RET expression and neuron-like differentiation of pheochromocytoma and normal chromaffin cells.

Receptor tyrosine kinase RET is normally expressed at low levels in chromaffin cells and high levels in sympathetic neurons. Paradoxically, it is overexpressed in subsets of pheochromocytomas. The overexpressed protein is usually wild-type, except in multiple endocrine neoplasia type 2. Possible explanations for overexpression include tumor origin from RET-expressing sympathoadrenal progenitors that escape developmental culling during embryogenesis, or reactivation of signaling pathways related to neuronal differentiation. Normal adult chromaffin and pheochromocytoma cells can undergo neuron-like differentiation in cell culture. In this investigation, cultured cells from two normal human adrenal medullas, two of three human pheochromocytomas, and one extra-adrenal paraganglioma showed RET induction corresponding with extensive nerve growth factor-induced outgrowth of neurite-like processes, while one pheochromocytoma showed neither processes nor RET induction. RET was similarly upregulated in parallel with process outgrowth in cultures of normal rat chromaffin cells and PC12 rat pheochromocytoma cells. In contrast, mouse pheochromocytoma cells that constitutively express high levels of wild-type RET together with other neuronal progenitor markers showed no further RET increase after cyclic AMP-induced process outgrowth. The RET-activating ligand GDNF was anti-apoptotic for mouse pheochromocytoma but not for PC12 cells. The findings suggest that overexpression of RET in pheochromocytomas could result either from a secondary event that activates signaling pathways mediating adult chromaffin cell plasticity or as a component of a persistent sympathoadrenal progenitor phenotype. Whether wild-type RET contributes to tumor development or is merely a lineage marker for cells at various stages of neuronal differentiation may vary, with other tumor characteristics.

Pheochromocytomas in Nf1 knockout mice express a neural progenitor gene expression profile.

Pheochromocytomas are adrenal medullary tumors that typically occur in adult patients, with increased frequency in multiple endocrine neoplasia type 2, von Hippel-Lindau disease, familial paraganglioma syndromes and neurofibromatosis type 1 (NF1). Pheochromocytomas arise in adult mice with a heterozygous knockout mutation of exon 31 of the murine Nf1 gene, providing a mouse model for pheochromocytoma development in NF1. We performed a microarray-based gene expression profiling study comparing mouse pheochromocytoma tissue to normal adult mouse adrenal medulla to develop a basis for studying the pathobiology of these tumors. The findings demonstrate that pheochromocytomas from adult neurofibromatosis knockout mice express multiple developmentally regulated genes involved in early development of both the CNS and peripheral nervous system. One of the most highly overexpressed genes is receptor tyrosine kinase Ret, which is known to be transiently expressed in the developing adrenal gland, down-regulated in adult adrenals and often overexpressed in human pheochromocytomas. Real-time polymerase chain reaction validated the microarray results and immunoblots confirmed the overexpression of Ret protein. Other highly expressed validated genes include Sox9, which is a neural crest determinant, and Hey 1, which helps to maintain the progenitor status of neural precursors. The findings are consistent with the recently proposed concept that persistent neural progenitors might give rise to pheochromocytomas in adult mouse adrenals and suggest that events predisposing to tumor development might occur before formation of the adrenal medulla or migration of cells from the neural crest. However, the competing possibility that developmentally regulated neural genes arise secondarily to neoplastic transformation cannot be ruled out. In either case, the unique profile of gene expression opens the mouse pheochromocytoma model to new applications pertinent to neural stem cells and suggests potential new targets for treatment of pheochromocytomas or eradication of their precursors.

Up-regulation of ret by reserpine in the adult rat adrenal medulla.

The receptor tyrosine kinase, ret, is activated by glial cell line-derived neurotrophic factor, neurturin and related ligands that bind to glycosylphosphatidylinositol-tailed receptors GFRalpha1-4. Ret expression is developmentally regulated and detectable only at very low levels in adult adrenal medulla. However, mutations of ret that cause constitutive activation or alter signal transduction give rise to adrenal medullary hyperplasia and pheochromocytomas in humans with hereditary multiple endocrine neoplasia (MEN) syndromes 2A and 2B and in animal models. These discordant observations pose the conundrum of how a molecule barely detectable in the adult adrenal can contribute to development of adrenal medullary pathology that typically occurs in adults. We recently reported that depolarization and phorbol esters that activate protein kinase C act synergistically with neurturin to up-regulate ret protein and mRNA expression in adult rat chromaffin cell cultures. Those findings suggested that ret expression in vivo is not static and might be regulated in part by neurally derived signals. We show here that the anti-hypertensive agent reserpine, which is known to cause a reflex increase in trans-synaptic stimulation of chromaffin cells, increases expression of ret mRNA and protein in adult rat adrenal medullary tissue in vivo. Elevated ret protein levels are detectable both by immunoblots and immunohistochemistry, which shows immunoreactive ret in chromaffin cells and neurons after reserpine administration. The finding that ret expression is subject to up-regulation by environmental signals in vivo suggests that epigenetic factors might influence the development of adrenal medullary disease by affecting the expression of ret. It is known that long-term administration of reserpine leads to the development of adrenal medullary hyperplasia and pheochromocytomas in rats. Our findings suggest potential utility of the rat model for studying the roles of ret in the adrenal medulla and the mechanisms of its involvement in MEN 2 and other pheochromocytoma syndromes.

Animal models of pheochromocytoma.

Pheochromocytomas are neuroendocrine tumors of adrenal chromaffin cells. They are rare in all species except rats but occur with increased frequency in several human familial tumor syndromes. Concurrence of pheochromocytoma with other tumors sometimes parallels these human syndromes in rats, bovines, horses and dogs but a shared genetic basis for human and spontaneously occurring animal pheochromocytomas has thus far not been established. Pheochromocytomas are inducible in rats by a variety of non-genotoxic substances that may act indirectly by stimulating chromaffin cell proliferation. They are not known to be similarly inducible in other species but arise with increased frequency in transgenic and knockout mice that to varying degrees recapitulate human tumor syndromes. Preliminary evidence suggests that homologous somatic genetic changes might contribute to pheochromocytoma development in humans and some mouse models. The nerve growth factor-responsive PC12 cell line, established from a rat pheochromocytoma, has for almost 30 years served as a research tool for many aspects of neurobiology involving normal and neoplastic conditions. Recently developed pheochromocytoma cell lines from neurofibromatosis knockout mice supplement the PC12 line and have generated additional applications. Advantages of the mouse lines include expression of substantial levels of the epinephrine-synthesizing enzyme, phenylethanolamine N-methyltransferase and expression of high levels of the receptor tyrosine kinase, Ret, which is characteristic of sporadic and familial human pheochromocytomas but not of PC12 cells. Disadvantages include an apparently less stable phenotype. It is difficult to establish pheochromocytoma cell lines from any species, although the tumor cells persist in culture for many months. Understanding of factors that permit pheochromocytoma cells to proliferate might itself provide important insights for tumor biology.

Potentiation of mitogenesis in adult rat chromaffin cell cultures by immunosuppressive agent FK506.

The immunosuppressive drugs FK506 and cyclosporin inhibit T- and B-lymphocyte proliferation and exert neuritogenic and/or cytoprotective effects on several types of neurons. While the immunosuppressive actions of both drugs are mediated in large part by inhibition of the Ca(2+)-dependent phosphatase, calcineurin, FK506 is known to exert additional effects. In the present study, FK506 is shown to potentiate mitogenic effects of the neurotrophic factor, neurturin, on normal adult rat adrenal chromaffin cells in culture. The effect is not seen with cyclosporin or with a non-immunosuppressive analog of FK506, GPI-1046. The finding of increased mitogenesis in response to FK506 may have applications to the study of normal and neoplastic neuroendocrine cells and to understanding the development of some types of tumors in transplant patients.

Chromaffin cell mitogenesis by neurturin and glial cell line-derived neurotrophic factor.

Neurturin and glial cell line-derived neurotrophic factor are novel mitogens for normal adult rat chromaffin cells in vitro. These neurotrophic factors differ from the previously described adult chromaffin cell mitogens, nerve growth factor and basic fibroblast growth factor, in that their effects are potentiated by depolarization and activation of protein kinase C. Neurturin and glial cell line-derived neurotrophic factor signal via the receptor tyrosine kinase, ret, but may also act independently of ret. Both depolarization and phorbol esters act synergistically with neurturin to up-regulate ret protein expression in chromaffin cell cultures, suggesting a mechanism for potentiation of mitogenesis. However, a direct role for ret in mitogenesis has not been established. Stimulation by neurturin causes increased phosphorylation of extracellular signal-regulated kinases 1 and 2 in cultured chromaffin cells, and mitogenesis is prevented by inhibitors of their phosphorylation. Inhibitors of phosphatidylinositol 3-kinase also prevent mitogenesis. The present findings suggest the hypothesis that neurotrophic factors and neurally derived signals might cooperatively regulate chromaffin cell proliferation in vivo in the rat. In addition, trans-synaptic stimulation might provide a route by which epigenetic factors could influence the development of adrenal medullary hyperplasia in humans with hereditary multiple endocrine neoplasia syndromes 2A and 2B by affecting expression and/or activation of ret.

Mitogenic signaling by cyclic adenosine monophosphate in chromaffin cells involves phosphatidylinositol 3-kinase activation.

Increase of intracellular cyclic adenosine monophosphate by the permeant cyclic adenosine monophosphate analog, 8-(4-chlorophenylthio)-adenosine 3':5'- cyclic monophosphate, is mitogenic for normal adult rat chromaffin cells. The mitogenic effect is blocked by the phosphatidylinositol 3-kinase inhibitor, LY294002, and is associated with accumulation of phosphorylated Akt and p70S6 kinase, suggesting that cyclic adenosine monophosphate activates Type l phosphatidylinositol 3-kinase. The mechanism of activation was examined in PC12 pheochromocytoma cells, which are neoplastic chromaffin cells that exhibit many of the biochemical characteristics of their normal counterparts. Incubation of PC12 cells with 8-(4-chlorophenylthio)-adenosine 3':5'- cyclic monophosphate led to a significant increase in total phosphatidylinositol 3-kinase activity that was sensitive to low concentrations of LY294002. The increase was maximal at 1 h and returned to basal levels within six hours. Immunoprecipitation studies showed no increase in phosphatidylinositol 3-kinase activity in anti-phosphotyrosine immune complexes from PC12 cells stimulated by 8-(4-chlorophenylthio)-adenosine 3':5'- cyclic monophosphate, in contrast to cells stimulated by nerve growth factor. Instead, activity was demonstrated in association with p110gamma and p85. These findings suggest that cyclic adenosine monophosphate causes activation of Types IA and IB phosphatidylinositol 3-kinase by a novel mechanism in chromaffin and pheochromocytoma cells. That activation may contribute to chromaffin cell proliferation and to the development and progression of pheochromocytomas. J. Cell. Biochem. Suppl. 36: 89-98, 2001.

C-cell hyperplasia, pheochromocytoma and sympathoadrenal malformation in a mouse model of multiple endocrine neoplasia type 2B.

Dominantly inherited multiple endocrine neoplasia type 2B (MEN2B) is characterized by tumors of the thyroid C-cells and adrenal chromaffin cells, together with ganglioneuromas of the gastrointestinal tract and other developmental abnormalities. Most cases are caused by substitution of threonine for Met918 in the RET receptor tyrosine kinase, which is believed to convert the RET gene to an oncogene by altering the enzyme's substrate specificity. We report the production of a mouse model of MEN2B by introduction of the corresponding mutation into the ret gene. Mutant mice displayed C-cell hyperplasia and chromaffin cell hyperplasia progressing to pheochromocytoma. Homozygotes did not develop gastrointestinal ganglioneuromas, but displayed ganglioneuromas of the adrenal medulla, enlargement of the associated sympathetic ganglia and a male reproductive defect. Surprisingly, homozygotes did not display any developmental defects attributable to a loss-of-function mutation. Thus, while our results support the conclusion that the Met918Thr substitution is responsible for MEN2B, they suggest that the substrate specificity of the RET kinase does not interfere with its normal role in the development of the kidneys and enteric nervous system.

Pheochromocytoma cell lines from heterozygous neurofibromatosis knockout mice.

Transplantable tumors and cell lines have been developed from pheochromocytomas arising in mice with a heterozygous knockout mutation of the neurofibromatosis gene, Nf1. Nf1 encodes a ras-GTPase-activating protein, neurofibromin, and mouse pheochromocytoma (MPC) cells in primary cultures typically show extensive spontaneous neuronal differentiation that may result from the loss of the remaining wild-type allele and defective regulation of ras signaling. However, all MPC cell lines express neurofibromin, suggesting that preservation of the wild-type allele may be required to permit the propagation of MPC cells in vitro. MPC lines differ from PC12 cells in that they express both endogenous phenylethanolamine N-methyltransferase (PNMT) and full-length PNMT reporter constructs. PNMT expression is increased by dexamethasone and by cell-cell contact in suspension cultures. Mouse pheochromocytomas are a new tool for studying genes and signaling pathways that regulate cell growth and differentiation in adrenal medullary neoplasms and are a unique model for studying the regulation of PNMT expression.

Vitamin D3-induced proliferative lesions in the rat adrenal medulla.

Adrenal medullary hyperplasia and pheochromocytomas are induced in rats by a variety of non-genotoxic agents, and we have hypothesized that these agents induce lesions indirectly by stimulating chromaffin cell proliferation. Vitamin D3, which has not been previously associated with adrenal medullary proliferative lesions, is the most potent in vivo stimulus to chromaffin cell proliferation yet identified. The present investigation utilized the vitamin D3 model to prospectively test the relationship between mitogenicity and focal proliferative lesions in the adrenal medulla and to determine early events in the pathogenesis of these lesions. Charles River Crl:CD BR rats were treated with 0; 5000; 10,000; or 20,000 IU/kg/day of vitamin D3 in corn oil (5 ml/kg) by oral intubation. Rats were killed after 4, 8, 12, or 26 weeks of treatment, following a final week of labeling with bromodeoxyuridine (BrdU) using a mini-pump. Adrenal sections were double-stained for BrdU and phenylethanolamine-N-methyl transferase (PNMT) to discriminate epinephrine (E) from norepinephrine (NE) cells or for vesicular acetylcholine transporter (VAchT) to identify cholinergic nerve endings. Vitamin D3 caused a 4-5-fold increase in BrdU labeling at week 4, diminishing to a 2-fold increase by week 26. An initial preponderance of labeled E cells gave way to a preponderance of labeled NE cells. By week 26, 17/19 (89%) animals receiving the 2 highest doses of vitamin D3 had focal adrenal medullary proliferative lesions, in contrast to an absence of lesions in control rats. The lesions encompassed a spectrum including BrdU-labeled "hot spots" not readily visible on H and E sections, hyperplastic nodules, and pheochromocytomas. Lesions were usually multicentric, bilateral, and peripheral in location, and almost all were PNMT-negative. The lesions were not cholinergically innervated, suggesting autonomous proliferation. Hot spots, hyperplastic nodules, and pheochromocytomas appear to represent a continuum rather than separate entities. Their development might involve selective responses of chromaffin cell subsets to mitogenic signals, influenced by both innervation and corticomedullary interactions. A number of non-genotoxic compounds that induce pheochromocytomas in rats are known to affect calcium homeostasis. The results of this study provide further evidence to support the hypothesis that altered calcium homeostasis is indirectly involved in the pathogenesis of pheochromocytomas, via effects on chromaffin cell proliferation.

Discordant effects of rapamycin on proliferation and p70S6 kinase phosphorylation in normal and neoplastic rat chromaffin cells.

Normal adult rat chromaffin cells show a robust proliferative response in vitro to nerve growth factor (NGF) and other mitogens. In contrast, PC12 rat pheochromocytoma cells proliferate in the absence of exogenous mitogens and undergo neuronal differentiation in response to NGF. We demonstrate in this work that the antiproliferative drug rapamycin suppresses normal chromaffin cell proliferation. This effect is blocked by FK 506, indicating that it occurs via interaction of rapamycin with its intracellular binding protein, FKBP. Rapamycin must be added within 2 days of mitogen stimulation in order to be fully effective. PC12 cells are refractory to the antiproliferative effect of rapamycin, although rapamycin does exert its expected inhibitory effect in PC12 cells on both basal and NGF-stimulated activation of one of its biochemical targets, the 70-kDa S6 protein kinase (p70S6K). The discordant findings suggest that a proliferative signal normally requiring activation of p70S6K either is unnecessary in PC12 cells or is provided by a downstream or cross-communicating pathway. They also suggest that p70S6K does not participate in the morphological responses of PC12 cells to NGF. Determining the basis for rapamycin resistance in PC12 cells might help to identify signaling abnormalities involved in the pathogenesis of pheochromocytoma.

Nerve growth factor receptor signaling in proliferation of normal adult rat chromaffin cells.

Adult rat chromaffin cells may proliferate or extend neurites when stimulated by nerve growth factor (NGF) but their response is predominantly proliferative, making them a unique model for studying how mitogenic specificity is achieved. We examined contributions of the NGF receptors trk and p75 and of the major NGF signaling pathways to proliferation versus neurite outgrowth. The type of initial NGF response does not correlate with intensity of immunoreactivity for trk or p75. However, proliferation is initiated at lower NGF concentrations than neurite outgrowth, suggesting that it requires a less intense signal. Mitogenic cooperativity between receptors at low NGF concentrations is suggested by inhibitory effects of p75-blocking antibodies, but responses to trk-agonist antibody indicate that trk activation alone can induce proliferation. NGF-induced phosphorylation of ras-mediated mitogen-activated protein kinases (MAPK) Erk1 and Erk2 is as prolonged in normal chromaffin cells as in PC12 cells, where NGF is neuritogenic. Trk-agonist antibody, which is as mitogenic as NGF but less neuritogenic, causes equally prolonged but less intense ERK phosphorylation. The MAPK kinase(MEK-1) inhibitor PD98059 partially inhibits Erk phosphorylation and does not inhibit chromaffin cell proliferation, while depolarization selectively inhibits proliferation without blocking Erk phosphorylation. Proliferation is markedly reduced by the phosphoinositol-3 (PI-3) kinase inhibitor LY294002 while downregulation of protein kinase C (PKC) causes no change. These findings suggest that low-level, rather than short-duration, stimulation of NGF signaling pathways causes NGF to be mitogenic. Ras-mediated MAPK activation may be more critical in neurite outgrowth than in proliferation and PI-3 kinase may be the major mitogenic determinant.

Subacute cocaine treatment changes expression of mouse liver cytochrome P450 isoforms.

Acute administration of single high doses of cocaine (50 or 60 mg/kg) produces liver injury in mice that have been pretreated with inducers of mixed function oxidases. Multiple low doses of cocaine (10-30 mg/kg) will produce hepatotoxicity without prior induction. To establish whether cocaine can induce its own activation, mice were given three daily injections of cocaine. Total cytochrome P450 content of the liver did not change. After 3 days the amount of cytochrome P450 2B10, as measured by pentoxy resorufin-O-dealkylase activity and immunoblotting, increased 3-fold. Cytochrome P450 2A5-catalyzed coumarin 7-hydroxylase activity and immunoreactive protein increased by about 50%. Enzyme activities and Western blotting of isoforms 1A, 2E, and 3A showed no change during this time. Chronic cocaine increased N-hydroxylation of norcocaine. Immunoinhibition studies showed that cytochrome P450 2A5 was the major isoform responsible for norcocaine N-hydroxylation. These results demonstrate that chronic cocaine can induce its own metabolism. Similar increases were also observed in mice not susceptible to liver injury from chronic cocaine.

Immunoreactivity of normal rabbit serum with epinephrine (E) cells of the rat adrenal medulla after microwave antigen retrieval.

Normal rabbit serum (NRS) produces intense staining of epinephrine (E) cells in microwave-heated sections of rat and mouse adrenal gland. This staining is not eliminated by liver adsorption, complement inactivation, high salt buffer, Triton X-100 or dilution in normal goat serum and bovine serum albumin (BSA), suggesting that it may result from specific antigen-antibody interactions. Western blots of adrenal medullary protein probed with NRS reveal several bands. The major band does not correspond to rat chromogranin A, which is a major constituent of E-cell secretory granules. The findings suggest that NRS may contain autoantibodies against a secreted rabbit E-cell protein with a homologous counterpart in rats and mice, and that this protein may be immunologically unmasked in situ by microwave heating. This phenomenon is a potential source of error in immunohistochemical studies of the adrenal medulla, and has potential biological significance in neuroimmunology.

Comparative studies of chromaffin cell proliferation in the adrenal medulla of rats and mice.

Spontaneous and drug-induced pheochromocytomas are common in rats and rare in mice. The antihypertensive drug reserpine has been shown to both induce pheochromocytomas and stimulate chromaffin cell proliferation in rats, leading to the hypothesis that reserpine causes pheochromocytomas indirectly by providing a proliferative setting in which DNA damage may occur. The present investigation was undertaken to obtain baseline information on the relationship across species between chromaffin cell proliferation and pheochromocytomas. Basal chromaffin cell proliferation was compared in age-matched young adult mice and rats. In addition, mice were studied for adrenal medullary responses to reserpine, and mouse chromaffin cells in vitro were studied for responses to agents that are mitogenic for cultured rat chromaffin cells. Concurrently maintained F-344 rats and several strains of mice showed no significant difference in basal BrdU incorporation over a 1-week period. Mice also showed an adrenal medullary proliferative response to reserpine that was comparable to the response previously reported for rats. However, there was a marked disparity between rat and mouse chromaffin cells in vitro, and cultured mouse chromaffin cells did not respond to any mitogens. The in vivo data indicate that interspecies differences in basal- or reserpine-stimulated chromaffin cell proliferation sufficient to account for different frequencies of pheochromocytomas are not detectable at a single time point in young adult animals. However, the possibility that such differences might emerge with aging has not been ruled out. These data further suggest either that stimulation of chromaffin cell proliferation might be necessary but not sufficient for development of pheochromocytomas or that stimulated proliferation in mice might not be sustained. The inability of cultured mouse chromaffin cells to respond to mitogens raises the speculation of whether mechanisms that prevent proliferation of normal chromaffin cells in vitro might also help to protect mice from developing pheochromocytomas.

Vitamin D3, lactose, and xylitol stimulate chromaffin cell proliferation in the rat adrenal medulla.

Chronic consumption by rats of diets rich in sugars or sugar alcohols leads to an increased incidence of pheochromocytomas. This relationship is hypothesized to be based on altered Ca2+ homeostasis due to increased intestinal Ca2+ absorption. Other agents associated with pheochromocytomas in rats in long-term toxicity studies have been shown to increase chromaffin cell proliferation, leading to the suggestion that the tumors occur secondarily to increased chromaffin cell turnover. We have demonstrated marked stimulation of chromaffin cell proliferation by vitamin D3, a potent stimulus to Ca2+ absorption not previously associated with adrenal medullary toxicity. This effect is detectable during the first week of dietary supplementation and persists throughout a 4-week time course. Lactose and xylitol, representative of sugars and sugar alcohols associated with pheochromocytomas, are also mitogenic but to a lesser extent, with their effects first detectable during Week 4 of dietary supplementation. Vitamin D3, its active metabolite calcitriol, lactose, and xylitol all fail to stimulate proliferation of rat chromaffin cells in vitro. The mitogenic effects of these agents may be mediated presynaptically in vivo. The data suggest that altered Ca2+ homeostasis may increase chromaffin cell proliferation and support the hypothesis that diets containing high concentrations of sugars and sugar alcohols cause pheochromocytomas in rats secondarily by this mechanism.

Procedure-related fetal losses in transplacental versus nontransplacental genetic amniocentesis.

OBJECTIVE: We hypothesize that loss rates after amniocentesis do not differ in transplacental and nontransplacental taps performed by experienced operators. STUDY DESIGN: Subjects were 1000 women undergoing second-trimester amniocentesis: 745 were referred for maternal age; 132 for positive maternal serum alpha-fetoprotein screens, 41 indicating a risk for fetal neural tube defect, 91 indicating a risk for fetal chromosome abnormality; and 123 were referred for other reasons. All procedures were videotaped. The placenta was anterior in 518 cases; in 306 of these the needle went through the placenta. All pregnancies were prospectively evaluated through delivery. RESULTS: There were 13 losses among the 1000 procedures (1.3%). The transplacental losses occurred from 4 to 71 days after procedure, median 26.5 days; the nontransplacental losses from 12 days after procedure to term, median 25 days. The loss rate was essentially similar in the two categories: six transplacental (1.96%) and seven nontransplacental (1%) (relative risk 1.52 [95% confidence limits 0.84 to 2.75], p = 0.23). If the three patients with elevated maternal serum alpha-fetoprotein values were excluded from data analysis, the loss rates in the two groups were virtually identical (relative risk 0.98 [95% confidence limits 0.38 to 2.54], p = 1.0000). CONCLUSION: Transplacental amniocentesis does not appear to increase the fetal loss rate in the hands of experienced surgeons. Moreover, in view of the time span between amniocentesis and loss in both groups, a procedural cause seems questionable.

Improved method for bile collection in unrestrained conscious rats.

We describe an improved method for continuous collection of bile from unrestrained rats. Use of an externally accessible, continuous-loop cannula when cannulating the common bile duct allows for full recovery from anesthetic effects and maintenance of a normal bile salt pool until the cannula loop is cut. Bile resulting from the cut cannula is diverted into a surgically implanted glass collection vessel and removed periodically via an externalized sampling port. Bile flow over a 24-hour collection period averaged 0.98 +/- 0.04 ml/hr (Mean +/- SEM, n = 9) with no gross pathological changes noted upon necropsy. This technique offers the capability of reestablishing conditions as close to physiologic as possible postsurgery to minimize potential artifacts during bile collection.

Hepatic morphologic and biochemical changes induced by subacute cocaine administration in mice.

The initial event and site of cocaine-induced hepatic injury have not been elucidated. In an attempt to identify the minimal effective dose and the site of injury, we have examined the livers of mice exposed to small daily doses of cocaine, using morphological and biochemical methods. All doses of cocaine greater than 5 mg/kg were able to cause significant elevation of serum glutamic pyruvic transaminase. Light microscopy revealed a progression of centrilobular necrosis as the dose increased from 10-30 mg/kg. The initial morphologic changes observed prior to necrosis included aggregation of intermediate filaments and dilation of rough endoplasmic reticulum with loss of ribosomes. Immunohistochemistry, using antibodies to cytokeratins, showed staining of individual hepatocytes in livers from cocaine-treated animals but not in controls. In contrast to earlier reports, we found little, if any, disruption of mitochondria. In vitro, the direct application of cocaine, norcocaine, and N-hydroxynorcocaine on isolated mitochondria had no effect on the ADP:O or respiratory control ratios, at concentrations up to 2.0 mM. Our studies demonstrate that any early cellular alterations in cocaine-induced hepatic injury are manifested in intermediate filaments and endoplasmic reticulum with no evidence of mitochondrial involvement.