The Prevalence of Proinsulin-Specific Antibodies in Diabetic Patients

Abstract

To establish the prevalence of proinsulin-specific antibodies in diabetic patients, thirty serum sampies were analyzed.Among the insulin-treated group, proinsulin antibodies were demonstrated in 20 cases (80%) by the significant binding of 1251-bovine proinsulin to the sera pretreated with excess of insulin and reduced binding when these sera were preincubated with unlabelled bovine proinsulin. Five other sera (20%) bound tracer pro insulin, but did not bind significantly after preincubation with insulin indicating tbat only cross-reacting insulin antibodies were present and there were no detectable proinsulin antibodies. Sera from five diet-controlled patients bad no antibodies. IntroduetionComrnercia! preparations of insulin contain variable amounts of proinsulin and related substances (Steiner, HaI/und, Rubenstein, Cho and Bayliss 1968, Chance 1972) which have been shown to have significant immunogenicity in anima! experiments (Rubenstein, Steiner, Lawrence and Kirsteins 1969, Kerp 1971, Schlichtkrull, Brange, Christiansen, Hallund, Heding and Jorgensen 1972. I t is very likely that the daily administration of proinsulin, present in therapeutic insulin will induce proinsulin-specific antibodies in diabetie patients. Although a small concentration of such antibodies have been demonstrated in a pooled human sera (Kerp, Steinhilber and Kasemir 1970), their ineidence in the individual patients remains uninvestigated.The purpose of this report is to establish the prevalence of proinsulin-specific antibodies in diabetic patients. Materials and MethodsFisher Endocrinology Laboratories, Chicago) were purified by gel filtration (Sephadex G-50, 0.5 x 20 in. column, borate buffer pH 7.4) and diluted to 0.2 ng/mL The same preparations of labelIed tracers were used for alI the tests.80vine proinsulin (lot no. 615-10708-212-2) and insulin (lot no. 615-10828-244) were supplied by Dr. R. Chance, Lilly Research Laboratories, Indianapolis. These proteins were dissolved in borate buffer, pH 7.4, containing 1% bovine serum albumin (Mann Laboratories, New York) and diluted to 500 ng/mL Anti-human immunoglobulin serum (AHI-RS) was prepared by hyperimmunizing rabbits to purified plasma Cohn fraction 11. This sera reacted strongly with both the antigenie types K and A light chains and was therefore a polyvalent anti-immunoglobulin (Feinstein, Rapaport and Chong 1969). A seven-fold concentration of this sera produced maximal precipitation of immunoglobulins.Detection of proinsulin and insulin antibodies: Two sets of tubes, each containing 0.1 ml of diluted (1: 16) serum were preincubated for 20 hr at 4 0 C with 0.1 ml of buffer, insulin (500 ng/mq or proinsulin (500 ng/ml). To one set of tubes, 0.1 ml of I 51-bovine proinsulin (0.2 ng/ml) and to the other set, 0.1 ml of 12SI_bovine insulin (0.2 !Jg/ml) were added. Tubes were reincubated for 40 hr at 4 0 C. To precipitate the immune complexes, 0.1 ml of AHI-RS 0:2) was then added. Forty hours later precipitate was filtered and washed on oxoid membrane (Amersham Searle). The radi...

Decrease of antibodies to insulin, proinsulin and contaminating hormones after changing treatment from conventional beef to purified pork insulin

“…Beef and pork proinsulins (provided by Nov 9 Laboratories Ltd.) were labelled with 125I by means of solid-phase lactoperoxidase [3] and purified on cellulose columns (Whatman CF11) [4]. In the determination of proinsulin binding insulin-binding antibody was blocked by including highly purified pork insulin in the incubation buffer at a concentration of 6.7 U/ml [5].…”

Characterization of circulating insulin and proinsulin-binding antibodies in autoimmune hypoglycemia.

“…The se(quences of the C-segments of the precursor molecules differ from each other and from their human counterpart to a much larger extent than the insulin A and B chains (18,21,22) and have been shown to be potent antigens in both experimental animals (23)(24)(25)(26) and humans (27)(28)(29)(30). A survey of insulin-treated diabetic patients demonstrated specific proinsulin antibodies in 80% of the cases (29). Our results have confirmed and extended this finding that C-segmentdirected proinsulin antibodies were detected in the sera of 100 and 90% of the insulin-treated diabetics tested with bovine and porcine proinsulins, respectively.…”

“…Commercial preparations of mixtures of bovineporcine insulin have small and variable amounts of contaminating bovine and porcine proinsulin (17)(18)(19)(20). The se(quences of the C-segments of the precursor molecules differ from each other and from their human counterpart to a much larger extent than the insulin A and B chains (18,21,22) and have been shown to be potent antigens in both experimental animals (23)(24)(25)(26) and humans (27)(28)(29)(30). A survey of insulin-treated diabetic patients demonstrated specific proinsulin antibodies in 80% of the cases (29).…”

Proinsulin and A-component antibodies in diabetics after long-term monocomponent insulin treatment

“…The immunogenicity of insulin preparations cannot be completely assessed by determination of insulin binding IgG alone, as patients treated with conventional (recrystallized) insulin develop specific antibodies against bovine and porcine proinsulin [1], and a-component [2], as well as antibodies against insulin [3]. This is due to the presence of non-insulin-like immunoreactive determinants in the contaminating protein substances e.g.…”