INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

ALTHOUGH THE ADRENAL CORTEX and the adrenal medulla are often considered separate functional units, the coexistence of the two embryologically distinct tissues within one organ and the demonstration of their physiological interactions suggest that communication between medullary and adrenocortical cells contributes to their optimal function (6, 18, 24, 39). Examples of functional interactions include the maintenance of chromaffin cell phenotype by corticosteroids (13, 39) and the modulation of steroidogenesis by the neurotransmitters and neuropeptides expressed by chromaffin cells (10, 11, 16, 38). The mammalian adrenal is organized such that a central medulla is surrounded by a zonated adrenal cortex with centripetally directed blood flow (37), providing the mechanism by which adrenocortical cell products directly affect medullary cells. On the basis of this organization, however, it is unlikely that medullary products affect adrenocortical cells via vascular perfusion. Alternatively, neuronal and paracrine interactions are the likely mechanisms for medullary control of adrenocortical function in mammals (4, 15, 19). Medullary ganglion cells, including a prominent vasoactive intestinal peptide (VIP)-expressing population (18), extend neurites radially through the inner zones of the cortex to form a plexus in the outermost zone, the zona glomerulosa, thus providing a morphological substrate for medullary (neuronal) control of adrenocortical function (19). On the other hand, paracrine interactions likely occur between the chromaffin and adrenocortical cells located in the innermost zone of the cortex, the zona reticularis, and/or between extramedullary chromaffin cells and adrenocortical cells (4). It is noteworthy that in the adrenals of those vertebrates in which chromaffin cells and adrenocortical cells are intermixed (e.g., some fishes, amphibians, birds, and reptiles), the paracrine mechanism is likely to account for much of the cellular interaction (24).

To study the cellular interactions between chromaffin and adrenocortical cells, we developed primary cocultures of frog adrenal cells (31). The frog adrenal is used as a model because its innate characteristics are likely to facilitate interactions between chromaffin and adrenocortical cells; these characteristics include the intermixing of the two cell types and the extension of short processes by chromaffin cells. In addition, we observed intrinsic neurons within the frog adrenal that may mimic the intrinsic innervation of the mammalian adrenal (unpublished observations and Ref. 18). Chromaffin and adrenocortical cells maintain many characteristics in coculture, including the expression and release of neurotransmitters, neuropeptides, aldosterone, and corticosterone; thus these cocultures are a useful model to study their cellular interactions (31).

In response to homeostatic challenge, many tissues undergo phenotypic changes such as the increased growth and secretion of the zona glomerulosa following dietary sodium restriction or the atrophy of the zona fasciculata and zona reticularis following hypophysectomy (7). Although comparable responses in chromaffin cell phenotype are not as well characterized, these cells clearly exhibit plasticity of neurotransmitter and neuropeptide expression during development and aging (9, 14). Acute depolarization of cultured bovine chromaffin cells induces rapid filopodial formation (26), whereas chronic depolarization induces neurite formation (35), suggesting that depolarization may stimulate the modification of morphology and functional connections. Environmental factors such as nerve growth factors (NGF) and glucocorticoids are also reported to affect chromaffin cell phenotype (1, 13).

In the process of our investigation of the effect of chromaffin cell activation on steroid secretion, we noted a resultant dramatic change in chromaffin cell morphology into a more neuronal-like configuration. In the present study, we characterize cellular interactions between chromaffin and adrenocortical cells following chromaffin cell activation using carbamylcholine (CCh), a cholinergic agonist that affects both nicotinic and muscarinic cholinergic receptors, and veratridine. Veratridine acts on voltage-sensitive sodium channels to increase sodium influx and thus depolarizes chromaffin cells and induces catecholamine release (21). Adrenocortical cells lack voltage-dependent sodium channels (27) and are, therefore, unaffected directly by veratridine, as was verified in our previous studies (30). In the present study, we show that short-term chromaffin cell activation increased Fos protein expression and corticosterone secretion, indicating that chromaffin cell modulation of adrenocortical cells occurs at the transcriptional level. The more neuronal-like morphology of chromaffin cells and increased apparent contacts between chromaffin and adrenocortical cells observed following prolonged chromaffin cell activation suggest that activity-related plasticity enhances their interactions.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Cell culture. Frog (Rana pipiens) adrenal cocultures were prepared as described previously (31). Briefly, the adrenal tissue was dissected from the ventral surface of the kidney, minced, and dissociated with enzymes (2 mg/ml collagenase A, 1.6 mg/ml dispase, and 0.1 mg/ml DNase I; all from Boehringer Mannheim, Indianapolis, IN), triturated, and then cultured at a density of 1.3 × 10⁴ cell/cm² on glass coverslips (Bellco Glass, Vineland, NJ) or 35-mm culture dishes (Falcon Plastics) on laminin (Sigma, St. Louis, MO) plus poly-D-lysine substrates (Sigma) in supplemented 55% Leibowitz-15 medium (L-15; Sigma) (31). The medium also contained 5% fetal bovine serum (Atlantic Biologicals, Norcross, GA), mouse NGF 2.5 s (50 ng/ml; Alomone Labs, Jerusalem, Israel), and basic fibroblast growth factor (human recombinant, 10 ng/ml; Bachem, Torrance, CA). Supplemented medium with histamine (2 µg/ml, Sigma) to optimize neurite outgrowth, adrenocorticotropic hormone (ACTH) 1-24 (10¹⁰ M, Sigma), 5% fetal calf serum to maintain basal steroidogenesis, and sometimes cytosine -D-arabinofuranoside (10⁶ M) to inhibit cell proliferation (31) were used for medium changes. Veratridine (2-100 µM), 50 µM veratridine plus 1 µM tetrodotoxin (TTX), or 2 mM CCh were added as described in RESULTS (all from Sigma).

Percoll purification of adrenal cells. For separation of adrenocortical from chromaffin cells, dissociated cells were washed and resuspended in modified frog Ringer solution that contained bovine serum albumin (BSA), low calcium, and no magnesium (114 mM NaCl, 2 mM KCl, 6.2 mM NaHCO₃, 50 µM CaCl₂, 2.5 mg/ml BSA, 5.5 mM glucose, 1 × 10⁵ U/l penicillin, 100 mg/l streptomycin, and 2.5 mg/l amphotericin, pH 7.3, saturated with air). The cell suspension was filtered through a 53-µm Spectra/Mesh nylon filter (Spectrum, Houston, TX) and then layered on top of a discontinuous Percoll gradient (22) modified for frog. Percoll (Pharmacia Biotech, Uppsala, Sweden) was adjusted to pH 7.3 and 230 mmol/kg osmolality, diluted to 1.025 g/ml, 1.055 g/ml, and 1.085 g/ml using the modified Ringer (vide supra), and underlayered in 3-ml increments beneath modified Ringer solution in 15-ml polypropylene conical tubes (Corning, Cambridge, MA). Cells from no more than three frogs were applied to each column. The gradient was centrifuged (600 g, 15 min), and the adrenocortical cell fraction located at the 1.025/1.055-g/ml interface and the chromaffin plus adrenocortical cell fraction at the 1.055/1.085-g/ml interface were collected and washed in modified Ringer. The cells were cultured in 24-well plates (Costar, Cambridge, MA) or 35-mm culture dishes (Falcon Plastics) as outlined above. The purity of each fraction was confirmed by identification of cell types with phase-contrast microscopy and immunocytochemistry for P450 11-hydroxylase and tyrosine hydroxylase (31).

Immunocytochemistry. In initial experiments, we found that 2-100 µM veratridine induced a dose-dependent alteration in neuronal morphology with no toxicity, as assessed by morphology and cell number; subsequently, 50 µM was used in all experiments. In experiments on the effect of activation on chromaffin cells, chromaffin cells were depolarized with 50 µM veratridine for 24, 48, 72, or 96 h (see Fig. 2A) and fixed in 2.6% paraformaldehyde in 0.067 M phosphate buffer, pH 7.4, on day 7 for immunocytochemistry (31). Chromaffin cells were identified and characterized using simultaneous immunostaining for tyrosine hydroxylase (1:1,000; Eugene Tech International, Ridgefield, NJ) and the neuronal marker linc (undiluted; Developmental Studies Hybridoma Bank, maintained by Dept. of Pharmacology and Molecular Sciences, Johns Hopkins Univ. School of Medicine, Baltimore, MD and Dept. of Biological Sciences, Univ. of Iowa, Iowa City, IA); adrenocortical cells were identified by histochemistry of 3-hydroxysteroid dehydrogenase or by immunocytochemistry for P450 11-hydroxylase (1:1,000; gift from Dr. Mitsuhiro Okamoto) as described previously (31). Primary antibodies were visualized using CY3-conjugated anti-rabbit IgG (1:2,000; Jackson ImmunoResearch, West Grove, PA) and FITC-conjugated anti-mouse IgG (1:1,000; Boehringer Mannheim). Parameters used to assess chromaffin morphology included neurite number, neurite length, number of branches per neurite (measured as the maximum number per neurite on each cell), and relative intensity (scale of 1 to 4) of tyrosine hydroxylase immunostaining measured on more than two coverslips per treatment and more than 50 cells per coverslip. The data were analyzed by analysis of variance followed by Fisher's protected least-significant differences test for multiple comparisons between groups. The percentage of immunoreactive chromaffin cells overlapping or directly adjacent to adrenocortical cells was also determined and analyzed by statistical methods for comparing two binomial populations.

Time-lapse video microscopy. The dynamic cellular interactions of chromaffin and adrenocortical cells were recorded with time-lapse video microscopy. Cocultures or Percoll-purified adrenocortical cells grown in 35-mm dishes were examined for up to 48 h under phase-contrast microscopy with a Nikon Diaphot inverted microscope controlled by an electronic timer to minimize exposure to light. Images were collected at 15- to 60-min intervals using a Sony charge-coupled device video camera [Quick Capture frame grabber board (Data Translation) on a Power Macintosh 7100/66 computer] and analyzed using the public domain NIH Image program version 1.61 (developed at the U.S. National Institutes of Health and available on the Internet at http://rsb. info.nih.gov/nih-image/). Time-lapse microscopy of identified adrenocortical and chromaffin cells (31) was first carried out in unstimulated (24 h) conditions followed by stimulated (50 µM veratridine, 24 h) conditions, thus controlling for individual variation in responsiveness to veratridine treatment. To control for sodium channel-related specificity of veratridine stimulation, another set of cultures was recorded in the presence of 2 µM TTX plus 50 µM veratridine (24 h) followed by veratridine alone (50 µM for an additional 24 h). The entire sequences of images were used for following individual identified cells in Figs. 3, 5, and 6; however, only frames from representative or greater intervals are shown.

HPLC analysis of catecholamines and indolamines. The effect of veratridine-induced depolarization on catecholamine and indolamine secretion was determined by HPLC (high-performance liquid chromatography; BioAnalytical Systems, West Lafayette, IN). Medium was collected every 24 h from untreated and 50 µM veratridine-treated cultures and immediately frozen. Epinephrine, norepinephrine, and the internal standard, 3,4-dihydroxybenzylamine (1.2 ng/ml), were extracted by adsorption to aluminum oxide; analysis of serotonin and homovanillic acid was performed on unextracted medium. The catecholamines and serotonin were separated by reverse-phase HPLC with a C18 column maintained at 37°C using mobile phase, comprising 8% acetonitrile and 92% 0.15 M monochloroacetate buffer, pH 3.0, with 0.86 mM sodium octyl sulfate and 0.7 mM disodium EDTA, and measured by electrochemical detection.

Aldosterone and corticosterone. Following 90-min stimulation with 50 µM veratridine and unstimulated controls, the medium was collected and immediately frozen for subsequent radioimmunoassay of corticosterone (ICN Biomedicals, Costa Mesa, CA) and aldosterone (Coat-A-Count; Diagnostic Products, Los Angeles, CA) content. The number of 11-hydroxylase immunoreactive adrenocortical cells was counted in three fields and extrapolated to estimate the total number per well. The Student's t-test was used for comparison of basal and stimulated cultures.

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Direct effects of veratridine on chromaffin and adrenocortical cell secretion. Adrenocortical cells were purified on a Percoll gradient to verify that veratridine did not directly affect steroid secretion. In purified cultures, 24-h veratridine treatment resulted in no significant increase in corticosterone secretion compared with untreated controls (control: 397 ± 47 pg/1,000 cells; veratridine: 518 ± 92 pg/1,000 cells; n = 4), whereas in cocultures it increased secretion approximately sevenfold.

1

1

1

1

1

1

1

1

1

1

1

1

Morphological effects of chromaffin cell activation. Chromaffin cell activation with 50 µM veratridine dramatically enhanced neuronal morphology (Fig. 1). Veratridine treatment of cocultures for 24, 48, 72, or 96 h (see Fig. 2A) resulted in time-dependent changes in chromaffin cell morphology, including increased 1) number of neurites per chromaffin cell, 2) neurite length, 3) number of branches per neurite, and 4) tyrosine hydroxylase immunoreactivity (Fig. 2, B-E). Chromaffin cell activation for >48 h (Groups IV and V) also significantly increased the number of contacts between chromaffin and adrenocortical cells (Fig. 2F). It is noteworthy that 48-h veratridine treatment immediately before fixation (Group VI) resulted in greater enhancement of neuronal morphology compared with those in which veratridine-induced activation had ceased for 3 days (Group II), indicating reversibility and plasticity of the cells.

View larger version (30K): [in this window] [in a new window]   Fig. 1.   Chromaffin cells differentiate to a more neuronal phenotype following depolarization with veratridine. Cocultures were treated for 36 h with 50 µM veratridine (C and D) and compared with untreated cocultures (A and B). The increased neurite length and branching of neurites evident in these linc-immunoreactive chromaffin cells are examples of differentiation following depolarization. Bar = 25 µm.

View larger version (47K): [in this window] [in a new window]   Fig. 2.   The effect of veratridine-induced depolarization on chromaffin cell morphology is time dependent and reversible. Adrenal cocultures were treated with 50 µM veratridine at times indicated by the bars in A. All groups were fixed on day 7. Chromaffin cell activation resulted in time-dependent enhancement of neurite number (B), neurite length (C), number of branches per neurite (D), tyrosine hydroxylase immunoreactivity (TOH-IR; on a relative scale of 1-4; E), and the number of contacts formed between chromaffin and adrenocortical cells (F). *P  0.05; **P  0.01.

Time-lapse microscopy. Time-lapse video microscopy of cocultures was used to document the dynamic interactions between chromaffin and adrenocortical cells under basal and stimulated conditions. Although chromaffin cells in coculture routinely extended neurites (Fig. 3A), these seldom developed into long neurites with growth cones, as observed with prolonged activation (Fig. 3B and Fig. 4). The pattern of neurite outgrowth in unstimulated conditions often showed repetitive extension and retraction, resulting in little overall neurite change. In contrast, in veratridine-treated cultures, neurites established growth cones and longer neurites, often forming contacts with adrenocortical cells (Fig. 3B and Fig. 4, A and B). The effect of veratridine stimulation was already evident within 30 min, and by 24 h, chromaffin cells established numerous contacts with adrenocortical cells and showed significant neurite outgrowth, growth cones, varicosities, and round soma. Figure 3A shows a characteristic unstimulated chromaffin cell that began with a neurite extended to an unidentified nonneuronal cell. The neurite was retracted after 2 h and remained so throughout the rest of the 18 h of observation. Within 3 h of veratridine treatment (Fig. 3B), this same chromaffin cell extended neurites that developed into complex processes and established contacts with adrenocortical cells (Fig. 3B and Fig. 4B). Although the distance between the chromaffin cell soma and the adrenocortical cell labeled "2" eventually increased after 12 h, contact was maintained by the neurite (see Fig. 4B). Neurite outgrowth in the presence of veratridine plus TTX, which antagonizes the effects of veratridine on sodium channels, was similar to that of control cultures (Fig. 5A). TTX reversibly blocked the effects of veratridine, as shown by the neurite outgrowth and branching that occurred during the 24 h following removal of TTX (Fig. 5B).

View larger version (163K): [in this window] [in a new window]   Fig. 3.   Time-lapse video microscopy shows dynamic interactions between chromaffin cells (arrows) and adrenocortical cells (arrowheads). Images were collected at 15-min intervals, but only representative frames from 3-h intervals are shown here. The time from the beginning of each sequence is shown (lower left corner of each frame); see Fig. 6A for the 15-min intervals between 3 and 6 h. The same field of chromaffin cells and adrenocortical cells was studied: first for 24 h under control/unstimulated conditions (A), followed by 24 h of veratridine treatment (B). Neurite outgrowth was minimal in unstimulated conditions (A) and was rapidly and dramatically enhanced with veratridine treatment, resulting in increased chromaffin-adrenocortical interactions (B). Adrenocortical cells (arrows labeled 1, 2, and 3) in the presence of both unstimulated and veratridine-stimulated chromaffin cells show migration toward the central chromaffin cell (arrow).

View larger version (95K): [in this window] [in a new window]   Fig. 4.   Images from time-lapse microscopy sequences demonstrate the enhancement of growth cones (arrows), complex neurites, varicosities, and chromaffin-adrenocortical cell contacts (arrowheads) induced by 14-h (A) and 22-h (B) treatment with 50 µM veratridine. Note that the cell shown in B is the same cell shown in Fig. 3.

View larger version (133K): [in this window] [in a new window]   Fig. 5.   Veratridine-induced changes in chromaffin cell morphology are reversibly blocked by tetrodotoxin (TTX). In this series of time-lapse images, TTX + 50 µM veratridine was added 0-24 h (A); TTX was then removed and replaced with veratridine (B). In the presence of TTX + veratridine, the chromaffin cell (white arrow) undergoes very little change in morphology and only a slight increase in neurite length. On removal of TTX, the veratridine-stimulated chromaffin cell develops and maintains multiple neurites (black arrows). Note that adrenocortical cells 2 and 4 (numbered arrowheads) in A divide into multiple cells that migrate. Adrenocortical cells 5 (A) and 1 and 3 (B) also undergo migration, confirming that neuronal depolarization appears to have little effect on adrenocortical cell migration. Only representative frames at 3-h intervals of the same field are shown. The time from the beginning of each sequence is indicated (lower left corner of each frame).

View larger version (166K): [in this window] [in a new window]   Fig. 6.   Adrenocortical cell migration is greater in cocultures (A) than in purified adrenocortical cell cultures (B), as illustrated by these images captured at 15-min intervals. For example, in A, adrenocortical cell 2 (numbered arrowhead) moves toward a chromaffin cell (arrow) and appears to develop filopodia directed toward the chromaffin cell. In later frames (see Fig. 2A), the adrenocortical cell moves away from the chromaffin cell. The images of purified adrenocortical cells in B demonstrate that little adrenocortical cell migration occurs in the absence of chromaffin cells. Adrenocortical cell 1 (numbered arrowhead) undergoes very little movement or change in morphology throughout the series while adrenocortical cells 2 and 3 show a small amount of movement away from the adrenocortical cell located between them. The purity of the Percoll-purified adrenocortical cell fraction is demonstrated by the presence of only adrenocortical cells in the micrographs, compared with the presence of adrenocortical cells, chromaffin cells, and other unidentified cell types in the other figures.

Short-term effects of chromaffin cell activation. Because ACTH-stimulated steroidogenesis was correlated with a twofold increase in the percentage of adrenocortical cells expressing the Fos protein, we used Fos expression to determine whether chromaffin cell stimulation of adrenocortical cell function occurred at the transcriptional level. In two separate experiments, veratridine and CCh stimulation of chromaffin cells resulted in an increase in the number of adrenocortical cells expressing Fos within 60-90 min after stimulation. In the first experiment, CCh and veratridine stimulation resulted in significant increases in the number of adrenocortical cells expressing Fos (P < 0.01, compared with control), with the greatest effect evident at 60 and 90 min after CCh (Fig. 7A; 37% increase vs. untreated) and veratridine (Fig. 7B; 25% increase), respectively. We suspect that the relatively high basal (unstimulated control) expression of Fos in this experiment was due to stimulation by basal catecholamine secretion (see Direct effects of veratridine on chromaffin and adrenocortical cell secretion). When we repeated this experiment, adrenocortical cells showed lower (23 ± 3%) basal (unstimulated control) expression of Fos and increased Fos expression at 60 and 90 min following veratridine or CCh (data not shown). Corticosteroid secretion apparently increased concomitantly with Fos expression (corticosterone, 1.5 h; basal: 4.9 ± 0.8 pg/1,000 cells, veratridine: 13.0 ± 5.5 pg/1,000 cells; aldosterone, basal: 3.8 ± 1.5 pg/1,000 cells, veratridine: 5.3 ± 3.5 pg/1,000 cells).

View larger version (29K): [in this window] [in a new window]   Fig. 7.   Expression of the immediate early gene, c-fos, shows adrenocortical cells (AC) are activated following chromaffin cell activation by veratridine and carbamylcholine (CCh). Time course following 30-min chromaffin cell activation with both CCh (A) and veratridine (B) shows stimulation of Fos expression in adrenocortical cells (P0.01 compared with control) with highest Fos expression after 1 h (CCh) and 1.5 h (veratridine).

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Chromaffin cell activation in the frog adrenal cocultures resulted in dramatic modification of chromaffin cell phenotype, including differentiation of a more neuronal morphology and increased tyrosine hydroxylase expression, both likely to facilitate the interactions between chromaffin and adrenocortical cells. Indeed, our studies show an increased number of chromaffin-adrenocortical cell contacts and increased synaptic efficacy (30) following chronic chromaffin cell activation. Veratridine-induced activation of chromaffin cells stimulated their release of norepinephrine, epinephrine, and serotonin, and, in turn, adenocortical cell release of corticosteroids. VIP and other neuropeptides, expressed by chromaffin cells in the cocultures, are also likely to be released with depolarization. Norepinephrine, epinephrine, serotonin, and VIP have all been shown to increase corticosteroid secretion (10-12, 16, 17, 25, 38) and, thus are likely to mediate corticosterone secretion following chromaffin cell activation in coculture (30).

Depolarization-induced modifications of neuronal phenotypes have previously been demonstrated for several different types of neurons, including chromaffin cells, and are speculated to have broad implications for neural development (26, 35) and plasticity. The effect of chronic depolarization on chromaffin cells in the frog adrenal cell cocultures was comparable to that reported in rat chromaffin cells (35). Similar to the frog (31), rat chromaffin cell neurite outgrowth was unaffected by dexamethasone (13, 36) or by CCh, which causes only transient activation of chromaffin cells (5). Chromaffin cell depolarization inhibits the increase in adrenal leucine-enkephalin and preproenkephalin mRNA that occurs following denervation (23) and induces the synthesis of tyrosine hydroxylase (29), suggesting that trans-synaptic activity can differentially regulate neurotransmitter and neuropeptide expression. Our studies provide further evidence that depolarization, in addition to environmental factors such as NGF and glucocorticoids (1, 13), affects chromaffin cell phenotype.

The physiological relevance of the adrenal cortical-chromaffin cell coculture model is underscored by the evidence for reciprocal interactions between chromaffin and adrenocortical cells. The possible presence of an adrenocortical neurotrophic factor that affects chromaffin cell growth and phenotype is indicated by the extension of chromaffin cell neurites to adrenocortical cells and the enhanced neurite outgrowth that occurs following ACTH stimulation of steroidogenesis (31). The nature of this neurotrophic factor is as yet unidentified, but may be growth factors, adhesive proteins, and/or extracellular matrix proteins. This neurotrophic effect is likely to be effective in vivo and important during development and in the regulation of adrenocortical cell steroidogenesis and proliferation. Accordingly, we have recently demonstrated that dietary sodium restriction stimulates expansion of the distribution of VIP-containing adrenocortical nerve fibers concomitantly with expansion of the zona glomerulosa and is possibly mediated by a trophic effect of the zona glomerulosa cells (20). Basic fibroblast growth factors, previously shown to be present in glomerulosa cells (2) and to be neurotrophic for chromaffin cells (33), may mediate this effect.

Chromaffin cells also modulate adrenocortical cell organization and function, as shown by the migration of adrenocortical cells toward chromaffin cells and their effect on adrenocortical Fos expression and corticosterone secretion. The increased expression of the Fos protein indicates that adrenocortical cells are affected at the transcriptional level and may alter adrenocortical cell phenotype. Indeed, ACTH stimulation, also shown to increase Fos expression (40; unpublished observations), stimulates the expression of steroidogenic enzymes and thereby exerts a long-term effect on steroidogenic capability (32). Our observations of the migration of adrenocortical cells also have possible implications for the histogenesis of the adrenal cortex. A current well-supported hypothesis proposes that adrenocortical cells of the mammalian adrenal gland proliferate in the outer zone and move inward either by migration and/or mitotic pressure, with cells undergoing the phenotypic changes characteristic for each zone, with apoptosis finally occurring in the innermost zone (3). Whereas chemotactic assays have previously ascertained adrenocortical cell migration (28), our studies directly demonstrate adrenocortical cell migration using time-lapse microscopy. Laminin is a chemotactic factor for bovine fasciculata cells; however, its uniform distribution throughout the adrenal cortex suggests that it is likely not the attractant driving adrenocortical cell migration (28). Our observations of cell migration of adrenocortical cells grown on a laminin substrate support this; yet, the migration rate of adrenocortical cells in cocultures is greater than that of purified adrenocortical cells. The directed movement toward chromaffin cells in coculture, in contrast to the random movement of purified adrenocortical cells, leads us to speculate that the adrenocortical cells migrate toward a trophic stimulus expressed in the adrenal medulla.

In conclusion, our results provide further evidence for the reciprocal interactions between chromaffin and adrenocortical cells that affect differentiation and modulate corticosteroidogenesis. Furthermore, adrenal neuronal and adrenocortical plasticity have significant implications during development and during acute physiological challenges such that the cohesive function of the adrenal gland is optimized.