Prolactin secretion from human prolactinomas perifused in vitro: effect of TRH, prostaglandin E1, theophylline, dopamine and dopamine receptor blockers

The expression of thyrotrophin-releasing hormone receptor 1 messenger ribonucleic acid in human pituitary adenomas

OBJECTIVE

Thyrotrophin-releasing hormone (TRH) paradoxically induces the release of growth hormone (GH) when injected intravenously into acromegalic patients, although the mechanism of this action is unknown at present. Several research groups have reported that the level of TRH receptor-1 (TRHR-1) mRNA expression is variable in pituitary adenomas, and does not correlate with the degree of paradoxical GH response to TRH administration in a limited number of acromegalic patients. We aimed to compare the expression levels of TRHR-1 mRNA among various types of pituitary adenoma and to clarify whether these levels correlate with the degree of pituitary hormone response to TRH.

PATIENTS

Pituitary adenoma tissue was obtained by surgery from 14 patients with acromegaly, four with prolactinomas, nine with nonfunctioning adenomas and one with a TSH-producing adenoma.

METHODS

The level of human TRHR-1 mRNA expression in each adenoma was quantified using the competitive reverse transcription polymerase chain reaction (RT-PCR) METHOD: For amplification of a TRHR-1 cDNA fragment, a sense primer was designed according to the sequence in exon 2 and an antisense primer designed according to the sequence located at the region in exon 3 that does not encode for the alternative splicing-generated short form of TRHR-1 mRNA.

RESULTS

TRHR-1 mRNA was detected in all pituitary adenomas examined and did not correlate with their size. The mean level of TRHR-1 mRNA expression was significantly lower in GH-producing adenomas than in prolactinomas and nonfunctioning adenomas (1.4 +/- 0.4 x 10(-2) attomol/microg total RNA, 10.7 3.4 x 10(-2) attomol/microg total RNA, and 7.2 +/- 3.3 x 10(-2) attomol/g total RNA, respectively). The ratio of plasma peak GH induced by TRH administration to the basal level of plasma GH in the patients with acromegaly correlated positively with the level of TRHR-1 mRNA expression in their GH-producing adenomas (r = 0.620, P = 0.0179). The responsiveness of plasma PRL and gonadotrophin to TRH in the patients with prolactinoma and nonfunctioning pituitary adenoma did not significantly correlate with the levels of TRHR-1 mRNA expression in their pituitary adenomas, respectively.

CONCLUSIONS

The findings of the present study suggest that the level of TRHR-1 mRNA expression varies among different types of pituitary adenoma. Furthermore, in acromegaly, the responsiveness of plasma GH to TRH administration appears to at least partially depend on the level of TRHR-1 mRNA expression in the GH-producing pituitary adenoma.

Human TRH receptor messenger ribonucleic acid levels in normal and adenomatous pituitary: analysis by the competitive reverse transcription polymerase chain reaction method

OBJECTIVE

Little is known about the mechanism of diversity in in-vivo hormonal responsiveness to TRH in patients with functional pituitary adenomas. In order to clarify the relation between the responsiveness to TRH and TRH receptor messenger ribonucleic acid (mRNA) expression, we attempted to measure TRH receptor mRNA levels in human pituitary adenoma tissues by competitive reverse transcription polymerase chain reaction (RT-PCR) method.

PATIENTS

Pituitary tissue samples were obtained at autopsy from 5 patients without pituitary disease. Pituitary adenoma tissue samples were obtained at surgery from 18 patients with pituitary adenoma (4 non-functioning, 8 prolactinoma, 4 acromegaly, 1 Cushing's disease and 1 FSH producing adenoma).

METHODS

Partial TRH receptor cDNA from a human GH producing adenoma cDNA library was amplified by PCR under low stringency conditions using primers encoding the transmembrane domains III and VI of pituitary TRH receptor cDNA. The partial sequence of the amplified cDNA determined by a dideoxy-chain termination method was identical to the corresponding sequence of human TRH receptor cDNA. A competitor was generated by deleting the inner 111 bp from the amplified TRH receptor cDNA and subcloning. RNA extracted from human pituitary was reverse transcribed and co-amplified with competitor by PCR under higher stringency conditions. The TRH receptor mRNA levels, expressed as the relative intensity against the amplified levels of competitor, were compared among various pituitary tissues.

RESULTS

The relative TRH receptor mRNA levels of pituitary tissues in patients without pituitary disease were detectable and variable (M +/- SD) (0.370 +/- 0.231, n = 5), and slightly but not significantly lower than those in patients with pituitary tumours (0.598 +/- 0.265, n = 18). In patients with prolactinoma, the relative levels of TRH receptor mRNA were quite variable (0.02-1.170, 0.604 +/- 0.358, n = 8) and not correlated with PRL responsiveness to TRH (responder 0.457; non-responder 0.340-0.950). In patients with acromegaly, TRH receptor mRNA was detectable not only in the paradoxical GH responder to TRH (0.718) but also in the non-responder (0.758 and 0.765). In one patient with Cushing's disease, a relatively low level of TRH receptor mRNA could be detected (0.415). In the patient with a FSH producing tumour whose plasma FSH did not respond to TRH, a small amount of TRH receptor mRNA was detectable (0.447).

CONCLUSIONS

In patients with functioning pituitary adenomas, hormonal responsiveness to TRH in vivo might not be assessable by TRH receptor mRNA levels in the adenoma cells.

Thyrotropin-releasing hormone and gonadotropin-releasing hormone-associated peptide modulation of [Ca2+]i in human lactotrophs

The effect of thyrotropin-releasing hormone (TRH) and gonadotropin-releasing hormone-associated peptide (GAP) was studied on both secretion and intracellular free Ca2+ concentrations ([Ca2+]i) in human pituitary cells cultured from prolactin (PRL)-secreting tumors. Secretion was measured during a 30-min incubation period and we used a microspectrofluorimetric method in individual cells and indo-1 as the fluorescent probe. TRH (10(-8) M) significantly increased PRL release in five out of the six cell populations. In these five cases, more than 68% of individual cells responded to TRH by an increase in [Ca2+]i. No significant increase in PRL secretion was found in another culture in which TRH increased [Ca2+]i in only 37% of the cells. The effect of GAP (10(-7) M) was studied in five cell populations. In three of them, a decrease of 20% to 51% of the PRL basal secretory rate was observed under GAP. GAP inhibited [Ca2+]i in respectively 59%, 46% and 94% of the cells from these cultures. The inhibitory effect of GAP was blocked by a pertussis toxin (PT) pretreatment which demonstrates the involvement of a PT-sensitive G-protein in GAP action. In two other cultures, GAP did not significantly alter PRL secretion or individual cell [Ca2+]i. These observations suggest that GAP might play a role in the control of PRL secretion in the human.

Studies of two thyrotrophin-secreting pituitary adenomas: evidence for dopamine receptor deficiency

Of 22 previously reported patients with TSH-secreting pituitary adenomas challenged with dopamine agonists, 18 showed no decrease in serum TSH. There have been few in-vitro studies of these rare tumours so the mechanism of the dopaminergic resistance has remained obscure. We describe two further patients with thyrotrophinomas; the first was thyrotoxic (T3 6.1 nmol/l, TSH 7 mU/l) and the second was diagnosed after radioiodine for presumed Graves' disease. The second patient had an alpha-subunit: TSH molar ratio less than unity (0.27). In-vivo TSH responses to TRH, bromocriptine and domperidone were compared with those of the resected tumour cells in vitro, the latter studied using a continuous perifusion system. Dopamine receptors were sought in membranes from each tumour using a radioreceptor assay employing 3H-spiperone. Patient 1 showed significant increases in serum TSH (7 to 13 mU/l) and alpha-subunit (18.7 to 385 ng/ml) after 200 micrograms TRH (i.v.) but patient 2 showed no such increases (TSH: 69 to 72 mU/l, alpha-subunit: 4.9 to 5.2 ng/ml). Neither patient showed a change in serum TSH following bromocriptine 2.5 mg (orally) or domperidone 10 mg (i.v.), though serum PRL responded normally. Serum TSH from patient 1 was of apparently normal molecular size but increased bioactivity (B/I ratio 3.8) and that from patient 2 was of increased molecular size but reduced bioactivity (B/I ratio 0.1). Tumour cells from each patient immunostained for TSH beta and alpha-subunit, and secreted TSH in vitro. The first showed dose-dependent TSH release after TRH (1-100 ng/ml) which could not be inhibited by dopamine (5 mumol/l) but the second was unresponsive to TRH in vitro. Neither tumour showed inhibition of TSH release by dopamine (5 mumol/l) or bromocriptine (0.01-10 nmol/l) and neither contained membrane-bound dopamine receptors. The results suggest that the dopaminergic resistance typical of most TSH-secreting pituitary adenomas may be due to altered or absent membrane-bound dopamine receptors.

Perifusion studies of bromocriptine-treated and untreated macroprolactinomas: effects of dopamine, bromocriptine and TRH

There have been no detailed in-vitro studies of PRL secretion by human macroprolactinoma cells exposed to bromocriptine (BC) to within a few days of surgical removal. We have studied cells from four such tumours (serum PRL 7.05-247 U/l) and six untreated tumours (serum PRL 4-80.35 U/l) using a perifusion technique. The BC-treated tumours had shown tumour shrinkage and were treated until 40-96 h before surgery, but in one patient serum PRL had not suppressed below 15 U/l despite chronic treatment. Pretreatment serum PRL responses to TRH were blunted in all 10 patients. During perifusion with dopamine (DA, 5 mumol/l) untreated prolactinomas had a higher PRL secretion rate (19.3 +/- 2.7 microU/mg tissue/min, mean +/- SEM) than BC-treated (3.9 +/- 0.7, P = 0.005). When DA was removed, PRL secretion from untreated tumours increased to 129.7 +/- 18.7 microU/mg/min, but in three of the BC-treated, little increase occurred. In the fourth (from the patient whose serum PRL had not fully suppressed) PRL secretion increased from 4.4 to 25.6 microU/mg/min after DA withdrawal, and DA and BC dose-related inhibition of PRL was similar to that observed in untreated tumours. TRH (10 ng/ml), without DA, provoked increased PRL release from both untreated (266% basal secretion, n = 3) and BC-treated (298%, n = 3) tumours; this effect was completely inhibited by DA (5 mumol/l). The absence of hormones other than PRL following potassium (55 mmol/l) excluded contaminating normal pituitary. We conclude: (1) The effects of BC on prolactinoma PRL secretion may persist for at least 4 days; (2) partial in-vivo BC resistance can be due to factors other than DA receptor malfunction; (3) the apparent discrepancy between in-vivo and in-vitro TRH responses was consistent with the presence of increased hypothalamic DA tone in vivo; and (4) BC may have differential effects on TRH and DA-controlled PRL pools in the tumourous lactotroph.