Abstract. Female pet dogs exhibiting either glucose intolerance alone or glucose intolerance and acromegaly were investigated. Some dogs developed the disorder(s) during dioestrus and some animals developed the disorder(s) after they were given medroxyprogesterone acetate (MPA). Elevated fasting plasma glucose levels (12.3 ± 1.9 mm, mean ± sem) were accompanied by fasting hyperinsulinaemia (144 ± 21 μU/ml, mean ± sem) and drastic elevation of plasma growth hormone (GH) levels (112.6 ± 45 ng/ml, mean ± sem). An iv glucose tolerance test (IVGTT) performed on all dogs revealed non-suppressibility of GH levels and glucose intolerance. Plasma concentrations of glucose, insulin and GH during IVGTT in affected dogs differed significantly from the concentrations measured in normal dogs during the same test. MPA withdrawal and/or ovariohysterectomy (OVx-HYx) in affected animals was followed by reversal of GH levels to normal and improved glucose tolerance. Acromegaly associated soft tissue changes were also reversible after MPA withdrawal and/or OVx-HYx when GH levels had dropped. In 5 dogs which had developed diabetes during dioestrus and in which a spontaneous decrease in plasma progesterone occurred during the investigation a concomittant decrease in GH levels was observed. Plasma GH measured at different stages of pregnancy in 45 dogs was found to be elevated in one animal only. The results show that the development of spontaneous diabetes/acromegaly occurring in some female dogs is related to progestagen (progesterone/MPA) exposure and that reversal of the signs is achieved by progesterone/MPA withdrawal. The results suggest that diabetes/acromegaly in the dogs studied was caused by progesterone/MPA-evoked GH elevation. Finally, the findings also suggest that the GH axis normally not appreciably responsive to progestagen exposure in some dogs becomes and/or is paradoxically controlled by physiologic levels of endogenous progesterone or low doses of MPA.

The relationships between body size, growth hormone (GH) secretory capacity and circulating insulinlike growth factor I (IGF I) levels were studied in genetically-determined subgroups of disparate size within one breed of dogs, the Poodle. Standard (large) Poodles exhibited six times the mean plasma IGF I concentration found in Toy Poodles. The mean IGF I level found in Standard Poodles significantly differed from the one found in Miniature and Toy Poodles (P < 0.001). The correlation between circulating IGF I levels and body size was found to be highly significant (P < 0.001; r = 0.88). All dogs secreted similar, normal amounts of GH in response to clonidine administration.The results show that body weight is correlated with IGF I levels rather than with the GH secretory capacity, thus providing indirect evidence for IGF I as an important in vivo growth-promoting principle.

A sensitive radioimmunoassay (RIA) for canine growth hormone (GH) was developed. Antibodies were elicited in rhesus monkeys. One antiserum exhi¬ bited a working titer at a dilution of 1:500000. Radioiodination was performed enzymatically employing lactoperoxidase. Logit-log transformation and least squares fitting resulted in straight line fitting of the standard curve between 0.39 and 50 ng/ml. Formation of largemolecular [12SI]GH during storage caused diminished assay sensitivity. Therefore [125I]GH was re-purified by gel chromatography. Using this procedure, high and reproducible assay sensitivity was obtained. Tracer pre¬ parations were used for as long as 3 months after iodination. Diluted plasma from normal and acromegalic dogs resulted in a dose-response curve parallel to the standard curve. Canine prolactin exhibited a cross-reac¬ tivity of 2%. The within-assay coefficient of variation (CV) was 3.8 and the between-assay CV was 7.2%. Mean plasma GH concentration in normal dogs was 1.92 \ m=+-\ 0.14 ng/ml (mean \ m=+-\sem). GH levels in acromegalic dogs were appreciably higher. Insulin-induced hypoglycaemia, arginine and ornithine administration resulted in inconsistent and sluggish GH increment. A better re¬ sponse was obtained by injecting a low dose of clonidine. Clonidine administration to hypopituitary dogs resulted in absent or poor GH increment.

Plasma insulin-like growth factor I concentrations from proportionate, chondrodystrophic and giant breeds were evaluated and compared with body size. IGF-I plasma concentrations were 91.2 \m=+-\10.9

A radioimmunoassay (RIA) devised for the measurement of human insulin-like growth factor I (IGF I) was employed for the measurement of canine IGF I. Canine IGF I was extracted from plasma specimens by gel chromatography. Columns were eluted with 1 m acetic acid and the fractions representing the 55 to 85% bed volume were pooled, lyophilized and reconstituted with assay buffer. Serial dilutions of canine IGF I from both normal and acromegalic dogs when added to the RIA system gave a similar displacement pattern of human [125I]IGF I as the one obtained by the addition of unlabelled human IGF I. The dose-response curve obtained by canine IGF I paralleled the one obtained by human IGF I. Logit-log transformation and least squares fitting resulted in straight line fitting of the standard curve between 0.039 and 5 ng IGF I added per tube. The within-assay coefficient of variation (CV) was 16.7% and the between-assay CV was 21.8%. Plasma IGF I concentrations in normal dogs appeared to be a function of body size. The concentrations were 36 \ m=+-\ 27 ng/ml in Cocker Spaniels, 87 \ m=+-\33 ng/ml in Beagles, 117 \m=+-\ 34 ng/ml in Keeshonds, and 280 \m=+-\23 ng/ml in German Shepherds (mean \ m=+-\SEM). The mean IGF I level in a group of dogs with growth hormone (GH) elevation was 700 \ m=+-\ 90 ng/ml. Though this group of dogs comprised both small and large dogs, the mean IGF I level significantly differed from the one found in German Shepherds, the largest breed studied (P < 0.01). IGF I levels in dogs with GH elevation were similarly elevated in both dogs exhibiting acromegaly and dogs exhibiting GH\ x=req-\ diabetes, but no signs of acromegaly. In dogs with GH elevation, drop in GH levels was associated with a significant drop in IGF I levels.