Changes in the central component of the hypothalamus-pituitary-thyroid axis in a rabbit model of prolonged critical illness.

INTRODUCTION: Prolonged critically ill patients reveal low circulating thyroid hormone levels without a rise in thyroid stimulating hormone (TSH). This condition is labeled "low 3,5,3'-tri-iodothyronine (T3) syndrome" or "nonthyroidal illness syndrome (NTI)" or "euthyroid sick syndrome". Despite the low circulating and peripheral tissue thyroid hormone levels, thyrotropin releasing hormone (TRH) expression in the hypothalamus is reduced and it remains unclear which mechanism is responsible. We set out to study whether increased hypothalamic T3 availability could reflect local thyrotoxicosis and explain feedback inhibition-induced suppression of the TRH gene in the context of the low T3 syndrome in prolonged critical illness. METHODS: Healthy rabbits were compared with prolonged critically ill, parenterally fed animals. We visualized TRH mRNA in the hypothalamus by in situ-hybridization and measured mRNA levels for the type II iodothyronine diodinase (D2), the thyroid hormone transporters monocarboxylate transporter (MCT) 8, MCT10 and organic anion co-transporting polypeptide 1C1 (OATP1C1) and the thyroid hormone receptors alpha (TRalpha) and beta (TRbeta) in the hypothalamus. We also measured the activity of the D2 and type III iodothyronine deiodinase (D3) enzymes. RESULTS: In the hypothalamus of prolonged critically ill rabbits with low circulating T3 and TSH, we observed decreased TRH mRNA, increased D2 mRNA and increased MCT10 and OATP1C1 mRNA while MCT8 gene expression was unaltered as compared with healthy controls. This coincided with low hypothalamic thyroxine (T4) and low-normal T3 concentrations, without a change at the thyroid hormone receptor level. CONCLUSIONS: Although expression of D2 and of the thyroid hormone transporters MCT10 and OATP1C1 were increased in the hypothalamus of prolonged critical ill animals, hypothalamic T4 and T3 content or thyroid hormone receptor expression were not elevated. Hence, decreased TRH gene expression, and hereby low TSH and T3 during prolonged critical illness, is not exclusively brought about by hypothalamic thyrotoxicosis, and infer other TRH suppressing factors to play a role.

Expression of TRH and TRH-like peptides in a human glioblastoma-astrocytoma cell line (U-373-MG).

The human glioblastoma-astrocytoma cell line U-373-MG shows morphological features typical of its neuroectodermal origin. Cells showed positive immunostaining for the glial fibrillary acidic protein. We used this cell culture for studying the putative production of TRH and TRH-related peptides. In a cell extract and conditioned medium, cation and anion exchange chromatography and HPLC revealed the presence of TRH and acidic TRH-like peptides which were identified, at least in part, as pGlu-Glu-ProNH(2). These findings demonstrated that U-373-MG cells are able to produce and release these peptides. Further evidence of TRH synthesis was obtained by amplification using RT-PCR of a 396 bp fragment that corresponds to the TRH precursor mRNA. Our results therefore suggest that the U-373-MG cell line may be a useful model for studying the regulation of TRH and TRH-related peptide production and the interaction of these peptides with other classical neurotransmitter systems. In fact, pilocarpine (a muscarinic cholinergic agonist) enhanced and nicotine (a nicotinic cholinergic agonist) decreased TRH and TRH-related compound production by this cell line. These data also point out that glia may produce substances with neuromodulatory action.

Peptides related to thyrotrophin-releasing hormone (TRH) in human prostate and semen.

The TRH-related peptide, pGlu-Glu-ProNH2, which was first identified in rabbit prostate has recently been named fertilization-promoting peptide (FPP) because of its ability to enhance the in vitro fertilizing potential of mouse epididymal spermatozoa. This study set out to examine the nature of the TRH-related peptides in human prostate and semen but, first, the optimal conditions for collection of semen samples were investigated. FPP was degraded slowly (t1/2 = 163 min, S.E. +/- 51.3, n = 6) in seminal plasma which has allowed us to measure accurately the concentrations of FPP, after extraction of the peptide in acidified acetone precisely 5 min after ejaculation. In this way, high levels of FPP (mean: 49.5 nmol/l) were detected in normal human semen, from young men, although other TRH-related peptides did not appear to be present. We have also examined the TRH-related peptides present in prostate samples from clinical patients both with and without evidence of benign prostatic hyperplasia (BPH), by ion-exchange chromatography followed by radioimmunoassay. Substantial concentrations of FPP were observed in normal (4.10 pmol/g tissue, S.E. +/- 1.46) and BPH prostate (6.27 pmol/g tissue, S.E. +/- 1.65). In addition, a second, neutral TRH-immunoreactive peptide was always detected in BPH tissue (7.40 pmol/g tissue, S.E. +/- 1.98) with only low levels generally present in normal prostate. The possibility that the presence of high levels of the neutral peptide in prostate may be used as an indicator of the onset of BPH deserves further scrutiny.

Processing of proTRH to its intermediate products occurs before the packing into secretory granules of transfected AtT20 cells.

The intracellular compartments where posttranslational processing of proTRH takes place have not been identified. Using AtT20 cells transfected with a complementary DNA for preproTRH, we have used purified antibodies that recognize the intact precursor, intermediate and end products of processing to identify the subcellular compartments in which cleavage occur. Further, pulse-chase experiments followed by subcellular fractionation were undertaken to determine the order of processing of proTRH during its transport to the secretory granules. Cells were homogenized by nitrogen cavitation and subjected to a centrifugation of 1.065 mg/ml density gradient of Percoll to separate secretory granules (SG) from rough endoplasmic reticulum (RER)/Golgi apparatus. The purity of the SG and RER fractions was assessed by assays of marker enzymes for mitochondria, RER, Golgi, and cytoplasm. ProTRH derived cryptic peptides and TRH in each fraction were determined by RIA. Golgi and SG fractions were subjected to polyacrylamide gel electrophoresis followed by extraction and RIA. Using the anti-pCC10 antiserum which recognizes intact (26 kd) as well as partially processed prohormone, the RER/Golgi fraction contained 0.3 pmol intact ProTRH and 0.2 pmol each 15 and 6 kilodalton (kDa) fragments; the SG contained the 15 kDa moiety (0.2 pmol) along with a 6 kDa (0.4 pmol) material but not the 26 kDa ProTRH. The SG were also enriched by 0.21 pmol pYE27 (PreproTRH 25-50), 0.23 pmol pFT (PreproTRH 53-74), 0.31 pmol pEH24 (PreProTRH 86-106), and 0.5 pmol TRH. None of these were present in the RER/Golgi. Pulse-chase studies also showed that the intact proTRH (26 kDa) precursor was only present in the RER/Gg fraction along with two of its N-terminal intermediate processing products, a 15 k mol wt peptide and a 6 k mol wt peptide, and two of its C-terminal processing products, a 16.5 k mol wt and a 9.6 k mol wt peptides. In addition, fully processed peptides as well as TRH were only detected in the neurosecretory granules. These observations suggest that after the initial conversion of proTRH in the RER/Golgi fraction, the peptides are delivered to the granules where processing to TRH and cryptic peptides takes place. Supporting this, our pulse-chase studies unequivocally showed that, pEH24, an end product of proTRH processing, was only produced in secretory granules. Thus, initial cleavage of the TRH precursor may be required for packing and sorting of the end products to occur.

Isolation and characterization of human thyrotropin-releasing hormone (TRH) from an endocrine pancreatic tumor.

An extract of a pancreatic carcinoid tumor obtained at autopsy from a patient who had suffered from Cushing's syndrome was found to have the ability to release thyrotropin (but not any other pituitary hormones) from cultured rat anterior pituitary cells, and to bind to a specific thyrotropin-releasing hormone (TRH) antiserum. The tumor contained 2.2 and 3.9 nmol/g of TRH bio- and immunoreactivity, respectively. The active material was purified and its amino acid composition and chromatographic properties were found to be identical with those of synthetic ovine/porcine TRH. This represents the first isolation of human TRH and the first established case of a 'TRHoma', a TRH-producing tumor.

Normal and adenomatous human pituitaries secrete thyrotropin-releasing hormone in vitro: modulation by dopamine, haloperidol, and somatostatin.

TRH is present in human normal pituitaries and in pituitary adenomas. In this study we demonstrated that the same tissues can release TRH in vitro. Fragments from seven normal pituitaries (10-15 mg/syringe) and dispersed cells from eight prolactinomas, four GH-secreting and two nonsecreting adenomas (1-3 x 10(6) cells/syringe) were perifused using a Krebs-Ringer culture medium. After 1 h of equilibration the perifusion medium was collected every 2 min (1 mL/fraction) for 3 h. TRH, PRL, and GH were measured by RIA under basal conditions and in the presence of 10(-10) to 10(-6) mol/L dopamine (DA), alone or concomitant with haloperidol, or in the presence of 10(-10) or 10(-6) mol/L somatostatin. Both normal pituitary fragments and pituitary adenomatous cells (from all types of adenomas studied) spontaneously released TRH in vitro. TRH was detected in the perifusion medium either immediately after the end of the equilibration period or 30-60 min later. The molecular identity of TRH was assessed by high pressure liquid chromatography. There was no difference in the profile and the rate of TRH secretion between normal and tumoral tissues, and no correlation was found between the level of TRH release and that of PRL or GH secretion. DA stimulated TRH release from normal pituitaries and from PRL- and GH-secreting adenomas at doses as low as 10(-10) mol/L. A concomitant decrease in PRL and GH release was observed from adenomatous cells and in one case of normal tissue. Haloperidol (10(-7) mol/L) antagonized the effect of 10(-8) mol/L DA on both TRH and PRL secretion in normal pituitary and in prolactinomas. DA had no effect on TRH release from two nonsecreting tumors. The amounts of TRH released during 1 h of perifusion were 60-1640 pg/2 mg wet wt tissue in normal pituitaries and 54-2174 pg/10(6) cells in adenomas; these values were very high compared to those precedently reported within the tissues. These results indicate that pituitary cells can release TRH in vitro and suggest that TRH might be synthesized in situ. We suggest that TRH could act on pituitary hormone secretion and/or cell proliferation via a paracrine and/or an autocrine mechanism.

Thyrotrophin-releasing hormone-related polypeptides in rabbit prostate and semen are different from those in rabbit hypothalamus.

TRH-related peptides were extracted from the hypothalamus and prostate gland of the rabbit. The peptides were fractionated by gel exclusion chromatography and located by trypsin digestion and radioimmunoassay with antibodies to TRH amide and TRH-Gly Lys. In the hypothalamus TRH-related peptides containing approximately 16 and 30 residues were observed: in these peptides the extensions to the TRH sequence were exclusively in the C-terminal direction. In addition, the three-residue form of TRH was also present. In the prostate complex, the predominant TRH-related peptide contained approximately 50 residues and the extension to the TRH tripeptide was on the N-terminal side; a three-residue form of immunoreactive TRH was also demonstrated. The same pattern of TRH-related peptides was shown to be present in rabbit semen. The results reveal the existence of a novel TRH-related polypeptide in the prostate and semen which does not occur in the hypothalamus. This peptide appears to undergo secretion.

Analysis of pancreatic peptide hormones by reversed-phase high-performance liquid chromatography.

Reversed-phase high-performance liquid chromatography (HPLC) is examined as a method for separating pancreatic peptides. The method was based on gradient elution with acetonitrile in an acid phosphate buffer (pH 3.10). Apart from human and porcine insulin all the other peptide standards tested (thyrotropin-releasing factor, vaso-active intestinal polypeptide, human C-peptide, porcine C-peptide, somatostatin, porcine glucagon, porcine proinsulin and porcine pancreatic polypeptide) could be separated simultaneously in 40 minutes with a binary gradient composed of five linear segments and increasing from 0 to 60% acetonitrile. Human and porcine insulin could be almost completely resolved by a minimal reduction in the steepness of the acetonitrile gradient. Repeated injections of human C-peptide and porcine insulin resulted in a coefficient of variation of less than 1.5% in the retention times. The use of 125I-labelled peptides gave recoveries exceeding 90%. HPLC of acid ethanol extracts of autopsy pancreases from three infants showed that the immunoreactivity of the peptides measured remained unaffected by the chromatography. Both immunoreactive C-peptide and immunoreactive insulin (IRI) were recovered in two peaks, the second common peak representing proinsulin and amounting to 6.5 to 8.4% of total IRI. Immunoreactive glucagon was eluted in a single peak. Chromatography of plasma extracts from two infants of diabetic mothers demonstrated that proinsulin accounted for 59-63% of total IRI, while insulin was separated into two peaks corresponding to the standards of human insulin and porcine insulin. These results indicate that reversed -phase HPLC is a method with a good reproducibility and a high recovery applicable to the rapid and effective separation of pancreatic peptides from biological extracts.

Luteinizing hormone-releasing hormone and thyrotropin-releasing hormone in the hypothalamus of women: effects of age and reproductive status.

In the study of the effects of age and reproductive status on LHRH and TRH content in the hypothalamus of women, we found that the amount of LHRH (58 +/- 5.5 ng; mean +/- SE) in the hypothalamus of young women (16-29 yr) was significantly greater (P less than 0.001) than that (28 +/- 3.0 ng) in postmenopausal women (50-78 yr). The hypothalamic content of LHRH (18 +/- 2.4 ng) of bilaterally ovariectomized women (39-47 yr) was significantly less (P less than 0.001) than that (60 +/- 12.6 ng) in younger ovulatory women (30-39 yr) or that (56 +/- 13.5 ng) in ovulatory women of comparable age (40-49 yr). In contrast, the hypothalamic content of TRH (121.4 +/- 32.8 ng) in postmenopausal women were similar to that (122.3 +/- 12.5 ng) in young women. Although aging in women is associated with a significant reduction in the amount of LHRH in the hypothalamus, such a reduction appears to be a consequence of ovarian failure and not of aging per se.

Hormonal control of prostatic thyrotropin-releasing hormone (TRH) testosterone modulates prostatic TRH concentrations.

The presence of extremely high concentrations of authentic TRH in the rat prostate prompted us to examine whether prostatic TRH concentrations were under hormonal control. Both the immunoreactive TRH content per prostate and TRH immunoreactivity expressed per 100 mg prostatic protein were lower in animals 2 weeks after hypophysectomy than in sham-operated and calorically restricted weight-matched controls. Since many of the prostatic functions are testosterone dependent, we assessed the possibility that testosterone modulated prostatic TRH concentrations. We measured prostatic TRH concentrations in the following four groups of sexually mature male Sprague-Dawley rats: group I, sham operated; group II, castrated; group III, castrated animals with 5-mm Silastic testosterone implants; and group IV, castrated animals with 20-mm testosterone implants. Prostatic TRH concentrations in these four groups 2 weeks after surgery were 600.5 +/- 33.3, 65.1 +/- 22.6, 169.4 +/- 55.3, and 609.5 +/- 144.3 (+/-SE) ng/100 mg protein. There was good linear correlation between prostatic TRH concentrations and serum testosterone concentrations (r = 0.65; P less than 0.01). By subjecting the pooled prostatic extracts from each group to ion exchange chromatography on a SP-Sephadex C-25 column and measuring the proportion of immunoreactivity coeluting with authentic TRH, it was shown that the fall in prostatic TRH immunoreactivity after castration and hypophysectomy was indeed due to a loss of authentic TRH. We conclude that the prostatic TRH concentrations are under hormonal control and appear to be modulated by serum testosterone concentrations. This is the first demonstration of hormonal regulation of a neuropeptide in a mammalian extrahypothalamic site and suggests a physiological role for this neuropeptide at this site.

Thyrotropin-releasing hormone immunoreactivity in the gastrointestinal tract of man.

TRH inhibits gastric secretion, gastric motility, and pancreatic enzyme secretion in man. We measured TRH immunoreactivity (TRH-IR) in mucosal and transmural biopsies from different parts of the stomach, pylorus, duodenum, and pancreas in patients undergoing surgery or gastroscopy. TRH-IR was found in pancreas (range, 1.64-3.27 pg TRH/mg protein) and in mucosa of the fundus (13.3 +/- 2.0 pg TRH/mg protein), antrum (22.5 +/- 4.5 pg TRH/mg protein), and duodenum (35.5 +/- 6.4 mg TRH/mg protein). Low levels of TRH-IR were found in transmural biopsies from fundus, antrum, and pylorus (range, 0.45-0.87 pg TRH/mg protein). Chromatography of gastrointestinal tissue extracts on a cation exchange column resulted in two different TRH-IR peaks, one eluted at pH 3.5 and the other eluted at pH 7.4. The peak at pH 7.4 corresponded to synthetic TRH. The presence and the action of TRH in human gut indicate a physiological role of TRH in the gastrointestinal tract in man.

Luteinizing hormone-releasing hormone and thyrotropin-releasing hormone in human and bovine milk.

Two hypothalamic peptide hormones, luteinizing hormone-releasing hormone (LHRH) and thyrotropin-releasing hormone (TRH), have been isolated from human milk and bovine colostrum. Acidified methanolic extracts, prepared from human milk, bovine colostrum and rat hypothalami, as well as synthetic LHRH and TRH markers were subjected to high-pressure liquid chromatography (HPLC). The eluates were tested for the presence of LHRH and TRH by specific radioimmunoassays. It was found that milk extracts contain significant amounts of LHRH (3.9 - 11.8 ng/ml) and TRH (0.16 - 0.34 ng/ml), which comigrate with the corresponding marker hormones and with those of hypothalamic origin. The HPLC-purified LHRH from both human and bovine milk was bioactive in a dose-response manner similar to synthetic LHRH.

Thyrotropin-releasing hormone-like bioactivity in placenta: evidence for the existence of substances other than Pyroglu-His-Pro-NH2 (TRH) capable of stimulating pituitary thyrotropin release.

The purpose of this study was to investigate the chemical nature of the TRH-like bioactivity and immunoreactivity in extracts of human placenta. Methanolic extracts of human placenta contained nearly 36 ng TRH-like bioactivity/g dried tissue. The immunoreactivity of this extract was only 3 ng/g dried tissue. Two molar acetic acid extracts of human placenta yielded 9 ng TRH-like bioactivity/g dried tissue. The immunoreactivity of these acetic acid extracts, however, was 5.5 ng/g dried tissue. When these placental extracts were subjected to gel filtration chromatography, the bioactivity was found to reside in two fractions which were distinct from synthetic TRH. Furthermore, the immunoreactivity present in these placental extracts was also chromatographically distinct from that of synthetic TRH. In conclusion, these experiments confirm the presence of substantial quantities of materials in placenta which possess TRH-like immunoreactivity and bioactivity. These results also argue that the immunoreactive fractions are not bioactive and provide firm evidence that neither the TSH-releasing substances nor the TRH-like immunoreactivity found in human placenta are identical to pyroglu-His-Pro-NH2.

[Radioimmunological aspects of neuropeptides]. / Aspects radioimmunologiques des neuropeptides.

Immunological aspects of neurosecretory peptides are relevant to antibody generation, labelling of peptides for radioimmunoassays and in particular the metabolism of neuropeptides. Antibodies were generally produced by adsorption or polymerization (using different approaches and various coupling agents) of peptides with intrinsic antigenic molecules preceding their administration to recepient animals (rats, guinea pigs, rabbits and sheep) with adjuvants. Antibodies were also obtained by injecting peptides alone with or without adjuvants. However, independent of the recipient species, the highest specific or nonspecific antibody titers were obtained by the former method. Up to now, only 3H-peptides or 125I-peptides or derivatives were used as tracers in RIAs. Highest sensitivities were generally obtained when using 125I-peptides. However, 125I-peptides may be adsorbed ("false titers") to some antiserum with sufficient specificity in terms of displacement by structural analogues of the peptide but with a loss of sensitivity. Finally, large molecular forms, cross-reacting with the antibodies prepared with different methods and peptidasic activities must be taken into account. These topics are presented with data on LHRH, TRH and substance P derived from our experiments and from those of other investigators.

[Hypothalamic peptide hormones]. / Die Peptidhormone des Hypothalamus.

After a review of the significance of the releasing and inhibiting polypeptide hormones of the hypothalamus, the clinical significance of the gonadotropin-releasing-hormone and the thyrotropin releasing hormone, is discussed in detail.