fall in mean binding of porcine insulin. The mean binding of bovine insulin seems independent of duration of treatment. Because of the high percentage of sera binding less than 40 FU/ml of human insulin, no mean values could be calculated. The relative number of sera with measurable binding was independent of duration of treatment. The daily insulin requirement in the patients studied varied greatly, although a positive correlation between insulin dosage in U/m 2 body surface and duration of treatment could be calculated (P 0.01). For a correlation between insulin requirement and binding capacity the patients were divided into three groups: (1) insulin requirement less than 18 U/mS/day, (2) 18--29 U/mS/day, and (3) more than 29 U/m2/day. Only the third group of patients requires insulin dosages clearly above normal endogenous production [8]. These results are summarized in Table 3. The patients in group 1 had a significant later onset of their diabetes and also of duration of treatment, whilst the mean chronologic age of the three groups showed Table 3. Comparison between insulin requirement and porcine, bovine, and human insulin binding capacity in serum. Mean and standard deviation Insulin requirement (U/m2/day) 1000 ~U/ml (mean binding capacity ~U/ml)
10.1•
3.1~2.9
6=22% (6270) 5 = 10% (1740) 3 = 12% (990) porcine
18 30 (n = 25)
7.8 -4- 4.0
6.0 :L 4.0
1---- 4% (140) 6 = 12% (1010) 3 = 12% (570) bovine
Patients withbinding capacity _->40 FU/ml 6=22% 10 = 19% 2= human
8%
220
P.W. Nars et al.
Table 4. Comparison between type of insulin used for treatment of diabetes mellitus and porcine, bovine, and human insulin-binding capacity in serum Insulin therapy
Mean dosage (U/m~/d)
Duration (years)
Patients with binding Patients with capacity > 1000 ~U/ml binding capacity (mean binding capacity __>40 f~U/ml ~U/ml)
Bovine (n~--32)
22.8
6.0
3 = 9% (730)
2 = 6% (320)
2=
20.9
3.8 4.7
3 ~- 10% (12so) 5 ~ 12% (560) bovine
3 = 10%
25.0
2 = 7% (900) 9 ~ 21% (4630) porcine
Bovine Jr porcine: Commercial mixture (~ = 29) Mixed individually (n -~ 43)
6%
12 ~ 28~o human
no difference. With greater insulin requirement there was on average a rise in binding capacity for bovine insulin, for porcine insulin the results were the reverse. The relative number of patients with a binding capacity of more than 1000 ~U/ml shows the same picture. Patients with high insulin requirement had relatively less frequently a measurable binding capacity for h u m a n insulin. The data on type of insulin used for treatment and binding capacity are summarized in Table 4. The number of patients with high binding capacities for porcine insulin was elevated in those using an individual mixture of bovine and porcine insulin, as was the mean binding capacity. Patients on bovine insulin alone showed somewhat lower binding capacities for bovine insulin as those on bovine plus porcine insulin. The same was the case as to binding of h u m a n insulin. All these figures failed to reach statistical significance.
Discussion As in the study of Murty et al. [10] there was a significant binding of iodinated porcine and bovine insulin in all sera of juvenile diabetics on insulin treatment for at least 6 months. I n contrast, Hiirter and Kiihnau [8] and Ortved Andersen [11] found no binding in 9% of their patients. Incorporation of iodine into the insulin molecule might lower immunoreactivity. Berson and Yalow [2] found no impairment with an incorporation of 10 ~0 ~Ci per ~g of hormone. The porcine, bovine, and h u m a n tracer preparations used in this study had a specific activity of 90--110 ~Ci/~g. High binding to guinea pig antiinsulin sera indicates intact immunoreaetivity of all tracer hormones. The physical characteristics of the h u m a n tracer preparation used were equal to the animal ones. Still 11.5% of the patients showed no significant binding of h u m a n insulin. Most of the sera binding iodinated h u m a n insulin had a binding capacity of less than 40 ~U/ml. Berson and Yalow [2] demonstrated t h a t sera from diabetics bind h u m a n insulin much less strongly t h a n porcine or bovine insulin. According to these authors the difference in binding characteristics is due to energetically different reactions with the same antibody combining sites and
Insulin Antibodies in 104 Children with Diabetes mellitus
221
not to various populations of antibodies. In contrast l~osselin et al. [12] studying adult diabetics and using human insulin as tracer found binding capacities for human insulin similar to those for porcine and bovine hormone. From our results it seems that antibodies induced in diabetic children do not strongly bind human insulin. Hiirter and Kiihnau [8] found comparable binding for porcine and bovine insulin in one third of their patients. In 53% bovine insulin capacity was higher, in 14% porcine. Meyer [9] and Murty et al. [10] also found more patients with higher binding for bovine than porcine insulin than the opposite. In the present study, only 17% of the patients had much higher (at least twice as high) binding of bovine insulin, in 14% the relation was reversed. Murty et al. [10] found no correlation between duration of insulin treatment and level of antibody capacity. In contrast Hiirter and Kiihnau [8] showed a significant rise in porcine and bovine insulin binding capacity after more than 4 years of treatment. The findings of Ortved Andersen also show a slow rise over 10 years of insulin application. In the present study some patients with only short duration of treatment had very high binding capacity of porcine insulin whilst binding of bovine insulin seems rather independent of duration of treatment. Ortved Andersen [11] demonstrated a maximal binding for porcine insulin in the first 2 years of treatment followed by a subsequent slow decrease. Several authors [3, 8, 10, 11] found a positive correlation between insulin requirement and insulin-binding capacity. Murty et al. [10] were unable to confirm this finding. In our patients there was no significant correlation between insulin dosage and binding of porcine, bovine, or human insulin. However, high insulin requirement was more often correlated with high binding of bovine insulin. As in the report of ttiirter and Kiihnau [8] there is a correlation between duration of treatment and insulin requirement. On the other hand, some patients with low insulin requirement had a very high binding of porcine insulin. Ditschuneit and Federlin [3] presumed a relatively high endogenous insulin production in patients with low insulin requirement not influenced by high binding of exogenous hormone. In contrast some of our patients with low insulin dosage tended to have measurable binding capacities for human insulin. Ditschuneit and Federlin [3] found no correlation between insulin requirement and calculated association constants which would indicate that the binding capacities measured reflect only antibody concentration and not biological function. The patients in the study of Hfirter and Kiihnau [8] had all received a mixture of bovine and porcine insulin. Waldh/~usl et al. [14] demonstrated lower insulinbinding capacity in patients on bovine insulin alone as compared to a mixture of bovine and porcine insulin. Our patients could be divided into three groups according to insulin preparations used. Patients on bovine insulin alone or on a commercial mixture of porcine and bovine insulin showed lower binding of bovine, porcine, and human insulin than those on individual mixture of porcine and bovine insulin. The second group probably contains patients who have a diabetes which is difficult to control. However, their daily insulin requirements are not significantly higher than in the other groups. The binding of porcine insulin was especially high in this group, the binding of bovine insulin somewhat lower than in the patients on a commercial mixture. The patients with "difficult-to-control"
222
P.W. Nars et al.
diabetes showed a binding of human insulin of 40 ~U/ml or more in 28% as compared to only 6--10~ in the patients on commercial mixtures or on bovine insulin alone. One explanation for the instability os the carbohydrate metabolism in these patients could be high binding of endogenous insulin. The present study and most data published show little and sometimes controversial correlations between antibody capacity and insulin requirements in diabetic patients. The presence of these antibodies seems to influence the action of exogenous and probably also os endogenous insulin. Circulating antibodies can block hormonal activity and lead to elevated requirements. They can also protect the insulin molecule from being destroyed and so prolong its action. Furthermore, the presence of circulating antibodies has been shown to play a role in the pathogenesis of diabetic complications as retinopathy and arteriosclerosis. Recent studies with monocomponent insulin preparations [5, 13] show only little or even no induction of antibodies. Although the biological role of such antibodies is not well understood and few estimations will be of clinical significance for the individual patient the use os such insulin preparations seems indicated.
References 1. Berson, S. A., Yalow, R. S. : Quantitative aspects of the reaction between insulin and insulin-binding antibody. J. clin. Invest. 88, 1996--2016 (1959) 2. Berson, S. A., Yalow, R. S. : Species-specificity of human anti-beef, pork insulin serum. J. clin. Invest. 88, 2017--2025 (1959) 3. Ditschuneit, H., Federlin, K.: Beitrag zur Pathogenese der Insulinresistenz. Dtsch. med. Wschr. 91, 853--859 (1966) 4. Dunlop, M., Court, J. M. : The measurement of total plasma insulin and insulin antibody levels in diabetic children. Clin. chim. Acta 52, 353--359 (1974) 5. Fankhauser, S., Michl, J.: 2. Internationales Donausymposium fiber Diabetes mellitus, Budapest 1971. Separatum (eds. I. Maggyar, A. Beringer). Wien: Verlag der Wiener Medizinischen Akademie 1971 6. Greenwood, F. C., Hunter, W. M., Glover, J. S. : The preparation of IiSi-labelled human growth hormone of high specific activity. Bioehem. J. 89, 114--123 (1963) 7. Herbert, V., Lau, K.-S., Gottlieb, C.W., Bleicher, S. J. : Coated charcoal immunoassay of insulin. J. clin. Endocr. 25, 1375--1381 (1965) 8. Hiirter, P., Kiihnau, J., Jr.: Die Aktivitit zirkulierender InsulinantikSrper bei kindlichen Diabetikern. Helv. paediat. Acta 25, 154---164 (1970) 9. Meyer, H.-W. : ~ber die Bindung yon Rinder- und Schweineinsulin durch zirkulierende AntikSrper im Serum yon Diabetikern. ivied. Welt 19, 1758--1763 (1968) 10. Murthy, D. Y. N., Guthrie, R.A., Womack, W.N., Jackson, R.L.: Insulin binding in children with diabetes mellitus. Pediatrics 48, 558--566 (1969) 11. Ortved Andersen, 0. : Insulin antibody formation. I. The influence of age, sex, infections, insulin dosage and regulation of diabetes. Acta endocr. 71, 126--140 (1972) 12. Rosselin, G., Tehobroutsky, G., Assan, R., Lelloueh, J., Dolais, J., Derot, M.: Etude quantitative d'anticorps humains anti-insulines animales par la m~thode radio-immunologique de Berson et u Diabetologia 1, 33--38 (1965) 13. Schlichtkrull, J., Brange, J., Christiansen, Aa. H., Hallund, 0., Heding, L. G., Jorgensen, K. H.: Clinical aspects of insulinantigenieity. Diabetes 21, Suppl. 2, 649--656 (1972) 14. Waldhiusl, W. K., Frisch, H., Haydl, H. : Vorkommen mid Bedeutung yon Insulinantik6rpern bei Diabetikern und ihre Beeinflussung durch Monocomponent-Insulin. Paper presented on the 8th annual meeting of the European Association for the Study of Diabetes, Madrid, September 6--8, 1972 Dr. P. W. I~lars Kinderspital Basel RSmergasse 8 CH-4005 Basel, Schweiz
