Transmembrane Potentials and Steroidogenesis in Normal and Neoplastic Human Adrenocortical Tissue JOHN LYMANGROVER,1 A. FRANCES PEARLMUTTER, ROBERTO FRANCOSAENZ,2 AND MURRAY SAFFRAN3 Departments of Biochemistry and Medicine, Medical College of Ohio, Toledo, Ohio 43614 Ommission of K+ from the medium caused hyperpolarization of the tumor cells, but the trans-membrane potentials did not reach the values of hyperpolarized nontumor cells. ACTH, added to the K+-free medium, caused little or no change in membrane potential of tumor cells except in one case of a virilizing adenoma, which responded very much like non-tumor tissue. Except for the virilizing adenoma, tumor tissue slices produced little or no detectable fluorogenic steroid, even in the presence of large amounts of ACTH or cyclic AMP. The virilizing adenoma responded with increased steroidogensis to ACTH and cyclic AMP. (J Clin Endocrinol Metab 41: 697, 1975)
ABSTRACT. Trans-membrane potentials and steroidogenesis were measured in superfused slices of non-tumor and neoplastic human adrenocortical tissue. Non-tumor tissue was obtained at the time of renal transplant or from tissue removed along with tumors. Non-tumor human adrenocortical tissue had electrophysiological and steroidogenic properties similar to those of the rat and rabbit. In normal medium ACTH stimulated steroidogenesis but had no effect on the membrane potential. In K+-free medium, the cells hyperpolarized, and subsequent addition of ACTH caused depolarization. Trans-membrane potentials of adrenocortical tumors were lower than those of non-tumor cells.
P
EARLMUTTER et al. (a) have shown that human adrenocortical tissue behaves remarkably like rat tissue in its steroidogenic response to ACTH and cyclic AMP. Does the human tissue also resemble other species in its electrophysiological properties? The voltage difference between the interior and exterior of a cell depends upon the existence of an ionic gradient across the cell membrane. Nerve and muscle cells maintain a gradient that results in a voltage of 70 or more mV, negative inside. Most visceral cells have membrane potentials of —20 to —40 mV. Activation of a nerve or muscle cell is accompanied by a characteristic sequence of changes in membrane potential, the action potential, which appears as a train of rapid depolarizations of the membrane. Received November 8, 1974. Supported by NIH grant AM 14132 and GRS grant 94357. 1 Present address: Department of Physiology, University of Cincinnati College of Medicine, Cincinnati, Ohio 45219. 2 Department of Medicine. 3 Address correspondence to: M. Saffran, Department of Biochemistry, Medical College of Ohio, Toledo, Ohio 43614.
The electrical properties of human adrenocortical cells have not been studied. We took advantage of the availability of samples of human adrenocortical tissue either at the time of renal transplant or removal of adrenal tumors to compare the electrophysiological and steroidogenic properties of non-tumor and tumor tissue. Our results show that the non-tumor human adrenal cortex is similar to that of other species. However, adrenal tumors had lower and more variable membrane potentials, depending on the type of tumor. Materials and Methods Non-tumor adrenal tissue. Non-tumor adrenal tissue was obtained from 2 kidney transplant donors, 3 non-tumor areas removed along with adrenal tumors, and the cortical areas removed with a pheochromocytoma and an adrenal medullary hematoma. Neoplastic adrenal tissue. Tumor tissue was obtained at the time of adrenalectomy or biopsy from 5 patients: 2 carcinomas, 1 probable carcinoma, and 2 adenomas (Table 1).
697
Tissue slices. The adrenal was sliced (0.5 mm) with a Stadie-Riggs tissue slicer and 30-100 mg portions were placed in the tissue holder. No
2 months
1 year
28
13
28
22
18
24.5
10
20
(36)
38
16
11
11
17
(310)
not done
16
1
219
27
29
4-8 mg incr.
not done
31
14
31
300500%
ACTH Stimulation 3 17 17 OH KS mg/24 hr.
195
not done
not done
3.5
35
21
0.7
not done
171
104
124
339
0.08
43
32°
0.14
22
not done
12
M Z N > Z
z n 9
w
r
33
W
33 PI 33
z o
00
CO
HUMAN ADRENAL MEMBRANE POTENTIALS attempt was made to separate the adrenal zones. Supervision with medium began within 30 min of obtaining the tissue from the operating room.
TABLE 2. Fluorogenic properties of testosterone and other C19 steroids in the analytical system, compared with corticosterone Steroid
Superfusion system ir steroid measurement. The superfusion system for the adrenal tissue has been described in detail (2,3). Slices were placed in a tissue holder and superfused at a rate of 4 ml/min with a solution containing Na+, 144 niM; K+, 4.7 mM; Ca ++ , 2.7 mM; Cl", HCXV, 25 mM; Mg ++ , 1.2 mM; H2PO4", 1.2 mM; glucose, 11 mM; bubbled with 95% O2/ 5% CO2. Potassium-free medium was prepared by omitting 4.7 mM KC1 from the solution. The effluent from the tissue holder was pumped into the analytical system, which is designed to record continuously the fluorescence formed from corticosteroids released into a stream of medium by adrenal slices. The analytical system extracts the steroids from the medium with a stream of methylene chloride, and the methylene chloride is in turn extracted with the ethanolH2SO4 reagent. Fluorescence formed by heating the ethanol-H2SO4 extract is detected in the flow cell of a Turner Model 111 fluorometer and is recorded on a strip chart. In this procedure cortisol has 44% of the fluorescence of corticosterone and aldosterone does not fluoresce. Testosterone and several other C19 steroids did not fluoresce very much (Table 2). Because the normal human adrenal secretes approximately 90% cortisol (4), steroidogenesis is expressed as cortisol. The apparatus is adjusted to zero fluorescence with the medium alone, and is calibrated with known amounts of cortisol. The fluorescence due to the steroid appears as a peak on the record. The area under the peak is directly proportional to the amount of steroid. After calibration of the system, the tissue holder containing adrenal slices is introduced into the apparatus. An increase in fluorescence appears immediately due to the steroids released by the tissue. This release becomes stabilized in about 1 h, and the base-line production of corticoids is established, approximately 0.16 ng/mg/min, expressed as cortisol. Trans-membrane potentials. The procedure for trans-membrane potential recordings was that described by Matthews (5). Finely drawn glass electrodes, filled with 1.5 M K-citrate solution and with initial resistances of 50-100 megohms, were aged in 0.9% NaCl solution for several days
699
Corticosterone Testosterone Dehydroepiandrosterone Androsterone Androstan-3,17-dione Androst-5-ene-3,7-dione Androst-5-ene-3y3, 17/3-diol Androst-4-ene-3,17-dione
% Fluorescence 100.00 0.41 0.37 0.01 0.05 0.47 0.43 0.01
prior to use. The electrodes were positioned with the aid of a dissecting microscope. Trans-membrane potentials were visualized and recorded with a Bioelectric P-1A amplifier system and a Tektronix 5031 storage oscilloscope and Westronics D 11A pen recorder. When the tip of the electrode penetrates the membrane, a sharp deflection of the trace on the oscilloscope screen or the recording paper occurs. If the seal around the electrode is imperfect, the tip is quickly ejected from the cell and the trace returns to base line. A good seal is signalled by a potential that rises quickly to a nearly final level, settles within a few seconds to a slightly higher value, and maintains that value for 10 s or more. The quality of the recording is dependent upon the quality of the electrodes. Often a good electrode will seal into a cell and remain there for an hour or more. The emergence of an electrode from a cell is signalled by the rapid return of the trace to base line. Under the dissecting microscope it is relatively easy to differentiate between medullary, capsular, and fasciculata-reticularis zones in the human adrenal gland and to direct the tip of the electrode into a population of cells in the desired zone. No attempt was made to distinguish between zona fasciculata and zona reticularis. Samples of tumor slices were taken from the center of a cross-section of the tumor to avoid, as much as possible, contamination with extratumor tissue. In some experiments tumor and nontumor tissue from the same patient were studied simultaneously. The tissue was superfused in Krebs medium at 37 C for 1-4 h until consistent potentials were observed before the experiments were begun. The tissue was usually studied for 24 h and in some instances for up to 48 h. Steroidogenesis and membrane potentials were measured simul-
700
LYMANGROVER, PEARLMUTTER, FRANCO-SAENZ AND SAFFRAN
taneously on different slices of the same tissue sample. Membrane potentials are expressed as mean and standard errors of a population of cells in the tissue slices. Statistical analysis was performed using Student's t test for unpaired data; P values