No abstract

SynopsisBilateral ligation of both the submandibular and parotid ducts of adult normal and mutant hyperinsulinemic diabetic mice resulted in a significant hypoglycemic effect. Therefore, we postulated that duct ligation may result in the removal of hyperglycemic factor (Hoshino et al., 1976) rather than a change in insulin sensitivity.Indeed, no change in specific binding of ,25I-insulin was observed in membrane fractions from several tissues obtained from mice of either sex or strains before and after duct ligation. After slices of the submandibular gland were incubated for4hr in Eagle's medium, an aliquot of the culture medium was injected i. p. into normal adult mice. A significant hyperglycemic effect was observed in30min in the injected animals.Eluates obtained by gel filtration of the crude extract of the submandibular gland were injected into normal adult mice, and hyperglycemia ensued. Thus, it is postulated that ligation of salivary ducts results in glandular atrophy and disappearance of the hyperglycemic factor which in turn leads to hypoglycemia and amelioration of diabetes mellitus, particularly of hyperinsulinemic type.

Utilizing an acid gel chromatography and insulin radioreceptor assay (RRA), serum levels of receptor assayable insulin-like activities were measured under various conditions. Acid gel filtration of sera on a Sephadex G-50 was adopted to separate small molecular ILAs from binding proteins before the assay by RRA. By employing 125I-pork insulin as the tracer, and pork insulin as the standard, an RRA for insulin was developed, in which kidneys of sacrificed pregnant guinea pigs were used as the source of the solubilized receptor. After gel-filtration of the sera, pooled fractions, which grossly corresponded to those of 125I-insulin marker, were assayed by RRA. The subjects consisted of fifty-nine cases: normal control subjects (n = 19), active acromegaly (6), Sheehan's syndrome (5), liver cirrhosis (7), chronic renal failure (10), non-insulin dependent diabetes mellitus (6), overt hyperthyroidism (5) and Nelson's syndrome (1). The average receptor assayable ILA of the normal control subjects was 40.2 +/- 12.2 ng/ml. As insulin RRA has a big interassay variation, receptor assayable ILA-ratio was used to minimize the variation, and each data was shown as the ratio to the average ILA of the normal controls. By this method, sera from normal adults had a mean (+/- SD) receptor assayable ILA ratio of 1.00 +/- 0.28. Four out of six cases of acromegaly revealed significantly high concentrations, and the average receptor assayable ILA-ratio of acromegaly was 1.30 +/- 0.28 (mean +/- SD, p less than 0.015). In the cases of Sheehan's syndrome, the ILA-ratio was 0.30 +/- 0.12, which was significantly low (p less than 0.001). Therefore, GH dependency was suspected from these two factors. However, the direct correlation was not indicated between GH and receptor assayable ILA. It was also considered that receptor assayable ILA was influenced not only by GH but also by some other factors. Furthermore, the subjects with liver cirrhosis indicated the low levels of receptor assayable ILA-ratio of 0.46 +/- 0.31, while the subjects with chronic renal failure showed the high ILA-ratio of 1.59 +/- 0.45 (p less than 0.05). No differences in ILA-ratio were found in the subjects with diabetes mellitus, hyperthyroidism and Nelson's syndrome, compared to the normal subjects.