Abdominal musculature segmentation and surface prediction from CT using deep learning for sarcopenia assessment.

PURPOSE: The purpose of this study was to build and train a deep convolutional neural networks (CNN) algorithm to segment muscular body mass (MBM) to predict muscular surface from a two-dimensional axial computed tomography (CT) slice through L3 vertebra. MATERIALS AND METHODS: An ensemble of 15 deep learning models with a two-dimensional U-net architecture with a 4-level depth and 18 initial filters were trained to segment MBM. The muscular surface values were computed from the predicted masks and corrected with the algorithm's estimated bias. Resulting mask prediction and surface prediction were assessed using Dice similarity coefficient (DSC) and root mean squared error (RMSE) scores respectively using ground truth masks as standards of reference. RESULTS: A total of 1025 individual CT slices were used for training and validation and 500 additional axial CT slices were used for testing. The obtained mean DSC and RMSE on the test set were 0.97 and 3.7 cm2 respectively. CONCLUSION: Deep learning methods using convolutional neural networks algorithm enable a robust and automated extraction of CT derived MBM for sarcopenia assessment, which could be implemented in a clinical workflow.

Diagnosis of focal liver lesions from ultrasound using deep learning.

PURPOSE: The purpose of this study was to create an algorithm that simultaneously detects and characterizes (benign vs. malignant) focal liver lesion (FLL) using deep learning. MATERIALS AND METHODS: We trained our algorithm on a dataset proposed during a data challenge organized at the 2018 Journées Francophones de Radiologie. The dataset was composed of 367 two-dimensional ultrasound images from 367 individual livers, captured at various institutions. The algorithm was guided using an attention mechanism with annotations made by a radiologist. The algorithm was then tested on a new data set from 177 patients. RESULTS: The models reached mean ROC-AUC scores of 0.935 for FLL detection and 0.916 for FLL characterization over three shuffled three-fold cross-validations performed with the training data. On the new dataset of 177 patients, our models reached a weighted mean ROC-AUC scores of 0.891 for seven different tasks. CONCLUSION: This study that uses a supervised-attention mechanism focused on FLL detection and characterization from liver ultrasound images. This method could prove to be highly relevant for medical imaging once validated on a larger independent cohort.

Detection and characterization of MRI breast lesions using deep learning.

PURPOSE: The purpose of this study was to assess the potential of a deep learning model to discriminate between benign and malignant breast lesions using magnetic resonance imaging (MRI) and characterize different histological subtypes of breast lesions. MATERIALS AND METHODS: We developed a deep learning model that simultaneously learns to detect lesions and characterize them. We created a lesion-characterization model based on a single two-dimensional T1-weighted fat suppressed MR image obtained after intravenous administration of a gadolinium chelate selected by radiologists. The data included 335 MR images from 335 patients, representing 17 different histological subtypes of breast lesions grouped into four categories (mammary gland, benign lesions, invasive ductal carcinoma and other malignant lesions). Algorithm performance was evaluated on an independent test set of 168 MR images using weighted sums of the area under the curve (AUC) scores. RESULTS: We obtained a cross-validation score of 0.817 weighted average receiver operating characteristic (ROC)-AUC on the training set computed as the mean of three-shuffle three-fold cross-validation. Our model reached a weighted mean AUC of 0.816 on the independent challenge test set. CONCLUSION: This study shows good performance of a supervised-attention model with deep learning for breast MRI. This method should be validated on a larger and independent cohort.

The pituitary V3 vasopressin receptor and the corticotroph phenotype in ectopic ACTH syndrome.

Ectopic ACTH secretion occurs in highly differentiated and rather indolent tumors like bronchial carcinoids or, in contrast, in various types of aggressive and poorly differentiated neuroendocrine tumors. We explored this phenomenon using the recently cloned human pituitary V3 vasopressin receptor as an alternate molecular marker of the corticotroph phenotype. Expression of V3 receptor, corticotrophin releasing hormone (CRH) receptor, and proopiomelanocortin (POMC) genes was examined in tumors of pituitary and nonpituitary origin. A comparative RT-PCR approach revealed signals for both V3 receptor and CHR receptor mRNAs in 17 of 18 ACTH-secreting pituitary adenomas, and 6 of 6 normal pituitaries; in six growth hormone- or prolactin-secreting adenomas, a very faint V3 receptor signal was observed in three cases, and CRH receptor signal was undetected in all. Six of eight bronchial carcinoids responsible for the ectopic ACTH syndrome had both POMC and V3 receptor signals as high as those in ACTH-secreting pituitary adenomas; in contrast, no POMC signal and only a very faint V3 receptor signal were detected in six of eight nonsecreting bronchial carcinoids. Northern blot analysis showed V3 receptor mRNA of identical size in ACTH-secreting bronchial carcinoids and pituitary tumors. Other types of nonpituitary tumors responsible for ectopic ACTH syndrome presented much lower levels of both POMC and V3 receptor gene expression than those found in ACTH-secreting bronchial carcinoids. In contrast with the V3 receptor, CRH receptor mRNA was detected in the majority of neuroendocrine tumors irrespective of their POMC status. These results show that expression of the V3 receptor gene participates in the corticotroph phenotype. Its striking association with ACTH-secreting bronchial carcinoids defines a subset of nonpituitary tumors in which ectopic POMC gene expression is but one aspect of a wider process of corticotroph cell differentiation, and opens new possibilities of pharmacological investigations and even manipulations of this peculiar ACTH hypersecretory syndrome.

Gastric electrical stimulation modulates hypothalamic corticotropin-releasing factor-producing neurons during post-operative ileus in rat.

High-frequency/low-energy gastric electrical stimulation (GES) is an efficient therapy to treat gastric emptying-related disorders but its mechanism of action remains poorly understood. We aimed to assess the effects of high-frequency/low-energy GES on corticotropin-releasing factor (CRF)-producing neurons in the paraventricular nucleus of the hypothalamus (PVN), which are involved in gastric ileus induced by laparotomy. Two electrodes were implanted in the rat gastric antrum during laparotomy, then stimulation (amplitude: 2 mA; pulse duration 330 micros; frequency: 2 Hz; 1 min ON/2 min OFF) or sham stimulation (control group) were applied. Using immunohistochemistry, the number of c-Fos protein-expressing neurons (c-Fos protein-immunoreactive cells, Fos-IR) was quantified in the PVN after 1 h of stimulation. The number of neurons expressing simultaneously c-Fos protein and CRF mRNA was measured by means of immunocytochemistry combined with in situ hybridization. Finally, c-Fos and CRF mRNA levels in the hypothalamus were determined by in situ hybridization or quantitative reverse transcriptase-polymerase chain reaction. Fos-IR in the PVN was significantly decreased 1 h after GES (P<0.05) but was not affected by sub-diaphragmatic vagotomy. The number of neurons containing c-Fos protein and CRF mRNA was lower in the GES group compared with the control group (P<0.05). In addition, c-Fos and CRF mRNA levels in the PVN were significantly decreased by GES (P

Predictors of responses to immune checkpoint blockade in advanced melanoma.

Immune checkpoint blockers (ICB) have become pivotal therapies in the clinical armamentarium against metastatic melanoma (MMel). Given the frequency of immune related adverse events and increasing use of ICB, predictors of response to CTLA-4 and/or PD-1 blockade represent unmet clinical needs. Using a systems biology-based approach to an assessment of 779 paired blood and tumor markers in 37 stage III MMel patients, we analyzed association between blood immune parameters and the functional immune reactivity of tumor-infiltrating cells after ex vivo exposure to ICB. Based on this assay, we retrospectively observed, in eight cohorts enrolling 190 MMel patients treated with ipilimumab, that PD-L1 expression on peripheral T cells was prognostic on overall and progression-free survival. Moreover, detectable CD137 on circulating CD8+ T cells was associated with the disease-free status of resected stage III MMel patients after adjuvant ipilimumab + nivolumab (but not nivolumab alone). These biomarkers should be validated in prospective trials in MMel.The clinical management of metastatic melanoma requires predictors of the response to checkpoint blockade. Here, the authors use immunological assays to identify potential prognostic/predictive biomarkers in circulating blood cells and in tumor-infiltrating lymphocytes from patients with resected stage III melanoma.

Pituitary adenylate cyclase-activating polypeptide directly modulates the activity of proopiomelanocortin neurons in the rat arcuate nucleus.

Pituitary adenylate cyclase-activating polypeptide (PACAP) and the proopiomelanocortin (POMC)-derived peptide alpha-melanocyte-stimulating hormone (alpha-MSH) both regulate multiple neuroendocrine functions and feeding behavior. Two subtypes of PACAP receptor mRNAs, pituitary adenylate cyclase-activating polypeptide-specific receptor (PAC1-R) and pituitary adenylate cyclase-activating polypeptide/vasoactive intestinal polypeptide mutual receptor (VPAC2-R), are actively expressed in the arcuate nucleus of the hypothalamus, where POMC cell bodies are located. This observation led us to investigate the possible regulatory action of PACAP on rat POMC neurons. Double-labeling in situ hybridization histochemistry revealed that approximately 50% of POMC-producing neurons express PAC1-R and/or VPAC2-R mRNAs. The proportion of POMC neurons that also contain PAC1-R mRNA was homogeneous along the rostro-caudal axis of the arcuate nucleus while POMC-positive cell bodies expressing the VPAC2-R subtype were more abundant in the rostral region. Incubation of mediobasal hypothalamic explants with PACAP (10(-7) M; 30 min) increased POMC mRNA expression, and this effect was blocked by PACAP6-38 (10(-6) M). In contrast, incubation with vasoactive intestinal polypeptide (10(-7) M) did not affect POMC mRNA level. Incubation of hypothalamic fragments with PACAP (10(-7) M) caused a significant increase in alpha-MSH content in the tissue and in the incubation medium. Altogether, the present results reveal that exogenous PACAP, acting probably through PAC1-R, regulates the activity of POMC neurons in the rat hypothalamus. These data suggest that the effects of PACAP on the gonadotropin-releasing hormone neuroendocrine axis and the regulation of feeding behavior may be mediated, at least in part, through modulation of POMC neurons.

Specific expression of the urotensin II gene in sacral motoneurons of developing rat spinal cord.

The neuropeptide urotensin II (UII) is expressed in motoneurons of the brainstem and spinal cord in adults. Here, the expression pattern of the UII gene was studied in the developing rat spinal cord. UII mRNA was detected by reverse-transcription-polymerase chain reaction (RT-PCR) as early as E10. From E14 to E21, in situ hybridization revealed intense expression of the UII gene specifically in sacral motoneurons, while only faint expression was detected at cervical and thoracic levels. After birth (P0, P4), the expression of UII mRNA increased in motoneurons at all rostrocaudal levels. Thus, UII is the first gene reported to show expression limited to the sacral pool of motoneurons, which are known to have particular properties in terms of targets and programmed cell death.

Effect of prolyl endopeptidase inhibition on arginine-vasopressin and thyrotrophin-releasing hormone catabolism in the rat brain.

Compound S 17092 is a potent and selective inhibitor of prolyl endopeptidase (EC 3.4.21.26, PEP) that may be of therapeutic value for the treatment of memory impairment associated with neurodegenerative diseases. In the present study, we investigated the effects of S 17092 on the catabolism of the promnesic neuropeptides thyrotrophin-releasing hormone (TRH) and arginine-vasopressin (AVP) in the rat brain. In vitro, bacterial PEP hydrolysed both TRH and AVP, and the breakdown of the two peptides was almost completely prevented by 10(-5) M S 17092. In vivo, a single oral administration of S 17092 provoked a significant increase in TRH-like immunoreactivity (TRH-LI) in the cerebral cortex (+63% for a 10 mg/kg dose and +72% for a 30 mg/kg dose), as well as AVP-LI in the hippocampus (+54% for a 30 mg/kg dose), but did not affect TRH-LI in the amygdala nor AVP-LI in the cerebral cortex. Chronic administration of S 17092 (10 or 30 mg/kg daily) lead to a significant increase in THR-LI in the cerebral cortex (+55% and +56%, respectively), but did not modify AVP-LI in the hippocampus, nor in the cerebral cortex. These results show that the selective PEP inhibitor S 17092 increases TRH and AVP content in discrete regions of the rat brain. The present data suggest that the promnesic and antiamnesic effects of S 17092 can be accounted for, at least in part, by blockage of AVP and TRH degradation by PEP.

In situ hybridization localization of TRH precursor and TRH receptor mRNAs in the brain and pituitary of Xenopus laevis.

We examined the distribution of the mRNAs encoding proTRH and the three TRH receptor subtypes (xTRHR1, xTRHR2, and xTRHR3) in the Xenopus laevis CNS and pituitary. A positive correlation was generally observed between the expression patterns of proTRH and xTRHR mRNAs. xTRHRs were widely expressed in the telencephalon and diencephalon, where two or even three xTRHR mRNAs were often simultaneously observed within the same brain structures. In the pituitary, xTRHR2 was selectively expressed in the distal lobe, and xTRHR3 was found exclusively in the intermediate lobe of white background-adapted animals, indicating that, in amphibians, the effect of TRH on alpha-melanotropin (alpha-MSH) secretion from melanotrope cells is mediated through the novel receptor subtype xTRHR3.

Frog chromogranin A messenger ribonucleic acid encodes three highly conserved peptides. Coordinate regulation of proopiomelanocortin and chromogranin A gene expression in the pars intermedia of the pituitary during background color adaptation.

Chromogranin A (CgA) is a neuroendocrine secretory protein that is widely used as a marker for endocrine neoplasms but whose function is not completely understood. In mammals, it is thought that CgA is a precursor for biologically active peptides. Here, we describe the cloning of a complementary DNA encoding CgA from a nonmammalian vertebrate, the frog Rana ridibunda. Sequence analysis revealed that frog CgA exhibits only 40-44% amino acid sequence similarity with its mammalian homologues. The amino acid identity is confined to three regions (70-80% identity) of the protein that are flanked by conserved pairs of basic amino acid residues, suggesting that proteolytic processing at these cleavage sites may give rise to three biologically active peptides whose sequences have been highly preserved during evolution. Tissue distribution analysis by Northern blot and in situ hybridization revealed the widespread expression of frog CgA messenger RNA in the brain and in endocrine tissues, the highest concentration occurring in the distal lobe of the pituitary. Adaptation of frog skin color to a dark background caused a concomitant increase in CgA and POMC messenger RNA levels in the intermediate lobe of the pituitary. Taken together, these data indicate that CgA may function as a precursor to three highly conserved peptides that may exert regulatory functions in the neuroendocrine system.

Regulation of the GABA(A) receptor by nitric oxide in frog pituitary melanotrophs.

Nitric oxide (NO) is implicated in the regulation of various endocrine functions, but the effect of NO on GABA(A) receptor transmission has never been reported in endocrine cells. In the present study, we have investigated the effects of various agents acting on the NO transduction pathway on GABA(A) receptor function in frog pituitary melanotrophs. Histochemical studies using the NADPH-diaphorase reaction and immunohistochemical labeling with antibodies against neuronal NO synthase (nNOS) revealed that nNOS is expressed in the intermediate lobe of the pituitary and in cultured melanotrophs. Whole-cell patch-clamp recordings showed that the specific substrate of NOS L-arginine (L-Arg, 10(-4) M) or the NO donor sodium nitroprusside (10(-5) M) provoked a long-lasting inhibition of the current evoked by GABA (5 x 10(-6) M). The NOS inhibitor L-nitroarginine (10(-5) M) produced a biphasic effect, i.e. a transient decrease followed by a delayed increase of the GABA-evoked current amplitude. Similarly, the specific nNOS inhibitor 7-nitroindazole and the specific inducible NOS (iNOS) inhibitor aminoguanidine (10(-5) M each) provoked a transient depression of the current followed by a sustained potentiation. Formation of cGMP in neurointermediate lobes was enhanced by L-Arg (10(-4) M) and by the calcium-releasing agent caffeine (10(-4) M), and inhibited by the calmodulin (CaM)/Ca2+ complex blocker W7 (10(-5) M). The GABA-evoked current was potentiated by the guanylyl cyclase inhibitor ODQ (10(-8)-10(-7) M) and inhibited by the protein kinase G (PKG) activator 8pCPT-cGMP (3 x 10(-7)-3 x 10(-5) M). The present data indicate that NO, produced by a CaM/Ca2+-dependent NOS in frog melanotrophs, exerts an autocrine inhibitory effect on the GABA-evoked current. The action of NO on the GABA(A) receptor function is mediated through activation of the cGMP/PKG pathway.

Immunohistochemical localization of gamma-endorphin in the rat pituitary gland and hypothalamus.

To gain more information about cells and organelles involved in the secretion of gamma-endorphin in the rat pituitary gland, an immunohistochemical localization of this peptide was conducted at both light and electron microscopic levels. Using an antiserum to gamma-endorphin which shows little cross-reactivity with beta-lipotropin (beta-LPH) and beta-endorphin, it has been demonstrated that gamm-endorphin is present in all cells of the intermediate lobe and the corticotropic cells of the anterior lobe. At the electron microscopic level, staining for gamma-endorphin was restricted to the secretory granules which also contain ACTH and beta-LPH, suggesting that all of these peptides are released concomitantly during exocytosis. In the hypothalamus, immunostaining possibly related to both gamma- and beta-endorphin but not to beta-LPH was also observed in nerve fibers and not in cell bodies.

A cloned frog vasoactive intestinal polypeptide/pituitary adenylate cyclase-activating polypeptide receptor exhibits pharmacological and tissue distribution characteristics of both VPAC1 and VPAC2 receptors in mammals.

Three receptor subtypes for the neuropeptides vasoactive intestinal peptide (VIP) and pituitary adenylate cyclase-activating polypeptide (PACAP) have been identified in mammals: the PAC1 receptor (PAC1-R) which is selectively activated by PACAP, and two VPAC receptors (VPAC1-R and VPAC2-R), which are equally stimulated by PACAP and VIP. The structures of PACAP and VIP have been well conserved during evolution, but little is known about VIP/PACAP receptors in nonmammalian species. An amphibian VIP/PACAP receptor complementary DNA (cDNA) has been cloned and characterized from a frog (Rana ridibunda) pituitary cDNA library. The predicted protein contains seven putative transmembrane domains and exhibits the highest sequence identity (65%) with the human VPAC1-R. The cloned cDNA was transiently expressed in LLC-PK1 cells, and its pharmacological profile was determined in comparison with the human VPAC1-R. Both PACAP and VIP stimulated cAMP accumulation through the cloned receptor with an EC50 of about 30 nM. In contrast, secretin, at concentrations that stimulate the human VPAC1-R, had no effect on cAMP production. RT-PCR analysis revealed the widespread distribution of this frog VIP/PACAP receptor in peripheral tissues. In situ hybridization histochemistry using a complementary RNA probe showed that the receptor gene is highly expressed in several hypothalamic and thalamic nuclei and to a lesser extent in the pallium and striatum. In the pituitary, the highest messenger RNA levels were detected in the distal lobe. Taken together, these data show that the cloned frog receptor shares several common features with both the VPAC1-R and VPAC2-R of mammals; the frog receptor exhibits the highest sequence identity with mammalian VPAC1-R, but the lack of effect of secretin and the brain distribution of the receptor are reminiscent of the characteristics of the mammalian VPAC2-R. The sequence of the frog receptor should thus prove useful to decipher the structure-activity relationships of the VIP/PACAP receptor family.

Visualization of gamma-aminobutyric acid A receptors on proopiomelanocortin-producing neurons in the rat hypothalamus.

It has recently been shown that gamma-aminobutyric acid (GABA) and central-type benzodiazepine receptor agonists inhibit the expression of the POMC gene and the release of POMC-derived peptides from hypothalamic neurons. To determine whether the inhibitory effect of GABA could be accounted for by a direct action on POMC neurons, we investigated the localization of the beta 1-subunit of the GABAA-benzodiazepine-receptor complex in the arcuate nucleus. Using a monoclonal antibody raised against a synthetic fragment of the beta 1-subunit, we demonstrate the presence of GABAA receptor on POMC neurons. The proportion of POMC neurons that exhibit immunoreactivity for the beta 1-subunit of the GABAA receptor was not significantly different in the posterior portion (73.0-76.0%) and anterior portion (61.3-62.7%) of the arcuate nucleus. The data also revealed that in the arcuate nucleus, a majority of neurons that were immunostained by the antibody to the beta 1-subunit were not POMC positive. The present results support the concept that GABAA and central-type benzodiazepine receptor agonists exert a direct inhibitory action on POMC neurons. The data also indicate the existence of subsets of POMC neurons within the arcuate nucleus.

Catecholaminergic control of alpha-melanocyte-stimulating hormone (alpha MSH) release by frog neurointermediate lobe in vitro: evidence for direct stimulation of alpha MSH release by thyrotropin-releasing hormone.

The role of dopaminergic and adrenergic innervation of the intermediate lobe of amphibian pituitary in the release of alpha MSH has been studied in vitro. Neurointermediate lobes of frog (Rana ridibunda Pallas) have been perifused in amphibian culture medium (ACM) for 5-7 h. alpha MSH released in the effluent perifusate was measured by means of a sensitive and specific RIA. No significant morphological alteration of neurointermediate lobe cells was observed during the perifusion experiment, even at the electron microscopic level. The existence of dopaminergic receptors, responsible for an inhibition of frog melanotrophs, was shown using the dopaminergic agonists apomorphine (10(-6) M) and bromo-2-ergocryptine (10(-8) and 10(-7) M), which initiated a marked reduction of alpha MSH secretion. The effect of apomorphine was obliterated by the dopaminergic antagonist haloperidol. Haloperidol itself induced a dose-related stimulation, and the monoamine oxidase inhibitor nialamide (4 x 10(-3) M) inhibited alpha MSH secretion. In addition, haloperidol led to a complete reversal of the inhibitory effect of nialamide on alpha MSH secretion. These results demonstrate the existence, in the parenchyme of the intermediate lobe, of dopaminergic nerve fibers that are functionally active. The beta-adrenergic agonist isoproterenol was responsible for a dose-related stimulation of alpha MSH secretion; the stimulatory effect was reversed by the beta-adrenergic antagonist propranolol. TRH is a potent stimulator of alpha MSH secretion in amphibians. Since haloperidol and propranolol did not abolish the stimulation of alpha MSH release induced by TRH, it appeared that TRH action was not mediated via an inhibition of dopamine release or via a stimulation of adrenergic nerve fibers.

Expression of GalR1 and GalR2 galanin receptor messenger ribonucleic acid in proopiomelanocortin neurons of the rat arcuate nucleus: effect of testosterone.

Previous studies have shown that galanin-containing fibers make synaptic contacts with POMC neurons in the arcuate nucleus. However, the ability of POMC neurons to express galanin receptors has never been assessed. The present study was designed to investigate whether POMC neurons express galanin receptor messenger RNA (mRNA) and whether testosterone could modulate galanin receptor gene expression. A dual-labeling in situ hybridization histochemistry, using 35S-labeled (galanin receptors GalR1 or GalR2) and digoxigenin-labeled (POMC) riboprobes, was performed on brain sections from intact, castrated, and testosterone-replaced adult male rats. For analysis, the arcuate nucleus was divided into four rostro-caudal areas. The results revealed that both GalR1 and GalR2 mRNAs were expressed in POMC neurons. Most POMC neurons expressing galanin receptor mRNAs were found in the rostral parts of the nucleus. Castration reduced the labeling density of galanin receptor mRNAs in POMC neurons, and testosterone prevented the effects of castration in all rostro-caudal subdivisions of the arcuate nucleus. Taken together, these data indicate that galanin can directly modulate the activity of POMC neurons, via an action on GalR1 or GalR2 receptors, particularly in the rostral-arcuate nucleus. In addition, testosterone can modulate the expression of GalR1 and GalR2. Because POMC neurons located in the rostral part of the nucleus are known to project preferentially to the preoptic area, POMC neurons expressing the galanin receptor genes may play an important role in the regulation of the GnRH neuroendocrine axis.

Evidence that TGF beta may directly modulate POMC mRNA expression in the female rat arcuate nucleus.

The purpose of the present study was to determine whether TGF beta, a cytokine secreted by hypothalamic astrocytes, was able to regulate POMC neurons in the arcuate nucleus. In a first set of experiments, mediobasal hypothalamic fragments were exposed to TGF beta(1), and the relative POMC mRNA expression was assessed by in situ hybridization using a radiolabeled POMC riboprobe. The results showed that 4 x 10(-10) M TGF beta(1) was efficient in decreasing significantly the amounts of POMC mRNA (P < 0.01). Interestingly, the decrease of relative POMC mRNA levels was higher in the rostral than in the caudal parts of the arcuate nucleus. In a second set of experiments, we examined the occurrence of TGF beta receptors expression in arcuate POMC neurons. Dual labeling in situ hybridization and in situ hybridization, coupled to immunohistochemical labeling, were performed to examine mRNA expression of the type I serine-threonine kinase receptor for TGF beta and the presence of type II receptor for TGF beta, respectively, in POMC neurons. The results indicated that TGF beta receptor I mRNA and TGF beta receptor II protein were expressed in numerous POMC neurons. Regional analysis revealed that the highest proportion of POMC neurons expressing TGF beta receptors was located in the rostral part of the arcuate nucleus. Using dual labeling immunohistochemistry, we also found that Smad2/3 immunoreactivity, a TGF beta(1) downstream signaling molecule, was present in the cytoplasm and nucleus of some POMC (beta-endorphin) neurons. We next examined whether the number of POMC neurons expressing TGF beta-RI mRNA was affected by sex steroids. Quantification of the number of POMC neurons expressing TGF beta receptor I mRNA in ovariectomized, ovariectomized E2-treated, and ovariectomized E2 plus progesterone-treated animals revealed that estrogen treatment decreased the expression of TGF beta receptor I mRNA in POMC neurons located in the rostral half of the arcuate nucleus, an effect reversed by progesterone in a subset of the most rostral cells. Taken together, these data reveal that TGF beta(1) may directly modulate the activity of POMC neurons through the activation of TGF beta receptors. Therefore, the present study provides additional evidence for the involvement of TGF beta(1) in the regulation of neuroendocrine functions and supports the existence of a glial-to-neurons communication within the arcuate nucleus.

Characterization of the somatostatin receptor subtype in a bronchial carcinoid tumor responsible for Cushing's syndrome.

Small ACTH-secreting carcinoid tumors responsible for Cushing's syndrome are often difficult to localize using available radiological investigations. Somatostatin receptors have been found in about 90% of carcinoid tumors studied, leading to a new approach for the localization of tumors or metastasis by using radiolabeled somatostatin analogs. We report a case of Cushing's syndrome due to an ACTH-secreting bronchial carcinoid tumor, completely suppressible with octreotide treatment and evidenced by body scintigraphy with 111In-labeled pentreotide. After removal, which led to patient recovery, the tumor was studied in vitro. In situ hybridization, using a complementary DNA probe, revealed POMC messenger ribonucleic acid in a subpopulation of tumor cells. These cells were labeled by immunochemistry using an antiserum directed against ACTH. Confocal laser scanning microscopy analysis showed that the ACTH-immunoreactive peptide was sequestered in secretory granules. Autoradiographic labeling using [125I-Tyrzero,D-Trp8]somatostatin-14 demonstrated the presence of somatostatin-binding sites in the whole tumor tissue. The relative affinities of various selective somatostatin analogs and the ability of GTP to inhibit radioligand binding suggested that the receptor expressed in the tumor cells belonged to the SSTR-2 subtype.

Cloning, sequence analysis and tissue distribution of the mouse and rat urotensin II precursors.

Urotensin II (UII) is a cyclic neuropeptide initially isolated from the caudal neurosecretory system of teleost fish. The recent cloning of the UII precursor in frog and human has demonstrated that the peptide is not restricted to the fish urophysis but that it is also expressed in the central nervous system of tetrapods. Here, we describe the characterization of the cDNAs encoding prepro-UII in mouse and rat. A comparison of the primary structures of mouse and rat UII with those of other vertebrate UII reveals that the sequence of the cyclic region of the molecule (CFWKYC) has been fully conserved. In contrast, the N-terminal flanking domain of prepro-UII has markedly diverged with only 48% sequence identity between the mouse or rat and the human precursors. In situ hybridization histochemistry showed that the prepro-UII gene is predominantly expressed in motoneurons of the brainstem and spinal cord, suggesting that UII may play a role in the control of neuromuscular functions.