Serine-protease inhibition during islet isolation increases islet yield from human pancreases with prolonged ischemia.

BACKGROUND: Islet isolation from the pancreatic tissue matrix remains highly variable. Recent evidence suggests that intrinsic human pancreatic proteases, including trypsin, may inhibit effective collagenase enzymatic activity during islet isolation, thereby impairing the isolation success. In this study we have hypothesized that serine protease inhibition applied during pancreatic digestion, could improve yield and/or functional viability of islets isolated from human pancreases. METHODS: Twelve organ donor pancreases with 12.9+/-0.6 hr cold storage (mean+/-SEM) were perfused via their ducts with Liberase-HI enzyme in the presence (n=6) or absence (n=6) of 0.4 mM Pefabloc. All were then gently dissociated and their purified islets separated with Ficoll density gradient centrifugation. RESULTS: Donor-related factors (age, gender, cold storage time, body mass index, and pancreas weight) did not differ significantly between the two experimental groups. Pefabloc supplementation did not affect the digestion time, islets remaining trapped in exocrine tissue, or final islet purity. Islet recovery was increased in the Pefabloc-treated group (mean+/-SEM yield 323.8+/-80.8 x 10(3) islet equivalents vs. 130.8+/-13.6 x 10(3) islet equivalents, P<0.05). Cellular composition, DNA and insulin content, and insulin secretory activity of the isolated islets was similar. CONCLUSIONS: Inhibition of intrinsic protease activity within pancreases after prolonged cold storage improves isolation of viable islets.

Interferon-gamma acts directly on rejecting renal allografts to prevent graft necrosis.

In transplant rejection interferon (IFN)-gamma regulates the recipient immune response but also acts directly on IFN-gamma receptors in the graft. We investigated these direct actions by comparing rejecting kidneys from donors lacking IFN-gamma receptors (GRKO mice) or control donors (129Sv/J) in CBA recipients. Beginning day 5, 129Sv/J kidneys displayed high major histocompatibility complex (MHC) expression, progressive infiltration by inflammatory cells, but no thrombosis and little necrosis, even at day 21. GRKO kidneys showed increasing fibrin thrombi in small veins, peritubular capillary congestion, hyaline casts, and patchy parenchymal necrosis, progressing to near total necrosis at day 10. Terminal dUTP nick-end labeling assays were positive only in the interstitial infiltrate, confirming that massive cell death in GRKO transplants was not apoptotic. Paradoxically, GRKO kidneys showed little donor MHC induction and less inflammatory infiltration. Both GRKO and 129Sv/J allografts evoked vigorous host immune responses including alloantibody and mRNA for cytotoxic T cell genes (perforin, granzyme B, Fas ligand), and displayed similar expression of complement inhibitors (CD46, CD55, CD59). GRKO kidneys displayed less mRNA for inducible nitric oxide synthase and monokine inducible by IFN-gamma but increased heme oxygenase-1 mRNA. Thus IFN-gamma acting on IFN-gamma receptors in allografts promotes infiltration and MHC induction but prevents early thrombosis, congestion, and necrosis.

Tissue distribution of calcineurin and its sensitivity to inhibition by cyclosporine.

The immunosuppressive activity of cyclosporine is mediated by inhibiting calcineurin phosphatase. However, calcineurin is widely distributed in other tissues. We examined the degree of calcineurin inhibition by cyclosporine in various tissues. In vitro, the cyclosporine concentration inhibiting 50% (IC50) of calcineurin was to approximately 10 ng/mL in human and mouse leukocytes suspensions. In vitro and in vivo IC50s of cyclosporine in homogenates of mouse kidney, heart, liver, testis, and spleen were also comparable (9-48 ng/mL). The maximum calcineurin inhibition by cyclosporine varied, from 83 to 95% of calcineurin activity in spleen, kidney, liver, and testis to 60% in heart and only 10% in brain. Maximum calcineurin inhibition was increased by the addition of cyclophilin A, indicating that cyclophilin concentrations were limiting in some tissues, at least in this assay. Western analysis of mouse tissues showed significantly less cyclophilin in heart than other tissues. cyclosporine concentrations per weight of tissue protein were highest in kidney and liver and lowest in brain and testis after oral dosing, with intermediate levels in spleen, heart, and whole blood. Thus each cyclosporine dose produces rapid and wide-spread inhibition of calcineurin in tissues, with differences in total susceptibility of each tissue.

The temporal profile of calcineurin inhibition by cyclosporine in vivo.

BACKGROUND: Cyclosporine (CsA) acts by inhibiting the phosphatase calcineurin (CN), but the time course and extent of inhibition in vivo are unknown. We examined the effect of single oral CsA doses on CN activity in humans and mice in vivo. METHODS: In humans, blood CsA levels were determined and CN activity was measured in whole blood and in blood leukocytes of patients up to 12 hr after CsA dosing (just before the second dose). Samples were collected from patients receiving a first single dose (2.5 mg/kg), and up to 14 days later after repeated dosing. In mice, after CsA dosing (12.5-200 mg/kg) by oral gavage, CsA levels in blood and tissue (spleen, kidney) were determined and CN activity was measured in spleen and kidney. RESULTS: In humans, peak CsA levels of 800-2285 microg/L at 1-2 hr produced 70-96% CN inhibition. Inhibition correlated closely with the rise and fall of CsA levels with no observable lag at the times sampled. Repeated doses showed similar CN inhibition to first dose, with no significant adaptation. In mice, CsA peaked at 1 hr in blood, spleen, and kidney, with higher concentrations in spleen and kidney than in blood. CN inhibition closely followed CsA concentrations/doses, and was greater in kidney than spleen. CONCLUSION: Thus CsA induces partial CN inhibition that varies directly with the blood and tissue levels, and may be greater in some tissues due to higher drug accumulation. The high CsA concentrations and CN inhibition in kidney may be relevant to nephrotoxicity.

Margatoxin binds to a homomultimer of K(V)1.3 channels in Jurkat cells. Comparison with K(V)1.3 expressed in CHO cells.

Voltage-gated potassium (K(V)) channels play key roles in setting the resting potential and in the activation cascade of human peripheral T lymphocytes. Margatoxin (MgTX), a 39-amino acid peptide from Centruroides margaritatus, is a potent inhibitor of lymphocyte K(V) channels. The binding of monoiodotyrosinyl margatoxin ([125I]MgTX) to plasma membranes prepared from either Jurkat cells, a human leukemic T cell line, or CHO cells stably transfected with the Shaker-type voltage-gated K+ channel, K(V)1.3, has been used to investigate the properties of lymphocyte K(V) channels. These data were compared with [125I]MgTX binding to heterotetrameric K(V) channels in rat brain synaptic plasma membranes [Knaus, H. G., et al. (1995) Biochemistry 34, 13627-13634]. The affinity for [125I]MgTX is 100-200 fM in either Jurkat or CHO/K(V)1.3 membranes, and the receptor density is 20-120 fmol/mg in Jurkat membranes or 1000 fmol/mg in CHO/K(V)1.3 membranes. In contrast to rat brain, [125I]MgTX binding to Jurkat and CHO/K(V)1.3 membranes exhibits an absolute requirement for K+, with no potentiation of binding by Na+. K(V)1.3 was the only K(V)1 series channel present in either CHO/K(V)1.3 or Jurkat plasma membranes as determined by immunoprecipitation of [125I]MgTX binding or by Western blot analyses using sequence-specific antibodies prepared against members of the K(V)1 family. The relative potencies of a series of peptidyl K(V) channel inhibitors was essentially the same for inhibition of [125I]MgTX binding to Jurkat, CHO, or rat brain membranes and for blocking 86Rb+ efflux from the CHO/K(V)1.3 cells, except that alpha-dendrotoxin was more potent at blocking binding to rat brain membranes than in the other assays. The characteristics of [125I]MgTX binding, the antibody profiles, and the effects of the peptidyl K(V) inhibitors all indicate that the [125I]MgTX receptor in Jurkat lymphocytes is comprised of a homomultimer of K(V)1.3, unlike the heteromultimeric arrangement of the receptor in rat brain.

Functional role of the beta subunit of high conductance calcium-activated potassium channels.

Mammalian high conductance, calcium-activated potassium (maxi-K) channels are composed of two dissimilar subunits, alpha and beta. We have examined the functional contribution of the beta subunit to the properties of maxi-K channels expressed heterologously in Xenopus oocytes. Channels from oocytes injected with cRNAs encoding both alpha and beta subunits were much more sensitive to activation by voltage and calcium than channels composed of the alpha subunit alone, while expression levels, single-channel conductance, and ionic selectivity appeared unaffected. Channels from oocytes expressing both subunits were sensitive to DHS-I, a potent agonist of native maxi-K channels, whereas channels composed of the alpha subunit alone were insensitive. Thus, alpha and beta subunits together contribute to the functional properties of expressed maxi-K channels. Regulation of co-assembly might contribute to the functional diversity noted among members of this family of potassium channels.

Tremorgenic indole alkaloids potently inhibit smooth muscle high-conductance calcium-activated potassium channels.

Tremorgenic indole alkaloids produce neurological disorders (e.g., staggers syndromes) in ruminants. The mode of action of these fungal mycotoxins is not understood but may be related to their known effects on neurotransmitter release. To determine whether these effects could be due to inhibition of K+ channels, the interaction of various indole diterpenes with high-conductance Ca(2+)-activated K+ (maxi-K) channels was examined. Paspalitrem A, paspalitrem C, aflatrem, penitrem A, and paspalinine inhibit binding of [125I]charybdotoxin (ChTX) to maxi-K channels in bovine aortic smooth muscle sarcolemmal membranes. In contrast, three structurally related compounds, paxilline, verruculogen, and paspalicine, enhanced toxin binding. As predicted from the binding studies, covalent incorporation of [125I]ChTX into the 31-kDa subunit of the maxi-K channel was blocked by compounds that inhibit [125I]ChTX binding and enhanced by compounds that stimulate [125I]ChTX binding. Modulation of [125I]ChTX binding was due to allosteric mechanisms. Despite their different effects on binding of [125I]ChTX to maxi-K channels, all compounds potently inhibited maxi-K channels in electrophysiological experiments. Other types of voltage-dependent or Ca(2+)-activated K+ channels examined were not affected. Chemical modifications of paxilline indicate a defined structure-activity relationship for channel inhibition. Paspalicine, a deshydroxy analog of paspalinine lacking tremorgenic activity, also potently blocked maxi-K channels. Taken together, these data suggest that indole diterpenes are the most potent nonpeptidyl inhibitors of maxi-K channels identified to date. Some of their pharmacological properties could be explained by inhibition of maxi-K channels, although tremorgenicity may be unrelated to channel block.

Effects of fasting, medium glucose, and amino acid concentrations on prolactin and growth hormone release, in vitro, from the pituitary of the tilapia Oreochromis mossambicus.

Previous investigations have shown that the release of PRL and GH from the tilapia pituitary is directly sensitive to osmotic pressure and a variety of endocrine and neuroendocrine factors. The present studies were aimed at determining whether the spontaneous release of PRL and GH, in vitro, is: (1) sensitive to the nutritional status of the fish, and (2) responsive to variations in the D-glucose and total amino acid content of the incubation medium. In the first series of experiments, male fish (50 to 60 g) were divided into two groups. One group was fed twice daily for 2 weeks while the second received no food. A nearly homogeneous mass of PRL-secreting cells was dissected from the rostral pars distalis (RPD) and incubated for 18 to 20 hr in either hyposmotic (300 mOsmolal) or hyperosmotic (355 mOsmolal) medium. Similarly, a mass of GH-secreting cells was dissected from the proximal pars distalis (PPD) and incubated for 18 to 20 hr in isosmotic (320 mOsmolal) medium. Fasting was found to alter the total amount of PRL and GH in the culture well (tissue + medium) at the end of the incubations, decreasing PRL and increasing GH. Fasting was also found to both reduce spontaneous PRL release in vitro and suppress its stimulation by reduced osmotic pressure (P less than 0.01). By contrast, fasting resulted in a substantial increase in spontaneous GH release from the PPD in vitro (P less than 0.01). In the second series of experiments, GH release was found to increase as the D-glucose concentration of the medium decreased (P less than 0.01), while prolactin release was unresponsive.(ABSTRACT TRUNCATED AT 250 WORDS)

The loss of 45Ca2+ associated with prolactin release from the tilapia (Oreochromis mossambicus) rostral pars distalis.

The relationship between tritium 3H-labeled prolactin (PRL) release and the loss of tissue-associated 45Ca2+ was examined in the tilapia rostral pars distalis (RPD) using perifusion incubation under conditions which inhibit or stimulate PRL release. Depolarizing [K+] (56 mM) and hyposmotic medium (280 mOsmolal) increased both the release of [3H]PRL and the loss of 45Ca2+. The responses to high [K+] were faster and shorter in duration than those produced by reduced osmotic pressure. The depletion of Ca2+ from the incubation medium with 2 mM EGTA suppressed the [3H]PRL response evoked by high [K+] or reduced osmotic pressure. Exposing the tissues to Ca(2+)-depleted medium in the absence of high [K+] or reduced osmotic pressure produced a sharp, but brief, increase in 45Ca2+ loss. Cobalt (10(-3) M), a competitive inhibitor of calcium-mediated processes, inhibited the [3H]PRL response to hyposmotic medium and to high [K+]. Cobalt also diminished the increased loss of 45Ca2+ evoked by exposure to reduced osmotic pressure, but was ineffective in altering responses to high [K+]. Methoxyverapamil (D600; 10(-5) M), a blocker of certain voltage-sensitive Ca2+ channels, did not alter either the [3H]PRL or the 45Ca2+ responses to high [K+] and reduced osmotic pressure. Taken together with our earlier studies, the present findings suggest that exposure to high [K+] or hyposmotic medium produces rapid changes in the Ca2+ metabolism of the tilapia RPD that are linked to the stimulation of PRL secretion. Nevertheless, the increased 45Ca2+ loss, but not [3H]PRL release, upon exposure to Ca(2+)-depleted media suggests that Ca2+ loss may not always reflect intracellular events that lead to PRL release.

Effects of osmotic pressure and somatostatin on the cAMP messenger system of the osmosensitive prolactin cell of a teleost fish, the tilapia (Oreochromis mossambicus).

Altered osmotic pressure and somatostatin (SRIF) rapidly alter prolactin (PRL) release from the pituitary gland of the euryhaline teleost, the tilapia. The present studies were undertaken to determine whether altered osmotic pressure and SRIF influence cAMP metabolism in a manner that is correlated with the pattern of PRL release observed previously. Although PRL release is stimulated within 10-20 min when medium osmotic pressure is reduced, cAMP metabolism was not altered. However, following 1 hr of incubation in the presence of IBMX, cAMP accumulation was higher in PRL tissue exposed to medium of reduced osmotic pressure. This suggests that cAMP does not initiate an increase in PRL release in response to reduced osmotic pressure. By contrast, SRIF reduced the forskolin-stimulated increase in cAMP levels in a manner consistent with its rapid effects on PRL release. Moreover, the ability of SRIF to suppress the forskolin-stimulated increase in cAMP levels suggests that SRIF may act to render adenylate cyclase less responsive to direct stimulation by forskolin.

Cortisol rapidly reduces prolactin release and cAMP and 45Ca2+ accumulation in the cichlid fish pituitary in vitro.

During in vitro incubation, prolactin release is inhibited in a dose-related manner by cortisol. This action is mimicked by the synthetic glucocorticoid agonist dexamethasone but not by other steroids tested. Perifusion studies indicate that the inhibition of [3H]prolactin release by cortisol occurs within 20 min. Cortisol (50 nM) also inhibits cAMP accumulation and reduces 45Ca2+ accumulation in the tilapia rostral pars distalis within 15 min. Cortisol's action on prolactin release is blocked in the presence of either the Ca2+ ionophore A23187 or a combination of dibutyryl cAMP and 3-isobutyl-1-methylxanthine, which increase intracellular Ca2+ and cAMP, respectively. Taken together, these findings suggest that cortisol may play a physiologically relevant role in the rapid modulation of prolactin secretion in vivo. Our studies also suggest that the inhibition of prolactin release by cortisol is a specific glucocorticoid action that may be mediated, in part, through cortisol's ability to inhibit intracellular cAMP and Ca2+ metabolism.

Effects of depolarizing concentrations of K+ and reduced osmotic pressure on 45Ca2+ accumulation by the rostral pars distalis of the tilapia (Oreochromis mossambicus).

The accumulation of 45Ca2+ into tilapia prolactin (PRL) tissue was examined under conditions which alter prolactin release. In initial experiments, PRL tissue was incubated in medium containing 12 microCi/ml 45Ca2+ in hyperosmotic medium (355 mOsmolal). Under these conditions, 45Ca2+ accumulated steadily, reaching a plateau within 15-20 min. Subsequent exposure to La3+, which displaces Ca2+ from superficial pools in a wide variety of tissues, rapidly (within 5 min) removed nearly 70% of the 45Ca2+ associated with the tissue. Following this initial removal of 45Ca2+, the level of 45Ca2+ in the PRL tissue remained constant, and is referred to as the La3(+)-resistant pool of Ca2+. This pool of Ca2+ is thought to reflect the entry rate of Ca2+ from extracellular sources. Prolactin tissue exposed to hyposmotic medium or to depolarizing [K+], which stimulates PRL release, significantly increased 45Ca2+ accumulation in this La3(+)-resistant pool. These results indicate that reduced osmotic pressure and depolarization may alter release from tilapia PRL cells, in part, through their ability to increase the entry of extracellular Ca2+.

The tilapia prolactin cell: A model for stimulus-secretion coupling.

The tilapia prolactin (PRL) cell responds rapidly (10-20 min) to small physiological changes in medium osmotic pressure (OP), releasing increasing quantities of hormone as medium OP is reduced. This release is rapidly (≤ 10 min) inhibited by somatostatin (SRIF). There is now extensive evidence that tilapia PRL cell function is regulated through the second messengers Ca(++) and cAMP. Our studies have shown that PRL release is augmented by treatments that lead to increased levels of intracellular Ca(++) or cAMP. On the other hand, PRL release is blocked when tissues are incubated in Ca(++)-depleted medium or upon the addition of Co(++), an inhibitor of Ca(++)-mediated processes. The use of(45)Ca(++) to characterize the movement of Ca(++) into PRL cells has provided evidence that an increase in the influx of extracellular Ca(++) may participate in PRL release upon exposure to hyposmotic medium. Our studies have also shown that SRIF suppresses the increase in(45)Ca(++) accumulation that is brought about when OP is reduced. We have also examined the effects of OP and SRIF on cAMP levels. The reduction of medium OP did not alter cAMP metabolism during 20 min of incubation. By contrast, cAMP accumulation in the presence of IBMX was enhanced at 1 hr of incubation in reduced OP. Thus, an increase in cAMP turnover may play a role in maintaining PRL release under sustained stimulation. SRIF reduced the accumulation of cAMP during 10 min of incubation with IBMX and also reduced the forskolin-stimulated increase in cAMP. Thus, SRIF may suppress adenylate cyclase activity. Finally, our studies have revealed that the forskolin-stimulated increase in cAMP levels is not dependent upon medium Ca(++). The presence of Ca(++) in the medium is required, however, for PRL release even when the cAMP messenger system has been activated. Moreover, cAMP accumulation was augmented when intracellular Ca(++) was increased. This raises the possibility that reduced OP may stimulate an increase in cAMP turnover indirectly through its action(s) on cytosolic Ca(++).

Thyrotropic activity of salmon pituitary glycoprotein hormones in the Hawaiian parrotfish thyroid in vitro.

The thyrotropic activities of salmon pituitary extract, thyroid-stimulating hormone (TSH), gonadotropins (GTH), and glycoprotein fractions obtained during purification of salmon TSH and GTH were measured using the parrotfish thyroid culture system. Purified salmon TSH was approximately 1,000 times more potent than bovine TSH in stimulating thyroxine release into the culture medium. Most of the forms of salmon GTH had no thyrotropic activity. One of the forms of salmon GTH (GTH-F) and three chromatofocusing fractions (CF-B, -C, and -E) that were devoid of activity in the coho salmon in vivo had some thyrotropic activity in the parrotfish thyroid culture. Whether the activity of these fractions was due to contamination with TSH, less potent forms of TSH, or inherent thyrotropic activity of a form of GTH is discussed. These results indicate that the parrotfish thyroid culture system can be used to detect thyrotropic activity of fractions obtained during the purification of teleost TSH.

Somatostatin and altered medium osmotic pressure elicit rapid changes in prolactin release from the rostral pars distalis of the tilapia, Oreochromis mossambicus, in vitro.

Prolactin (PRL) cells in the rostral pars distalis of the tilapia Oreochromis mossambicus respond to somatostatin (SRIF) and reduced medium osmotic pressure within 10-20 min of exposure during perifusion incubation. Pieces of rostral pars distalis tissue were removed from freshwater-adapted tilapia and were preincubated in [3H]leucine in static culture (355 m phi smolal) for 48 hr. Following preincubation, they were placed in the perifusion apparatus and baseline release was established for 3 hr in hyperosmotic medium (355 m phi smolal). Exposure to hyposmotic medium (280 m phi smolal) resulted in a rapid and steep rise in the release of [3H]PRL, which remained elevated for more than 2 hr. When SRIF was added simultaneously with hyposmotic medium, the rise in PRL release normally initiated by reduced osmotic pressure was prevented. Somatostatin also quickly reduced release that had been previously elevated by exposure to hyposmotic medium. The time course of these changes suggests that SRIF and altered osmotic pressure act on PRL secretion in at least partial independence of effects which they may have on PRL synthesis in the tilapia pituitary.

Studies on the regulation of growth hormone release from the proximal pars distalis of male tilapia, Oreochromis mossambicus, in vitro.

The in vitro effects of several factors, including cortisol, somatostatin (SRIF), and medium osmotic pressure, on growth hormone (GH) release from the tilapia pituitary were examined in relation to fish size. Spontaneous GH release from the proximal pars distalis (PPD) of approximately 60-g fish was significantly less than that from tissue of fish weighing either approximately 120 or approximately 280 g when incubated in 340 m phi smolal medium. While GH content of the PPD cultures (tissue + medium measured by densitometry) increased consistently with fish size, GH concentration (per microgram of tissue protein) was variable, being highest in 120-g fish and lowest in 280-g fish. Moreover, GH concentration was not related to GH release. Fish size also appeared to be important in the responsiveness of GH cells to stimulation by cortisol (Nishioka et al., 1985) and by increased osmotic pressure. In cultures of PPD from approximately 60-g fish, in which spontaneous release was relatively low, cortisol and increased medium osmotic pressure significantly enhanced release. Cortisol and hyperosmotic medium were without significant effect, however, on GH release from PPD of approximately 120-g fish, which showed high spontaneous release. In contrast, SRIF, a potent inhibitor of GH secretion, was effective in lowering GH release regardless of fish size. Nevertheless, SRIF was apparently more effective in inhibiting GH release from tissue of 60-g fish than from tissue of 120-g fish. Our data suggest that GH secretion may be augmented when smaller tilapia (approximately 60 g) are transferred to seawater, a situation in which blood cortisol and osmotic pressure would presumably be elevated.

The role of calcium in prolactin release from the pituitary of a teleost fish in vitro.

The release of PRL from the pituitary of a teleost fish, the tilapia (Oreochromis mossambicus), has been previously shown to be dependent on calcium. However, the source(s) and specific action(s) of calcium in the secretory process have not been identified. Also undefined are the mechanisms by which regulators of PRL cell function may alter calcium distribution. In the present investigation, the elevation of medium K+ concentration during static incubations to a depolarizing concentration (56 mM) produced no change in cumulative PRL release over control levels during the 18-20 h of incubation. During perifusion incubation, exposure to high K+ concentrations briefly stimulated (less than or equal to 10 min) and then depressed PRL release. In contrast, reduced medium osmotic pressure elicited a rapid elevation in PRL release that was sustained for 2 h or more. D600, a calcium entry blocker, at 10(-5) M diminished the K+-induced pulses of PRL release. The same concentration, however, did not alter the release of PRL evoked by reduced osmotic pressure. In contrast, CoCl2, which blocks a range of calcium-mediated processes in addition to calcium influx, suppressed PRL release during perifusion and static incubations in hyposmotic medium. These findings suggest that while PRL secretion from the tilapia pituitary is calcium dependent, calcium entry through voltage-regulated plasmalemma channels may not be a prerequisite to the actions of reduced osmotic pressure.

The thyroid gland of the Hawaiian parrotfish and its use as an in vitro model system.

Thyroid tissue in the Hawaiian parrotfish, Scarus dubius, is organized into two discrete lobes, one anterior to the first pair of afferent branchial arteries and the other between the first and second pairs. Lobes or pieces of thyroid lobes from S. dubius were incubated in static or perifusion culture using a simple defined medium. In static culture, this thyroid tissue releases thyroxine (T4) responding to bovine thyrotropin (bTSH) in a dose-related manner during 4- and 24-hr incubations. At 24 hr, however, substantially lower concentrations of bTSH are required to evoke a significant elevation of T4 than at 4 hr. Exposure to bTSH also elicits morphological changes within 4 hr, which include colloid resorption and an increase in the height of the follicular epithelium. During perifusion culture, T4 release rises rapidly within 20 to 30 min following the initiation of exposure to bTSH and remains elevated for between 4 and 8 hr thereafter. In spite of high plasma triiodothyronine (T3) concentrations, the parrotfish thyroid releases no detectable T3 during in vitro culture. This is the first direct evidence in support of the notion that plasma T3 in a teleost fish may be derived principally, perhaps exclusively, from the peripheral monodeiodination of T4.