Search Results

Angiostatin, a 38 kDa fragment of plasminogen, potently inhibits the growth of blood vessels. Angiostatin is generated from plasminogen by urokinase-type (uPA) and tissue-type (tPA) plasminogen activators in the presence of free sulphydryl donors. Angiogenesis inhibitors may be important in regulating angiogenesis in developing goitre. We have examined angiostatin formation in human primary thyrocyte cultures and a rat thyrocyte cell line (FRTL-5). We found that human thyroid cells in culture secrete plasminogen activators (both tPA and uPA) as well as matrix metalloproteinase 2 into the medium. When human thyrocyte conditioned medium was incubated with plasminogen (10 microg/ml) and N-acetylcysteine (100 microM) for 24 h, a 38 kDa fragment of plasminogen, which is consistent with angiostatin, was generated. The appearance of the 38 kDa fragment was increased by agents that increase cAMP (forskolin and 8 BrcAMP). FRTL-5 cells, which do not secrete uPA or tPA, did not generate angiostatin. Thyroid cells produce several angiogenic growth factors, and human thyrocyte conditioned medium stimulated growth of endothelial cells. When the conditioned medium was incubated with plasminogen and N-acetylcysteine, this stimulatory effect was lost, consistent with the production of a growth inhibitory factor. We conclude that thyroid cells can produce angiostatin from plasminogen in vitro, and this may play a role in vivo in limiting goitre size.

Patients with critical illness or hepatic failure exhibit impaired cortisol responses to ACTH, a phenomenon known as ‘relative adrenal insufficiency’. A putative mechanism is that elevated bile acids inhibit inactivation of cortisol in liver by 5α-reductases type 1 and type 2 and 5β-reductase, resulting in compensatory downregulation of the hypothalamic–pituitary–adrenal axis and adrenocortical atrophy. To test the hypothesis that impaired glucocorticoid clearance can cause relative adrenal insufficiency, we investigated the consequences of 5α-reductase type 1 deficiency in mice. In adrenalectomised male mice with targeted disruption of 5α-reductase type 1, clearance of corticosterone was lower after acute or chronic (eightfold, P<0.05) administration, compared with WT control mice. In intact 5α-reductase-deficient male mice, although resting plasma corticosterone levels were maintained, corticosterone responses were impaired after ACTH administration (26% lower, P<0.05), handling stress (2.5-fold lower, P<0.05) and restraint stress (43% lower, P<0.05) compared with WT mice. mRNA levels of Nr3c1 (glucocorticoid receptor), Crh and Avp in pituitary or hypothalamus were altered, consistent with enhanced negative feedback. These findings confirm that impaired peripheral clearance of glucocorticoids can cause ‘relative adrenal insufficiency’ in mice, an observation with important implications for patients with critical illness or hepatic failure, and for patients receiving 5α-reductase inhibitors for prostatic disease.