Search

Interleukin-6 (IL-6), a pleiotropic cytokine, is postulated to be involved in the pathogenesis of sick euthyroid syndrome, although the direct in vitro effects of IL-6 on human thyroid function are controversial. Because IL-6 signal can be transduced when the complex of IL-6 and soluble IL-6 receptor (sIL-6R) binds to gp 130, an IL-6 signal transducer, we studied the effects of IL-6 and sIL-6R on thyroid function, using human thyroid follicles obtained from patients with Graves' disease. IL-6 alone had no inhibitory effect on TSH-induced thyroid function (125I incorporation and organic 125I release), even at supraphysiological concentrations. However, in the presence of physiological concentrations of sIL-6R (100 ng/ml), IL-6 inhibited thyroid function dose dependently and completely, accompanied with the decreased ratio of 125I-T3/125I-T4 not only in the thyroid follicles but also in the culture medium. Thyroid follicles did not secrete sIL-6R but produced IL-6 constitutively. Consistent with these findings, sIL-6R inhibited thyroid function slightly at high concentrations. Furthermore, RT-PCR analyses revealed that human thyroid follicles expressed the messenger RNAs for IL-6 and gp130 but scarcely messenger RNA for IL-6R. These in vitro findings suggest that IL-6 alone hardly affects thyroid function in thyroid follicles in which IL-6R gene is scarcely expressed. However, because sIL-6R is present abundantly in serum, IL-6 in vivo would be capable of inhibiting the synthesis and release of T4 and, to a greater extent, T3 from the thyroid gland. These in vitro findings are at least partly related to the development of sick euthyroid syndrome.

Arai M, Tsushima T, Isozaki 0, Demura H, Shizume K, Emoto N, Miyakawa M, Nozoe Y, Murakami H, Ohmura E. Effects of transforming growth factor α (TGF-α) on DNA synthesis and thyrotropin-induced iodine metabolism in cultured porcine thyroid cells. Eur J Endocrinol 1995;132:242–8. ISSN 0804–4643 Transforming growth factor α (TGF-α) is a potent mitogen that is similar structurally to epidermal growth factor (EGF). As EGF is a potent growth stimulator and an inhibitor of iodine metabolism in cultured thyroid cells of several species, we studied whether TGF-α has similar effects using porcine thyroid cells in culture. Recombinant human TGF-α dose-dependently stimulated DNA synthesis of thyroid cells, with maximal stimulation (eight- to ninefold above basal) occurring at 2 nmol/l. The potency was approximately 50% that of mouse EGF and correlated with the ability to compete with EGF for receptor binding, suggesting that the action of TGF-α is mediated by interaction with EGF receptors. When thyroid cells were cultured for 3 days with thyrotropin (TSH) in the presence of TGF-α, TSH-induced iodide uptake was inhibited in a dose-dependent manner. The potency of TGF-α again was approximately 50% that of EGF. Transforming growth factor α did not inhibit TSH-stimulated cAMP production. Moreover, iodide uptake stimulated by either forskolin or 8-bromo-cAMP also was inhibited by TGF-α. Thus, we conclude that TGF-α inhibits TSH-induced iodine metabolism largely by acting at the steps distal to cAMP production. Northern blot analysis revealed expression of TGF-α mRNA in porcine thyroid cells. These observations suggest that TGF-α acts as an autocrine modulator of growth and differentiated functions in porcine thyroid cells. T Tsushima, Department of Medicine 2, Tokyo Women's Medical College, Kawadacho 8–1, Shinjukuku, Tokyo 162, Japan

Thirty-nine children with Noonan syndrome were treated with growth hormone (GH). Prepubertal height velocity was 4.25 ± 1.31 cm/year (M ± SD) (n=38) before GH therapy, which was increased to 6.23 ± 0.87 cm/year (n=33) in the first year, to 4.92 ± 1.15 cm/year (n=22) in the second year and 4.76 ± 1.25 cm/year (n=13) in the third year. Height velocity SDS for chronological age and height velocity (SDS) for bone age were also improved in the 1st year, but declined progressively during the subsequent years. No adverse effects were observed during GH treatment.These data indicate that GH may be effective in Noonan syndrome, but further studies are required to evaluate the efficacy of long-term GH therapy on adult height and body proportions.

To investigate the role of basic fibroblast growth factor (FGF) in renal cell carcinoma growth, we have analyzed the expression of mRNA of basic FGF. In 7 of 15 cases, basic FGF mRNA level in renal cell carcinoma tissues was higher than that in corresponding normal tissues. However, the tumor-to-normal ratios of expression levels are chiefly less than 2.0 and, in 5 cases, are even less than 1.0. Furthermore, there was no correlation between the ratio and the clinical stage. In protein analysis, we could not find any difference between basic FGF extracted from renal cell carcinomas and that from normal kidney tissues in bioactivity, immunoreactivity, molecular weight and affinity to heparin. On the other hand, anti-basic FGF monoclonal antibody inhibited the growth of a renal cell carcinoma cell line, VMRC-RCW, and this inhibition was reversed by an extraphysiological amount of exogenous basic FGF (100 ng./ml.). These results suggest that basic FGF itself may have no pivotal role in renal cell carcinoma etiology but is involved in the growth of renal cell carcinomas in an autocrine manner.

Fifty-seven children with non-endocrine short stature (NESS) were treated with recombinant human growth hormone (GH) for three years. GH treatment significantly increased height velocity, height SD score (SDS), and height velocity SDS for chronological age (CA). The mean height velocity was increased by GH treatment from 4.4cm/year to 7.2cm/year in the first year, 6.3cm/year in the second year and 5.6cm/year in the third year; the mean height SDS from -3.10 SD to -2.72 SD in the first year, to -2.50 SD in the second year and to -2.41 SD in the third year; and the mean height velocity SDS for CA from -1.72 SD to 1.86 SD in the first year, to 0.78 SD in the second year and to 0.52 SD in the third year. However, evaluated in terms of the height SDS for BA, no improvement in the effect was observed.Twenty-two children developed puberty during the three-year GH treatment: 7 in the first year, 5 in the second year and 10 in the third year. The mean height and age at the onset of puberty in 17 patients were 133.5 (±4.9) cm and 12.31 (±1.13) years in the males, and 126.0 (±7.1) cm and 11.87 (±1.02) years in the females. The height at onset of puberty was shorter than that of normal children. Since it is known that the height at onset of puberty correlates well with the final height, it is probable that the patients who were short at puberty in this study will remain short as adults. The effect of GH treatment in children with NESS should be carefully evaluated by conducting a long-term follow-up.

Semisynthetic tetracyclines (TCNs) are used for the management of malignant pleural effusions as sclerosing agents. However, their precise mechanism of actions are uncertain. In the present study, the mechanism of inhibitory effects of minocycline (MINO) and doxycycline (DOXY), on the accumulation of ascitic fluid induced by mouse fibrosarcoma (Meth-A) cells were investigated using male mice. Meth-A cells inoculated intraperitoneally elicited 2.5–4 ml of bloody ascites 10 days after implantation. The production of ascitic fluid was suppressed in a dose-related manner by daily intraperitoneal injections of MINO or DOXY, whereas vehicle (normal saline with 0.01N HC1) did not exert a significant effect. The inhibitory activity of these two substances was quite similar; one mg/mouse of MINO or DOXY inhibited the accumulation of fluid by 87% and 84%, respectively. The survival rate of Meth-A-bearing mice treated with MINO or DOXY was higher than that of the controls. Macroscopic examination of the peritoneal cavity did not reveal any obvious effects, such as adhesions, in mice treated with either MINO or DOXY. In vitro studies showed that MINO and DOXY suppressed Meth-A cell growth with IC 5 0 s of 5 μM and 8 μM, respectively. Maximal suppression (95%) was achieved at MINO and DOXY concentrations of 25 μM. The above observations suggest that MINO and DOXY inhibit the accumulation of ascites by a direct effect on Meth-A cell growth. Therefore, it appears that TCNs injected into the pleural cavity to manage malignant effusions in man exert their activity, at least in part, by suppressing malignant cell growth.

Interleukin-4 (IL-4) potently inhibits bone resorption by preventing the differentiation of osteoclast precursors to osteoclasts. To elucidate the role of IL-4 in bone formation, we studied the effects of human IL-4 on human osteoblast-like cells obtained from trabecular bone, which showed increased osteocalcin production in response to 1,25-(OH)2D3 in more than 10 passages. IL-4 stimulated the proliferation of osteoblast-like cells in a concentration-dependent manner, showing the minimal and maximal stimulatory effects at 10 pg/ml and 100-1000 pg/ml, respectively. IL-4 also stimulated the expression of alkaline phosphatase mRNA (1.7-fold) and the enzyme activity to the same extent at 10-100 pg/ml. Furthermore, IL-4 stimulated collagen type I mRNA expression in human osteoblast-like cells. The cytokine did not affect osteocalcin production in a short culture period (3 days). These in vitro findings suggest that IL-4, a bone-resorption-inhibitory cytokine produced by activated T cells in bone marrow, may exert an anabolic effect on osteoblast-like cells in trabecular bone through a paracrine mechanism.

Studies have demonstrated the presence of epidermal growth factor receptors in human parathyroid tumors. However, there is little information on the effect of other peptide growth factors on parathyroid cell growth. We therefore studied the interaction of insulin-like growth factor I (IGF-I) with human parathyroid tumor cells. Parathyroid tissues were obtained from 24 patients with primary or secondary hyperparathyroidism. There were 15 solitary adenomas, 5 carcinomas, and 4 hyperplastic tissues. First, the binding of [125I]IGF-I to the crude membrane fractions was studied by competitive inhibition with unlabeled IGF-I. Second, isolated parathyroid cells were cultured with IGF-I and examined for DNA synthesis. The IGF-binding protein (IGFBP) content of tissue homogenates was determined by ligand blot analysis. The binding of [125I]IGF-I to parathyroid membranes was dependent on time, temperature, and pH of the medium. Maximum binding was obtained after incubation for 18 hours at 4 degrees C. Specific binding to parathyroid cancer membranes (mean +/- SE, 10.75 +/- 10.55%/mg protein) was significantly (p0.05) greater than that in adenoma tissues (3.71 +/- 2.11%/mg). The value in hyperplastic tissues (4.78 +/- 2.97%/mg) was not different from that in adenomas. Affinity cross-linking and autoradiography demonstrated the type I IGF receptors. Cultured parathyroid cells responded to IGF-I with increased DNA synthesis. The parathyroid tumor tissues expressed IGFBPs. These results suggest that IGF-I and IGFBPs are involved in the growth regulation of parathyroid tumor cells.

The effect of one year of human growth hormone (hGH) treatment on height velocity, height SDS, height velocity SDS and bone age was studied in 71 children with non-endocrine short stature. At the start of hGH treatment, mean chronological age (CA) was 8.75 yr, mean bone age (BA) 6.63, mean height SDS -3.09SD, mean height velocity 4.4 cm/yr, mean height velocity SDS for CA -1.66SD and mean height velocity SDS for BA was -2.19SD. Subjects were divided into four groups: 22 cases with familial short stature (M 14, F 8). 16 cases with SFD short stature (M 11, F 5), 5 cases with both familial and SFD short stature (M 5), and 28 cases with idiopathic short stature (M 20, F 8). There was no significant difference in height SDS, height velocity, mean peak GH and mean sleep GH before treatment among these four groups. The children received rhGH at a dose of 0.5U/kg/week 6-7 times a week subcutaneously.During hGH treatment, 8 cases developed pubertal sign. Since there was no significant difference in the effect hGH treatment between prepubertal group and pubertal group, both group were analyzed together. After one-year hGH treatment, mean height SDS increased significantly to -2.63SD (+0.44SD), mean height velocity to 7M5 cm/yr (+2.63cm/yr), height velocity SDS for CA to+ L62 SD (+3.84SD) and height velocity SDS for BA to +1.63SD (+ 3.845D). There was no significant difference in these values among four groups. During hGH treatment, plasma IGF- I levels increased significantly, but serum osteocalcin levels remained at the same level. Bone age advanced 1.57 years and height age advanced 1.20 years during one-year hGH treatment. The mean ratio of δ bone age/δ height age was 1.34. The effectiveness judged by δheight velocity ≥2cm was 67.6% and that was 62.0% judged by Aheight velocity SDS≥2.5SD. There was no significant difference in these effectiveness among four groups.Short-term hGH treatment was useful in children with non-endocrine short stature. However, long-term effects remained to be clarified.

Insulin-like growth factor II (IGF-II) in serum and tumor extracts from five patients with non-islet-cell tumor hypoglycemia (NICTH) has been characterized. These tumors contained large quantities of IGF-II (2.4-14.2 micrograms/g tissues). The serum IGF-II levels in four of five patients were a little high and the serum IGF-I levels in five patients were low. The serum IGF-II/IGF-I ratios in these patients ranged from 24.1 to 64.2, and the values were significantly greater than those in normal subjects (1.7-7.1). When the sera were gel-filtered on a Sephacryl S-200 column under neutral conditions, the proportion of the free form of IGF-II was not increased. However, in four of five patients, an abnormal IGF-II-IGF binding protein complex was found. When serum IGF binding proteins (IGFBPs) were analyzed by Western ligand blotting, serum IGFBP-2 increased in these patients. When the tumor extracts and sera were gel-filtered on a Biogel P-60 column under acidic conditions, the majority of IGF-II in these sera was a big form of IGF-II. As compared to authentic IGF-II, insulin receptor reactivities and IGF-II receptor reactivities of tumor extracted IGF-II increased in two of three patients. These data indicate that in patients with NICTH, heterogenous IGF-II is produced in respect of size and bioactivities, and that the characteristics of IGF binding protein are altered. Thus, to find IGF-II producing tumors among extrapancreatic tumors associated with hypoglycemia, the quality of IGF-II as well as the quantity should be studied.

Transforming growth factor-alpha (TGF-alpha) in pleural and peritoneal effusions was assayed by homologous radioimmunoassay (RIA) and radioreceptor assay (RRA) using human placental membrane. Effusions were obtained from 24 patients with and 17 patients without cancer. Most of the effusions were found to contain TGF-alpha by RIA and RRA, but immunoreactive epidermal growth factor (EGF) was not detected. Effusions were chromatographed on Bio-Gel P-60 with several peaks of TGF-alpha activity by both RIA and RRA. A discrepancy in the chromatographic pattern of TGF-alpha between RIA and RRa suggested the existence of EGF-like substances capable of binding to EGF receptors which lack immunoreactivity for EGF or TGF-alpha. The TGF-alpha concentration of the acetic acid-extracted malignant effusions assayed by RRA significantly exceeded the value obtained from benign effusions (17.0 +/- 8.7 vs. 9.2 +/- 6.3 ng/ml; mean +/- SD: p0.02). However, the concentrations obtained by RIA did not differ.

Basic fibroblast growth factor (bFGF) increases DNA synthesis in rat FRTL-5 thyroid cells, as measured by increased incorporation of tritiated thymidine into DNA. We show that this action is associated with the ability of bFGF to increase cytosolic Ca2+ levels and transiently increase c-fos mRNA levels. Other agents that increase c-fos mRNA levels and DNA synthesis in FRTL-5 cells include TSH, insulin, insulin-like growth factor-I, phorbol esters, A23187, and alpha 1-adrenergic agents; the last two agents also act by increasing cytosolic Ca2+ levels. Despite its enhancement of DNA synthesis, however, bFGF decreases TSH-induced cAMP-mediated iodide uptake. This action appears to reflect two separate actions of bFGF. First, bFGF decreases TSH receptor mRNA levels and the ability of TSH to acutely increase cAMP levels in FRTL-5 cells. The ability of bFGF to negatively regulate TSH receptor mRNA levels is additive to and independent of the ability of TSH and its cAMP signal to negatively autoregulate TSH receptor mRNA levels. This is consistent with the effect of bFGF on cytosolic Ca2+ levels and the ability of increased cytosolic Ca2+ to decrease TSH receptor mRNA levels. Second, bFGF inhibits cAMP signal expression, as evidenced by its ability to inhibit (Bu)2cAMP-induced iodide uptake in FRTL-5 cells. Both effects are, presumably, associated with the ability of bFGF to counteract TSH/cAMP-induced increases in thyroid peroxidase mRNA levels, which we demonstrate. We suggest, therefore, that bFGF causes opposite effects on DNA synthesis and iodide uptake because of its effect on cytosolic Ca2+ levels and because increases in cytosolic Ca2+ can have opposite effects on gene transcription, particularly in the case of the TSH receptor and c-fos genes.

We summarized the clinical course of 10 patients with silent thyroiditis and evaluated the clinical usefulness of ultrasonography, in combination with clinical and laboratory findings, for the differentiation from Graves' disease. Serum T3 and T4 were increased in all cases, and the ratio of T3/T4 (ng/micrograms) was 17.8 +/- 3.6 (SD). But in 3 of 10 patients the ratio was greater that 20. TSH receptor antibody (TRAb) and thyroid stimulating antibody (TSAb) were negative in all cases. The estimated thyroid volume by ultrasonography was 18.4 +/- 5.5 ml, which was slightly increased but significantly lower than those in Graves' disease (p less than 0.05). The internal texture of the thyroid showed a decreased echogenicity with a mean echo level of 70.4 +/- 15.4. There was a weak positive correlation between the echo level at the onset of thyrotoxicosis and the lowest T3 level during the clinical course (p less than 0.05). It is suggested that ultrasonography gives a useful information to the diagnosis and outcome of patients with silent thyroiditis.

Data for a total of 942 new cases of hypopituitarism, out of 3493 patients treated with recombinant human growth hormone (GH) for more than 1 year, have been analysed. The mean peak GH correlated well with clinical variables related to growth and was considered to be a good index of GH secretory capacity. Mean peak GH was correlated inversely with obesity (r = -0.253, p less than 0.01). The lower the height velocity before treatment, the mean peak GH and the height SDS, the greater was the therapeutic effect achieved. The patients with mean peak GH less than or equal to 5 ng/ml were defined as having complete GH deficiency (GHD), and those with a mean peak GH of 5-10 ng/ml as having partial GHD. The clinical entity of GH neurosecretory dysfunction could not be identified from either the clinical variables examined or the therapeutic effect. The appearance of GH antibody was considered to be of no clinical significance because its incidence and titre were both low.

Effects of retinoids on DNA synthesis, iodine metabolism, and thyroid peroxidase messenger RNA levels were studied in cultured porcine thyroid cells. Retinol (10(-8)-10(-5) M) alone did not affect DNA synthesis but potentiated that induced by epidermal growth factor or insulin-like growth factor-I without changes in the number or affinity of receptors for the growth factors, suggesting that retinol stimulates postreceptor events responsible for DNA synthesis. Retinol was an inhibitor of TSH-stimulated iodine metabolism. Iodide uptake and release of organified iodine stimulated by TSH or forskolin were inhibited dose dependently by treatment with retinol. The inhibition was detected at 10(-8) M and was approximately 50% at 10(-6) M. The potency of retinoic acid was comparable to that of retinol. The inhibitory effect of retinol was detected after treatments of thyroid cells for 24 h, and the maximal effect occurred after 48 h incubation. The cAMP accumulation in cultures treated with TSH plus retinol was lower than that of control cultures treated with TSH alone. However, iodide uptake stimulated by 8-bromo-cAMP was also inhibited by retinoids. Retinol reduced TSH- or 8-bromo-cAMP-stimulated gene expression of thyroid peroxidase. Thus, the data suggest that retinoids inhibit TSH-stimulated iodine metabolism by reducing cAMP accumulation and also by acting on the steps subsequent to cAMP production.

Glucocorticoids are widely used for the treatment of malignancy-associated hypercalcemia to delay the occurrence of an escape phenomenon inherent in calcitonin therapy. Using parathyroid hormone-related protein (PTHrP)-producing squamous carcinoma cells (T3M-1 and EC-GI) established in our laboratory, we investigated the in vitro effects of glucocorticoids and calcitonin on PTHrP mRNA expression in the cells and release of PTHrP into the culture medium. The PTHrP gene was constitutively expressed in the logarithmic growth phase in both squamous carcinoma cell lines. When these cells became superconfluent, PTHrP mRNA expression was greatly diminished in T3M-1 cells but was not distinctly diminished in EC-GI cells. Hydrocortisone inhibited the PTHrP mRNA expression in T3M-1 cells and EC-GI cells in a dose-dependent manner. In accordance with the decreased expression of PTHrP mRNA, the release of immunoreactive as well as bioactive PTHrP also decreased in the conditioned medium of glucocorticoid-treated cells. The minimal effective concentration of prednisolone was about 10(-7) M, which is readily attainable in the serum of patients treated with the agent. Calcitonin and indomethacin did not affect the PTHrP mRNA expression or PTHrP release into the medium. Calcitonin did not modulate the hydrocortisone-induced inhibition of PTHrP production. These in vitro findings suggest that the combined use of glucocorticoids and calcitonin plays a beneficial role in the treatment of malignancy-associated hypercalcemia, since the steroid hormone can suppress PTHrP mRNA expression and release of bioactive PTHrP in certain PTHrP-producing tumors.

Insulin-like growth factor (IGF)-II receptors were demonstrated in normal and neoplastic tissues of human thyroid. Specific binding of [125I]IGF-II to thyroid membranes was dependent on the time and temperature of incubation, and a steady state was achieved after 22 h of incubation at 4 degrees C. The binding of [125I]IGF-II was dose-dependently displaced by unlabelled IGF-II with 50% inhibition at an IGF-II concentration of 6 ng/ml. IGF-I had a relative potency of 1% compared to IGF-II, and insulin showed no inhibition at concentrations as high as 2000 ng/ml. Scatchard analysis of the binding data revealed a single class of IGF-II receptor with high affinity. Affinity crosslinking and autoradiography demonstrated the type II IGF receptors. Specific binding of [125I]IGF-II to thyroid papillary cancer tissues (mean [S.D.]13.2[1.3]% per 200 micrograms protein, n = 8) was significantly (P less than 0.01) higher than that to the surrounding normal tissues (4.8 [0.5]%). The binding in follicular cancer and follicular adenoma was also significantly higher than that in the corresponding normal tissues. The higher IGF-II binding to neoplastic tissues was due to an increase in the number of binding sites without any change of affinity. These results suggest that the increased IGF-II receptors may be involved in growth or functions of thyroid neoplasms.

Methimazole (MMI) increases thyroglobulin (Tg) mRNA levels in FRTL-5 rat thyroid cells. The increase reflects a transcriptional action of the antithyroid agent and is inhibited by cycloheximide, as is the transcriptional action of TSH. It takes several hours to be apparent, is maximal between 24-48 h, and is specific, in that thyroid peroxidase and beta-actin mRNA levels are not increased simultaneously. The increased mRNA levels are associated with increased recovery of immunoprecipitable Tg in the medium of cells exposed to [35S]methionine. The MMI effect appears to be independent of the action of TSH or its cAMP signal, since the MMI-induced increase in Tg mRNA levels is evident in cells treated with TSH or maintained in its absence and is associated not with increases in cAMP levels but, rather, under some circumstances with a decrease. The effect is evident under conditions in which the ability of insulin or insulin-like growth factor-I to increase Tg mRNA levels is already maximal. The MMI-induced increase is inhibited by concentrations of iodide associated with autoregulation of FRTL-5 rat thyroid cells, is inhibited but not mimicked by propylthiouracil, and is not altered by T3. The increase in Tg mRNA levels does not correlate with increased DNA synthesis as a function of MMI concentration either in cells treated with TSH or in those maintained in its absence. A concentration of MMI (5 mM) that increases Tg mRNA levels can also inhibit 8-bromo-cAMP- or phorbol ester-induced increases in [3H]thymidine incorporation into DNA.

A 73-yr-old patient with idiopathic hypoparathyroidism was admitted to our hospital in May 1981. The immunoreactive PTH (iPTH) level determined by RIA using antiserum specific for the C-terminal region of PTH-(65-69) was in the upper normal range (0.6 ng/mL) and over the next 7 yr increased gradually to 6 ng/mL. Since iPTH levels determined using other commercial RIA kits remained constantly decreased or in the undetectable range, we studied the mechanism of false elevation of iPTH in this patient. The patient's serum contained no binding protein to the tracer ([125I]) [Tyr45] human PTH-(46-84)), nor was any heterophilic antibody to the first [guinea pig immunoglobulin G (IgG)] or the second antibody (goat IgG) detected. Consistent with these findings, the dilution curve of the serum was parallel with that of standard bovine PTH-(1-84). Gel filtration analysis revealed that the iPTH-like substance was eluted in the void volume (apparent mol wt, greater than 70,000). Almost all of the iPTH-like substance was adsorbod by a protein-A-Sepharose column. When the IgG fraction purified by protein-A-Sepharose affinity chromatography was applied to an antihuman IgG lambda-Sepharose column, 72% of the iPTH-like substance was detected in the IgG lambda. These results suggest that the falsely elevated iPTH in the patient's serum was due to IgGs (mainly IgG lambda), which were cross-reactive with the antiserum highly specific for the C-terminal region of human PTH-(65-69).

The effects of endothelin on systemic hemodynamics and renal functions were investigated in anesthetized dogs. Infusion of endothelin at a dose of 1 ng/kg/min decreased renal blood flow and increased renal vascular resistance and filtration fraction. Endothelin at doses higher than 10 ng/kg/min significantly decreased cardiac output, glomerular filtration rate, renal plasma flow, urine volume, and urinary sodium excretion. Mean arterial pressure showed a transient decrease at doses higher than 50 ng/kg/min. These results showed that endothelin in systemic administration has effects on renal functions as well as on systemic hemodynamics.

To investigate the mechanism of the inhibitory effects of interferon-gamma (IFN-gamma) on bone resorption, the effects of murine IFN-gamma on 45Ca release from prelabeled fetal mouse forearm bones were investigated. Murine IFN-gamma usually did not affect basal 45Ca release but almost completely and equipotently inhibited bone resorption induced by PTH(1-34), PTH-rP(1-34), 1,25(OH)2D3, and interleukin 1 (IL-1). The half-maximal concentration for inhibition of bone resorption induced by IL-1 alpha was 25.8 +/- 14.6 U/ml (mean +/- SD for 13 experiments), which is not different from those for PTH, PTH-rP, and 1,25(OH)2D3. There was no correlation between prostaglandin E2 concentration in the conditioned medium and 45Ca release from the IFN-gamma-treated forearm bones. The inhibitory effect of IFN-gamma on bone resorption induced by PTH-rP (1-34) or IL-1 alpha continued during 6 days of culture, whereas that of calcitonin disappeared after 2 days of culture. These findings suggest that IFN-gamma non-preferentially inhibits bone resorption induced by various bone-resorbing factors in fetal mouse forearm bones via a PGE2-independent mechanism. As no escape phenomenon developed in IFN-gamma-treated bones, the cytokine may be potentially useful for treatment of certain patients with malignancy-associated hypercalcemia.

The levels and molecular form of atrial natriuretic peptide-like immunoreactivity (ANP-LI) in human atrial tissue were investigated. The levels of right atrial ANP-LI were significantly higher in mitral disease than in other cardiac or noncardiac diseases. The increased ANP-LI was mainly accounted for by an increase in beta-human ANP-LI (beta-hANP-LI). Both total ANP-LI and beta-hANP-LI levels were associated with the presence of atrial fibrillation and with increased atrial pressure. Plasma beta-hANP-LI levels were also increased in mitral disease. These results suggest that human atrium with hemodynamic overloads is characterized by increased tissue levels of ANP and by a shift to the beta-hANP form.

The effects of endothelin on renal hemodynamics and excretory functions were investigated in anesthetized dogs. Infusion of endothelin at a rate of 1 ng/kg·minresulted in a slight but significant decrease in renal blood flow and an increase in renal vascular resistance and filtration fraction. Endothelin at doses higher than 10 ng/kg·min significantly decreased cardiac output, glomerular filtration rate, urine volume, and urinary sodium and potassium excretion, whereas it increased systematic vascular resistance. Mean arterial pressure and heart rate showed a transient decrease and increase, respectively, at doses higher than 50 ng/kg·min. Plasma renin activity and plasma aldosterone concentrations were increased only at the dose of 100 ng/kg·dmin. These effects lasted for more than 60 min. These results suggest that endothelin may have an important role in the modulation of renal functions as well as in the modulation of systemic hemodynamics.

Patients with Turner syndrome have many somatic characteristics, including short stature. We report the results of a 6-year multicentre clinical trial of recombinant human growth hormone (GH) therapy

Vascular endothelial growth factor (VEGF), a secreted endothelial cell-specific mitogen, is produced in endocrine organs and regulated by trophic hormones. Because angiogenesis and osteogenesis are closely regulated, we studied whether human osteoblast-like cells produce VEGF, and if so, what factors regulate VEGF mRNA expression. Human osteoblast-like cells (HObLC) derived from trabecular bone explants were cultured in alpha-MEM supplemented with 10% fetal calf serum. Northern blot analysis revealed that HObLC expressed VEGF mRNA, as did several human osteosarcoma cells. 1,25-(OH)2D3 increased the steady-state levels of VEGF mRNA in a time- and concentration-dependent manner in HObLC and one of the osteosarcoma cell lines, SaOS-2, accompanied by an increase in the concentration of immunoreactive VEGF in the conditioned medium. PTH and IGF-I also increased the level of VEGF mRNA in HObLC and SaOS-2 cells. Furthermore, 12-O-tetradecanoylphorbol ester stimulated VEGF mRNA in a time-and concentration-dependent manner. The VEGF mRNA expression induced by 1,25-(OH)2D3 was completely inhibited by H-7, but only partially by staurosporine. We have demonstrated that PTH, IGF-I, and most potently 1,25-(OH)2D3 stimulate the mRNA expression and secretion of VEGF in human osteoblast-like cells, suggesting that one of the anabolic effects of 1,25-(OH)2D3 on skeletal tissue may be mediated by VEGF produced by osteoblasts.

The role of cell surface heparan sulfate proteoglycans in the effect of basic fibroblast growth factor (bFGF) on FRTL-5 rat thyroid cells was investigated and compared with that of endothelial cells. FRTL-5 cells were incubated for 2 h with heparitinase (0.5-5.0 mU/mL), which specifically degrades heparan sulfate proteolgycans, and then stimulated by bFGF. The mitogenic effect of bFGF was estimated by measuring [3H]thymidine incorporation. Although cell surface heparan sulfates have been believed to be necessary for bFGF binding to its high affinity receptors, the heparitinase treatment had no significant effect on the DNA synthesis of FRTL-5 cells stimulated by bFGF. The binding study revealed that heparitinase treatment decreased low affinity bindings of [125I]bFGF to FRTL-5 cells by only 50% and did not attenuate the high affinity binding, while the same treatment abolished the high and low affinity binding to bovine pulmonary artery endothelial (CPAE) cells. Analysis of trypsin accessible cell surface 35SO4-labeled materials by Q-sepharose anion-exchange column chromatography showed that heparan sulfate proteoglycans, peaked at 0.55 M NaCl elution, disappeared from the surface of FRTL-5 cells after treatment with 2.0 mU/mL of heparitinase, indicating that the heparitinase resistant low-affinity binding sites are not heparan sulfates. These results demonstrate that cell surface heparan sulfates are not required for the high affinity binding of bFGF to FRTL-5 rat thyroid cells, while proteoglycans are necessary for binding to endothelial cells, and suggest that the mechanism of the action of bFGF is different in rat thyroid cells compared with endothelial cells.

To elucidate the pathogenesis of thyroid gland hypervascularity in patients with Graves' disease, we studied the expression of mRNAs for vascular endothelial growth factor (VEGF) and its receptor, Flt family, using human thyroid follicles in vitro and thiouracil-fed rats in vivo. Human thyroid follicles, cultured in the absence of endothelial cells, secreted de novo-synthesized thyroid hormone in response to thyroid-stimulating hormone (TSH) and Graves' IgG. The thyroid follicles produced VEGF mRNA but not flt-1 mRNA. The expression of VEGF mRNA was enhanced by insulin, tumor-promoting phorbol ester, calcium ionophore, dibutyryl cAMP, TSH, and Graves' IgG. When rats were fed thiouracil for 4 wk, their serum levels of TSH were increased at day 3. VEGF mRNA was also increased on day 3, accompanied by an increase in flt family (flt-1 and KDR/ flk-1) mRNA expression. These in vitro and in vivo findings suggest that VEGF is produced by thyroid follicles in response to stimulators of TSH receptors, via the protein kinase A and C pathways. VEGF, a secretable angiogenesis factor, subsequently stimulates Flt receptors on endothelial cells in a paracrine manner, leading to their proliferation and producing hypervascularity of the thyroid gland, as seen in patients with Graves' disease.

Using a highly sensitive bioassay for TSH, in which human thyroid follicles incorporate 125I and release de novo synthesized thyroid hormone into the culture medium, the thyrotropic activities of various hCG preparations were studied. Under the culture conditions employed, bovine TSH (bTSH) was approximately 6- to 9-fold more active than human TSH (hTSH). Highly purified hCG prepared from urine of normal pregnant women (CR 127) had only a trivial thyrotropic activity equipotent to 0.00022 microU bTSH/U hCG or 0.0013 microU hTSH/U hCG (19.7 microU hTSH/mg hCG). Hybrid hCG (AB1ER) also elicited low thyrotropic activity (14.0 microU hTSH/mg), whereas crude hCG had moderate thyrotropic activity (0.041 hTSH microU/U hCG or 127 microU/mg protein). Deglycosylated hCG, a very weak LH/hCG receptor agonist, was the most potent agonist in thyroid follicles (588 microU hTSH/mg protein). hCGs purified from urine of patients with trophoblastic tumors had greater TSH-like activity (37-84 microU hTSH/mg protein) than purified hCG. Asialo-hCG purified from a patient with choriocarcinoma had very potent TSH-like activity (468 microU hTSH/mg). Submaximal doses of bTSH and hCG variants produced additive stimulation of thyroid function. Furthermore, the thyrotropic effect of hCG was inhibited by anti-TSH receptor antibody obtained from patients with myxedema. These in vitro findings suggest that although hCG is reported to exert potent cAMP-stimulating activity on rat thyroid-like cells (FRTL-5) and Chinese hamster ovary cells transfected with hTSH receptor complementary DNA (0.092-0.72 microU hTSH/U hCG), the thyrotropic activity induced by authentic hCG in human thyroid follicles is too weak to cause hyperthyroidism in normal pregnancy. However, hCG produced by some trophoblastic tumors, particularly asialo-hCG, has potent thyrotropic activity sufficient to cause clinically overt hyperthyroidism when produced excessively.

Seven patients with growth hormone (GH)-secreting pituitary adenoma were treated preoperatively with octreotide (Sandostatin or SMS 201-995; a somatostatin analogue), and were compared with 18 non-treated patients in their clinical courses and adenoma analyses. Octreotide treatment improved the endocrinological data in all 7 cases. The octreotide-treated adenomas were soft and easily removed by suction and curettage. The postoperative normalization of endocrinological data was encountered more often in the octreotide-treated cases than in the non-treated, although the statistical significance was not observed by the limited number of cases. The adenoma tissues were examined with conventional histology and immunohistochemistry, and the amount of GH messenger ribonucleic acid (mRNA) was quantitatively assessed. The studies demonstrated: 1) No fibrosis nor necrosis was observed in the adenomas from the octreotide-treated patients. 2) Immunohistochemistry for human GH revealed no remarkable differences between the octreotide-treated and the non-treated adenomas. 3) The amounts of GH mRNA in the adenoma from the octreotide-treated patients were 4.2 +/- 1.8 (mean +/- SEM; expressed in an arbitrary unit) and were significantly less than those from the non-treated (33.6 +/- 9.1). These data suggest that octreotide inhibits not only GH release from the adenoma but also its biosynthesis.

Recently, with an availability of recombinant human insulin-like growth factor I (IGF-I), IGF-I has been tried for various clinical therapeutic applications. IGFs are found in blood complexed to specific binding proteins (IGFBPs), and these IGFBPs might influence the biological effect of administered IGF-I. Therefore, it is interesting to study the responses of serum IGFBPs to IGF-I administrations. We have previously reported that serum IGFBP-I levels increased after single sc injection of IGF-I, and serum IGFBP-2 levels increased and decreased serum IGFBP-3 and acid labile subunit levels after repetitive IGF-I injections in normal subjects1. We report here the responses of IGFBPs to single and repetitive IGF-I sc injections in patients with growth hormone (GH) deficiency.

Malignancy-associated hypercalcemia is mainly caused by excessive production of parathyroid hormone-related protein (PTHrP) by the tumor. Using anti-PTHrP-(1-34) monoclonal murine antibody (anti-PTHrP MoAb), we studied whether repeated injection of the homologous antibody would continuously decrease the serum calcium concentration in hypercalcemic nude mice bearing transplanted human PTHrP-producing tumors, leading to prolongation of their survival time. Daily SC injections of anti-PTHrP MoAb decreased the serum calcium concentration almost to within the normal range in nude mice bearing transplanted human PTHrP-producing tumors (T3M-1, EC-GI, PC-3, and FA-6) but not in a nude mouse bearing a transplanted parathyroid carcinoma. The antibody did not affect FA-6 tumor growth either in vitro or in vivo. Pancreatic carcinoma cells (FA-6), which caused the most severe hypercalcemia, were inoculated into 6-week-old nude mice. When severe hypercalcemia (approximately 19 mg/dl) had developed, daily SC injection of anti-PTHrP MoAb was started. Within 18 days of this time point, all untreated tumor-bearing mice (n = 10) died of hypercalcemia and cachexia, whereas all the treated mice (n = 10) showed an increase in body weight and survived for at least 25 days. Histologic examination of the treated mice revealed a marked decrease in osteoclastic bone resorption, without toxicologic findings in the kidney and liver. These results suggest that passive immunization against PTHrP can continuously ameliorate the hypercalcemia and markedly prolong the survival time of severely hypercalcemic, tumor-bearing mice. If a human monoclonal antibody against PTHrP-(1-34) could be developed, then passive immunization would be potentially one of the most effective therapies for patients with malignancy-associated hypercalcemia due to excessive production of PTHrP.

Recently, interleukin 4 (IL-4) was reported to inhibit bone resorption in mouse long bone culture. To test the effect of IL-4 on the formation of osteoclast-like cells, we used a mouse bone marrow culture system that formed mononuclear and multinucleated cells with osteoclast characteristics. Recombinant mouse IL-4 dose-dependently inhibited the formation of tartrate-resistant acid phosphatase-positive multinucleated cells [TRAP(+)MNC] induced by 1,25(OH)2D3, PTH(1-34) or Interleukin-1 alpha (IL-1 alpha). The minimum effective inhibitory concentration was 0.01 ng/ml, and 1 ng/ml IL-4 completely inhibited TRAP(+)MNC formation. IL-4 also dose-dependently inhibited the formation of tartrate-resistant acid phosphatase-positive mononuclear cells. Treatment of cultures with IL-4 for the first or last 48 h of an 8-day culture period inhibited TRAP(+)MNC formation to the same extent, whereas IFN-gamma and calcitonin suppressed TRAP(+)MNC formation mainly at the early and the late phase, respectively. IL-4, as a macrophage fusion factor, at higher concentrations (0.1-10 ng/ml), increased formation of tartrate-resistant acid phosphatase-negative multinucleated cells [TRAP(-)MNC] with giant macrophage characteristics. The half-maximal concentrations inhibiting TRAP(+)MNC formation and stimulating TRAP(-)MNC formation were 0.05 ng/ml and 2 ng/ml, respectively. These results demonstrate that IL-4 inhibits bone resorption by inhibiting the recruitment of osteoclast precursor and formation of multinucleated osteoclast-like cells, and by stimulating the formation of macrophage polykaryons.

Lymphokine-activated killer (LAK) cell therapy frequently results in primary hypothyroidism. To elucidate the responsible mechanism, we investigated the effects of interleukin-2 (IL-2) on thyroid function of cultured human thyroid follicles in the presence or absence of autologous thyroid-derived lymphoid cells (TDLC). Human thyroid follicles, obtained by subtotal thyroidectomy from patients with Graves' disease, were cultured in serum-free medium containing bTSH and various concentrations of human IL-2, with or without TDLC. After 5 days of culture, 125I was added, and after an additional 3 days of culture, 125I incorporated into thyroid follicles and organic 125I (125I-T4 + 125I-T3) released into the culture medium were estimated. In the absence of TDLC, IL-2 did not affect bTSH-induced thyroid function. In the presence of TDLC, however, IL-2 inhibited the bTSH-stimulated thyroid function in a concentration-dependent manner. The minimum IL-2 concentration required to inhibit thyroid function was 1 U/ml. At 100 U/ml, IL-2 inhibited thyroid function completely. IL-2 increased the concentration of IFN-gamma in the culture medium conditioned by TDLC but not by thyroid follicles. When both were cocultured, IL-2 more significantly increased the concentration of IFN-gamma to an extent sufficient for inhibiting thyroid function per se. Furthermore, the addition of anti-IFN-gamma antibody to the medium partially restored the IL-2-inhibited thyroid function. In summary, by coculturing human thyroid follicles and autologous intrathyroidal lymphocytes with IL-2, it was possible to reproduce the thyroid dysfunction associated with LAK cell therapy. Our data indicate that IL-2 inhibits thyroid function, at least partly, by stimulating production of IFN-gamma, a potent inhibitory cytokine for thyroid cells.

Hypercalcemia with adult T-cell leukemia (ATL) is chiefly caused by an excessive production by tumor cells of parathyroid hormone-related protein (PTHrP). We have previously reported hypercalcemic patients with solid tumors to excrete a large amount of the C-terminal fragments of PTHrP (C-PTHrP) into their urine. To elucidate whether PTHrP production correlates with or predicts the development of hypercalcemia, we studied the urinary excretion of C-PTHrP in 36 ATL patients. The urinary excretion of C-PTHrP was in the normal range (< 0.40 nmol equivalent to PTHrP (109-141)/g creatinine) in HTLV-1-positive carriers (n 3), ATL patients in complete remission (n 2) and chronic type ATL patients (n 2). It was marginally increased in seven patients in partial remission, and gradually increased as the disease progressed. In 20 patients who died without or with hypercalcemia, it was increased to 1.98 +/- 0.69 (n 9) and 7.6 +/- 2.1 nmol/g creatinine (mean +/- SD, n 11, P < 0.01), respectively. Urinary C-PTHrP excretion was significantly correlated with serum calcium and LDH levels as well as with CD25-positive cells in the peripheral blood. In four patients whose urinary excretion had been serially determined, it increased prior to the development of hypercalcemia. The findings suggest the urinary excretion of C-PTHrP to be of use as a predictor of the development of hypercalcemia in ATL patients. In ATL patients whose urinary excretion of C-PTHrP is progressively increasing, the serum calcium concentration should be carefully monitored to prevent hypercalcemic crisis.

Periodic paralysis has been known to be associated with thyrotoxicosis in Japan. The incidence was 8.6% among male and 0.4% among female thyrotoxic patients according to a survey performed in the three major thyroid clinics in Japan in 1957. To determine the changes in the incidence during the intervening 34 years, the same type of survey was carried out again in 1991 at the same three major thyroid clinics previously involved. The incidence of paralysis in 1991 was 4.3% among male and 0.04% among female thyrotoxic patients, indicating more than a 40% decrease in the incidence. The possible cause of the decrease is related to the changes in food consumption, namely, to the fact that less carbohydrate and more potassium were taken in 1991 than in 1957.

In FRTL5 thyroid cells, endothelin(ET)-1 alone had no effect on DNA synthesis but caused a transient increase in c-fos mRNA levels and stimulated IGF-I induced DNA synthesis and cell proliferation. By contrast, ET-1 inhibited the stimulatory effects of TSH actions on DNA synthesis, cell proliferation and c-AMP production. 8-Bromo-cAMP-induced DNA synthesis was also inhibited by ET-1, suggesting that ET-1 exerts its inhibitory effects at step(s) involving cAMP production and post cAMP pathway. ET-1-induced suppression of TSH actions were reversed by a C-kinase inhibitor, H-7. These results suggest that the effect of ET on functions of FRTL5 cells is, at least, in part mediated by C-kinase dependent pathway.

An established cell line (TC-cell, clone 78) derived from human thyroid papillary cancer cells was investigated for production of peptide growth factors. The cells had specific binding sites for insulin-like growth factor-I (IGF-I) and responded to this growth factor with increased proliferation. Culture medium conditioned by TC cells was found to contain insulin-like growth factor (IGF)-I and IGF-binding protein(s). Furthermore, reverse transcription-polymerase chain reaction revealed expression of IGF-I mRNA. When monoclonal antibody to IGF-I receptors (αIR3) was added, the growth of TC cells cultured in serum-free medium was significantly reduced. The growth rate of the cells was restored when the antibody was removed from the medium. These results strongly suggest that TC cells produce IGF-I, which is involved in the regulation of their own growth.

In order to study the mechanism of GH secretion from somatotroph adenoma cells, we have compared the effect of 12-O-tetradecanoyl phorbol-13-acetate (TPA) with that of growth hormone releasing factor (GRF) on GH secretion from human somatotroph adenoma cells cultured in monolayer. Pituitary adenoma cells were obtained from 13 patients with acromegaly undergoing surgery. On the 7th day of culture, the cells were exposed for 2 h to secretagogues. All 13 adenoma cell cultures (100%) responded to TPA (1.6-16.0 nmol/l) with a two- to six-fold increase in GH release (240 +/- 37% increase of control: mean +/- SE). The response was detectable within 10 min, and was maximal at 2 h. Phospholipase C (7.7 mmol/l) also stimulated a two- to ten-fold increase in GH release in all four adenomas examined (100%). GH release was stimulated by GRF (2.0 nmol/l) in eight out of 12 adenoma cells (67%), but the magnitude of the responses to GRF (60 +/- 18% increase of control: mean +/- SE) were much smaller than that of TPA. Five out of 13 adenomas secreted detectable amount of PRL into the medium and these five adenomas (100%) responded to TPA (16.0 nmol/l) with a two- to six-fold increase. These observations indicate that the activation of protein kinase C is the consistent stimulator in GH and PRL secretion in human somatotroph adenoma cells. However, it is not determined whether the protein kinase C is involved in the in-vivo production of GH in patients with acromegaly.

To elucidate the physiological role of atrial natriuretic peptide (ANP) in plasma during mineralocorticoid escape, we investigated the effects of passive immunization with ANP-specific antiserum on deoxycorticosterone (DOCA)- treated rats. Sodium was retained in excess of intake during the first day after treatment with DOCA, and sodium balance returned to control values by the second day, whereas the excretion of potassium exceeded the intake during all days after DOCA treatment. These changes in electrolyte balance were associated with a significant increase in plasma levels of ANP. Administration of ANP-specific antiserum significantly impaired the return to normal sodium balance as well as the augmented kaliuresis that were observed on the second day after injection of DOCA. No significant effect was observed on either sodium or potassium balance after the injection of normal rabbit serum. These results suggest that plasma ANP plays an important role in mineralocorticoid escape. (Endocrinology 128: 22...

Insulin-like growth factor I (IGF-I) has been synthesized by recombinant DNA technology (1). With the availability of large quantities of the biosynthetic peptide, the biological effects have been studied in animals in vivo. We have previously reported that IGF-I has growth promoting, anabolic, and insulin-like effects in hypophysectomized and normal rats, and IGF-I inhibits catabolism in fasted rats in vivo (2–5). Recently, the biological effects of IGF-I have been studied in normal men and patients with Laron-type dwarfism (6–9). We have also studied effects of IGF-I single sc administration in normal men (10). In this paper, the results of the single sc administration of IGF-I combined with our previous data (10) are reported.

Fibroblast growth factor (FGF) is one of the most potent endothelial cell growth factors and has been found in almost all tissues in the body. However, there is no definitive report showing that FGF exists in the thyroid. In this report, we describe heparin binding endothelial cell growth factor activity in porcine thyroids. The extract from normal adult porcine thyroids was loaded onto a heparin-sepharose affinity column. The mitogenic activity on endothelial cells was found to elute from the column with 0.9 M-2.0 M NaCl. Polyacrylamide gel electrophoresis and an immunoblotting of bioactive fractions showed the basic FGF immunoreactivity in a double band with a molecular weight of 18K and 20K. These results strongly imply that the adult porcine thyroid-derived heparin binding endothelial cell growth factor may be authentic basic FGF and one of its high molecular weight forms. In support of this conclusion, basic FGF mRNA was detected in poly A+ RNA isolated from cultured porcine thyroid cells. On the other hand, recombinant human basic FGF at the concentration of 0.1-10 ng/ml increased the incorporation of 3H-thymidine into FRTL-5 cells and porcine thyroid follicular cells in culture. The stimulatory effects were also observed when the biologically active fractions of heparin-sepharose column of the thyroid extract were added to the culture. In addition, both recombinant human basic FGF at 10 ng/ml and heparin binding fractions from porcine thyroids inhibited TSH-induced iodide uptake by porcine thyroid cells in culture. These results suggest that basic FGF may be one of the important local modulators of thyroid function, cell growth, and angiogenesis.

We sought to clarify the mechanisms of growth hormone (GH) secretion induced by insulin hypoglycemia, L-dopa, and arginine in man. The secretion of GH as measured by increased plasma level, in response to oral administration of 500 mg L-dopa or 30 min-infusion of arginine, was not modified by prior intravenous administration of 200 micrograms GH-releasing hormone (GHRH). It was, however, completely blocked by preadministered 50 micrograms SMS201-995, a long-acting somatostatin (SRIH) analog. GH release with 200 micrograms GHRH was completely blocked by 100 micrograms SMS201-995. GH secretion caused by insulin-induced hypoglycemia was significantly reduced but still present after administration of 100 micrograms of the analog. These results suggest that a suppression of SRIH release may be partially involved in the stimulatory mechanism of GH secretion by L-dopa. Coadministration of GHRH accentuated the stimulatory effect of arginine on GH secretion. Arginine significantly raised plasma TSH levels. These findings suggest that arginine suppresses SRIH release from the hypothalamus to cause GH secretion because SRIH suppresses TSH secretion. It is also suggested that some factor (or factors) other than GHRH and SRIH are involved in the mechanism by which insulin-induced hypoglycemia stimulates GH secretion, because the effect of insulin was not fully blocked in the presence of SRIH analog. Thus all the tests for GH release appear to act via different mechanisms.

To elucidate the mechanism of decreased 131I uptake by the thyroid gland in patients with subacute thyroiditis and painless thyroiditis, human thyroid follicles were cultured with interleukin-1 (IL-1), tumor necrosis factor-alpha (TNF alpha), and/or interferon-gamma (IFN gamma), and the effects of these cytokines on thyroid function were studied in vitro. When human thyrocytes were cultured in RPMI-1640 medium containing 0.5% fetal calf serum and TSH for 5-8 days, the cells incorporated 125I, synthesized de novo [125I]iodotyrosines and [125I]iodothyronines, and secreted [125I]T4 and [125I]T3 into the medium. IL-1 alpha and IL-1 beta inhibited 125I incorporation and [125I]iodothyronine release in a concentration-dependent manner. The minimal inhibitory effect was detected at 10 pg/ml. Electron microscopic examination revealed a marked decrease in lysosome formation in IL-1-treated thyrocytes. TNF alpha and IFN gamma also inhibited thyroid function in a concentration-dependent manner. Furthermore, when thyrocytes were cultured with IL-1, TNF alpha and IFN gamma, these cytokines more than additively inhibited thyroid function. Although the main mechanism of 131I uptake suppression in the thyroid gland in subacute thyroiditis is due to cellular damage and suppression of TSH release, our present findings suggest that IL-1, TNF alpha, and IFN gamma produced in the inflammatory process within the thyroid gland further inhibit iodine incorporation and at least partly account for the decreased 131I uptake by the thyroid gland in destruction-induced hyperthyroidism.

Recombinant human insulin-like growth factor I (IGF-I) was administered subcutaneously to each of 5 normal human subjects at doses of 0 mg/kg (control), 0.06 mg/kg, or 0.12 mg/kg successively at one week intervals. After 0.06 mg/kg or 0.12 mg/kg IGF-I injections, plasma IGF-I levels increased from 185 +/- 17 ng/ml (mean +/- SEM) to maximal levels of 396 +/- 21 ng/ml at 3 hours and from 169 +/- 14 ng/ml to 480 +/- 27 ng/ml at 4 hours, respectively. These two peak values were statistically different (p less than 0.05). After 0.06 mg/kg and 0.12 mg/kg IGF-I administration, blood glucose levels decreased from 85 +/- 2 mg/dl to minimal levels of 73 +/- 3 mg/dl at 3 hours and from 83 +/- 1 mg/dl to 50 +/- 4 mg/dl at 2 hours, respectively. These two minimal values were statistically different (p less than 0.001). Serum insulin and C-peptide levels were decreased in a dose dependent manner after IGF-I administration. There were no changes between blood urea nitrogen levels before and 4 hours after IGF-I administration. The urinary GH concentration decreased after 0.06 mg/kg IGF-I administration, but increased and maintained normal values after 0.12 mg/kg IGF-I administration.

Evaluation of 24-hour endogenous growth hormone (GH) secretion was carried out in 62 children, aged 7-16 years, who did not have classic GH deficiency (GHD). The mean 24-hour GH concentration, determined at 20-minute intervals over 24 hours, was variable, ranging from 1.28 to 11.39 μg/l with a mean of 4.95 ± 2.55 μl (± SD). There was a positive correlation between mean 24-hour GH concentration and plasma insulin-like growth factor I (IGF-I) values (r= 0.54; p < 0.01). Recombinant human GH, 0.1 IU/kg/day was administered to 30 of the 62 children for 6 months followed by 6 months’observation without treatment. Thereafter, GH was administered at the same dose for a further 6 months to 16 children. The mean height velocities before, during, and after the first treatment period were 4.3 ± 0.9, 7.3 ± 1.9 and 4.9 ± 2.0 cm/year (mean ± SD), respectively. The height velocity during treatment was greater than pre- and post-treatment values (p < 0.001). The height velocity Increased again during the second treatment period to a mean of 8.5 ± 2.0 cm/year (p < 0.001). Nine other children were treated continuously in a similar manner for 1 year and their height velocity increased significantly from 4.1 ± 1.4 to 6.0 ± 1.9 cm/year (p < 0.001). According to our criteria, 29 of the 39 children (74.4%) who were treated for 6-12 months showed a GH-dependent height increase during therapy. There were no differences between the children who responded to GH treatment and those who did not in terms of Chronological age, bone age, plasma IGF-I level, maximal GH level to insulin-induced hypoglycaemia, or mean 24-hour plasma GH concentration. These data indicate that some short children without GHD respond to GH treatment with an increased height velocity. Further investigations are required to determine the effect of GH on final height.

We investigated the direct effect of lithium on porcine thyroid cells in culture to exclude the secondary regulatory factors. First we have studied the effect of lithium on TSH-induced iodide uptake. Significant suppression was seen at 0.1 mmol/liter, and half-maximal suppression was obtained at the pharmacological concentration reported in patient serum. The suppression was dose dependent and reversible. Besides the suppression of cAMP production stimulated by TSH, lithium also inhibited iodide uptake stimulated by forskolin or 8-bromo-cAMP. These results demonstrated that lithium inhibits TSH-induced iodide uptake not only by reducing cAMP production, but also by acting on the steps of post-cAMP production. Next, we studied the effect of lithium on DNA synthesis of the cultured porcine thyroid cells. Lithium stimulated [3H]thymidine incorporation of the thyroid cells in the basal condition (0.5% fetal calf serum) as well as those stimulated by insulin-like growth factor-I (100 micrograms/liter). The minimal concentrations for the significant increase were 0.5 and 0.1 mmol/liter, respectively. These results suggest that lithium might contribute to the formation of the goiter directly at the cellular levels in patients treated with the agent.

125 boys with idiopathic GH deficiency who received GH treatment were followed until they reached their final height. In 85 patients who had spontaneous puberty (group I), the mean final height was 151.5 +/- 6.6 cm. In 23 patients with combined GH and gonadotropin deficiency (group II) whose pubertal development was induced artificially at age 19.4 +/- 2.1 years, the mean final height was 163.7 +/- 3.9 cm (p less than 0.01 vs. group I). Final height was strongly related to height at the onset of pubertal development in combined groups I and II when the time of gonadal replacement treatment was taken as the onset of pubertal development in group II. In 17 patients (group III) who developed spontaneous puberty, gonadal suppression treatment was started at their early stage of puberty and was continued for a mean duration of 4.3 +/- 1.1 years. The mean final height in group III was 157.9 +/- 3.0 cm (p less than 0.01 vs. group I, group II). Longitudinal growth pattern analysis demonstrated that this beneficial effect on final height by gonadal suppression treatment was attributed to the elongation of pubertal growth spurt and pubertal height gain.