Can. J. Physiol. Pharmacol. Downloaded from www.nrcresearchpress.com by UNIV CHICAGO on 11/15/14 For personal use only.
An Adrenal Cell System Suitable for Long Term Study of Factors Affecting Steroidogenesis CARMAN S. PRICE,^ JAMESL. RUSE, AND JOHN C. LAID LAW^ Department of Medicine and In.rtit~rteo f Medical Science, University of Toronto, Toronto, Cbntario M5G 1.48 Received August 2 1, 1974
PRICE,C. S., RUSE,J. I,., and I,AIDI~AW, J . C. 1975. A11 adrenal cell system suitable for long term study of factors affecting steroidogenesis. Can. J. Yhysiol. Pharmacol. 53,531-539. Cell suspensions of normal adult rat adrenals have been prepared by trypsinization and incubated in Ham's nutrient mixture F10 containing horse serum, fetal calf serum, and lima bean trypsin inhibitor. In most experiments culture medium was not changed during incubation. In this system the number of cells fell to 5W% after 2 days, then slowly declined to 2% after I month of incubation. A corticosterone (B) response was seen to as little as 5 pU of ACTH per millilitre, a concentration which is within the range found in normal rat serum. With maximal stimulation (100 mU ACTHIml) the rate of accumulation was highest during the first 24 h then slowly decreased over the fc>llowing9 days. When in separate experiments AC'TH was added after various times of incubation up to 3 weeks, there was a B response which continued for as long as 1 week after the ACTH was added; the later the time at which ACTH was added the lesser was the initial B response and the longer the lag period before a substantial response occurred. In cell suspensions in medium containing 5.0 mequiv. of KS per litre, aldosterone content increased for approximately 24 h, then showed little or no change over the next 9 days. With increased Kt concentration, aldosterone was found in greater amounts and accumulation continued for longer periods, both without and with ACTH. This adrenal cell system appears suitable for long term study of factors affecting steroidogenesis. PRICE, C. S., RUSE.J. I,. et L,AIDL.~\w, J . C. 1975. An adrenal cell system suitable for long term study of factors affecting steroidogenesis. Can. J. Physiol. Pharmacol. 53,53 1-539. Des suspensions cellulaires de surrenales de rat adulte sont preparees par trypsinisation, et incubees dans un milieu nutritif F,, de Ham contenant du serum de cheval, du siri~rnfoetal de veau et l'inhibiteur de la trypsine de la feve "lima". Dans la plupart des experiences, le milieu de culture n'est pas changi au cours de I'incubation. Dans ce systPme, le nomhre des cellules tombe a 50% apres 2 jours. puis diminue lentement jusqu'a 20% apres un mois d'incubation. Une secretion de corticosterone est constatee en reponse a la concentration rninime de 5 pUlml (reponse B), qui est dans les limites des taux seriques normaux chez le rat. Apres une stimulation maximale (100 m u d'ACTH/ml), le taux d'accumulation est le plus eleve durant les premieres 24 h, puis decline dans la periode des 9 jours suivants. Dans des experiences separees, on ajoute I'ACTH apres des temps d'incubation allant jusqu'i trois semaines. On constate alors une reponse B prolongie jusqu'8 une semaine aprks I'addition d' ACTH. Plus I'ACTH est ajoute tard, moins la reponse B initiale est elevie, et plus la periode de latence de cette reponse est longue. Dans les suspensions cellulaires en milieux contenant 5.0 inequiv. de potassium par litre, le contenu en aldosterone est augmente pendant environ 24 h, mais au cours des 9 jours suivants, il y a peu de changements. La concentration d'aldostirone dans le milieu augmente avec la concense fait sur de plus longues periodes, avec ou sans ACTH. tration de potassium, et I'acct~n~ulation Ce systeme de suspensions cellulaires snrrenaliennes semble utilisable pour I'etude a long terme des facteurs affectant la steroidogCntse. [Traduit par le journal]
Preparations of isolated normal adult adrenal cells (Halkerston and Feinstein 1968; Kloppenborg ct al. 1968; Swallow and Sayers 1969; 'Supported by a grant from the Medical Research Council of Canada. "resented in part at the 55th meeting of the Endocrine Society, June 1973. "older of a Fellowship from the Medical Research Council of Canada. 'Please send reprint requests to: Dr. 1. C . Laidlaw, Medical Sciences Building, University of Toronto. Toronto, Ontario. ,%
Saycrs et al. 1971; Peytremann ct al. 1973) have proved to be of great value in the study of the function of the adrenocortical cell. Systems such as these were designed for experirnents requiring only short term incubations of the cells and are not suitable for the in vitro study of events which in vivo require many hours or even days for their expression. Two such events are the suppression of ddosterone production (Ruse et ". 1972) and the fall of aldosterone secretion, after an initial rise, in the face of continuing adrenocortico-
Can. J. Physiol. Pharmacol. Downloaded from www.nrcresearchpress.com by UNIV CHICAGO on 11/15/14 For personal use only.
532
CAN. J. PHYSIOL. PHARMACOL. VOL. 53, 1975
tropic hor~none(ACTH) infusion (Tucci et al. 1967). To study the adrenal component of these phenomena, we have developed a system which, like that of Armato and Nussdorfer (1972), Kahri ( 1973), O'Hare and Neville (1973a, 1973h), and Roos (1974), permits survival in vitro of functioning adrenocortical cells for several weeks. Isolated cells were prepared from whole normal adrenal glands of adult rat by a modification of the method described by Swallow and Sayers (1969) and incubated in Ham's nutrient mixture Ff 0, supplemented with horse and fetal calf serum, for periods up to 30 days. In this paper we present our observations on cell survival, corticosterone and aldosterone accumulation. and the effects of ACTH and K + concentration on the accumulation of these steroids.
Materials and Methods 1-yophilized trypsin (TRL) and lima-bean trypsin inhibitor (LRI) were obtained from Worthington Biochemical Corporation, Freehold, N.J., and heatinactivated horse and fetal calf serum from Grand Island Biological Company, Grand Island, N.Y. Sterile disposable plastic tubes and petri dishes (Optilux) were obtained from Falcon Plastics, Oxnard, Calif. Sterile disposable pipettes and membrane filter units (0.45 p n ~ ) were purchased from Fisher Scientific. Toronto, Ont. Corticosterone and aldosterone were purchased from Tkapharm, Ramat-Gan, Israel and used as received. 14-"C]Corticosterone (S.A. 57.3 mCi/mmol) , [I ,2-"Hlcorticosterone (S.A. 50 Ci/mmol) , [4-'CCldosterone (S.A. 55 mCi/mmol ), and [I ,2-Wlaldosterone (S.A. 54 Ci/mmol) were supplied by New England Nuclear Canada, Dorval. Que. Purity checks and storage conditions were as described by Levine et ul. (1972). Male Wistsr rats weighing from 250 to 300 g were obtained from High Oak Ranch, Aurora, Ont. They were maintained on Purina rat chow and water ail lihitltnl for approximately 1 week before sacrifice. Solvents and chemicals were certified or reagent grade. All solvents except n-butyl acetate, petroleum ether, and propylene glycol were distilled prior to use. Bovine albumin (fraction V ) was purchased from .4rmour Pharmaceutical Co., Chicago. Ill. Synthetic ACTHI-., (Cortrosyn, 0.1 U.S.P. unit per microgram, Organon Tnc., Montreal, Que.) was used throughout. Radioactivity was measured as previously described (Levine et al. 1972). Aqueous samples were counted in 15 ml of a scintillation mixture consisting of 40 ml of ethanol, 80 ml of Liquifluor (New England Nuclear), 1880 ml toluene and 500 ml of Biosolv (BBS11 1, Beckman Instrument Co., Fullerton, Calif.). Incuburion Mediirrn Adrenal cells were incubated in Ham's nutrient
mixture F1 0 (Grand Island Biological Co. ) supplemented with 15% horse serum and 2.5% fetal calf serum. F10 was purchased in the powder form, without sodium and potassium chloride. Except where noted, sodium, potassium, and calcium concentrations were adjusted to 140 mequiv. /litre, 5.0 mequiv. /litre, and 2.5 mM, respectively. These values approximate the concentrations measured in the serum of the rats used in our studies. The final chloride concentration was 115 mequiv. /litre, about 12 mequiv. higher than the mean concentration in the serum of the rats. Immediately prior to use, lima-bean trypsin inhibitor (1 g/litre) was added and the solution sterilized by filtration through a 0.45 pm membrane filter. In the presence of 5% COz in air, the pH of the medium was 7.4.
Preyuration and Incubaticrrz o f CclI Suspension F Adrenal cell suspensions were prepared by a minor modification of the method described by Swallow and Sayers ( 1969). Rats were killed by cervical dislocation. The adrenals were removed and cleaned as quickly as possible, then minced and placed in 20 ml of 0.25% trypsin solution in Krebs-Ringer bicarbonate with glucose. The solution was agitated with a magnetic stirrer and stirring bar. After 20 min the supernatant was removed. The incubation procedure was performed a total of six times. The pooled supernatants were centrifuged for 30 min at 100 X g and the resultant pellets resuspended in the incubation medium. Approximately 5 X 1 0 cells were obtained per adrenal gland. Cells were plated in 35 X 10 mm petri dishes in a total volume of 1.0 ml: 0.9 ml of the cell suspension and either 0.1 ml of medium or 0.1 ml of medium eontaining ACTH. Two of these small dishes, together with a third dish containing water, were placed in a larger covered petri dish and incubated at 37 " C in a water-saturated atmosphere consisting of 5 % CO:! in air. Evaporative losses and pH changes were negligible throughout the incubations. No antibiotics were added to the incubations; about 10% of incubations showed visible signs of infection and were discarded. Except where noted, medium was not changed during the course of incubation. The entire contents of a given dish were used either for cell counting or for steroid measurements. Cell Counting Aliquots of the original unincubated cell suspensions were counted directly using a Fuchs-Rosenthal chamber. Cells which had been incubated were counted after treatment with trypsin (final concentration 0.7 % ). These cell counts, therefore, included both cells that were free in the medium and cells that had been released from the bottom of the dish by the trypsin treatment. The time of the exposure to trypsin required to release adherent cells varied from a few minutes to 4 h. Further incubations with trypsin for as long as 3 h did not affect cell counts. Only cells excluding Trypan Blue were included in the count. At each time point in the experiments reported in this paper, cell counts on duplicate incubations were obtained. The coefficient of variation of these duplicate counts was 7.1 %, (n = 39).
Can. J. Physiol. Pharmacol. Downloaded from www.nrcresearchpress.com by UNIV CHICAGO on 11/15/14 For personal use only.
PRICE E'T AL.: L,ONCi TERM INCUBATICIN O F AIIKENAL, CELLS
Harvesting oj Cells Cells were harvested after exposure to trypsin (final concentration (3.7%) for 1-3 h. The pooled cells were centrifuged at 4 " C and 100 X g 30 min, resuspended in fresh medium and re-incubated in the same way as the original suspension. Recovery of cells varied from 75-9096. Meusurernetzt of Steroids The contents of incubation vessels were transferred to 60 ml extraction tubes, together with 1 ml of medium and 3 ml of water (for rinsing the vessels), and extracted with 15 ml of dichloromethane. The latter was washed once with 1 ml of 0.1 N NaBH. Extraction of a mixture of 2 ml of medium and 3 ml of water provided material for blank measurements. Corticosterone was measured by the method of Silber et nl. (1958). Standard solutions, prepared by dissolving known amounts of corticosterone in medium, were extrhcted in the same manner as the samples and blanks. Fluorescence was measured in an Aminco-Bowman spectrophotofluorometer et excitation and emission wave lengths of 464 and 520 pm, respectively. All measurements were done on duplicate incubations. The coefficient of variation of these duplicate measurements was 1196. Inter-assay variation, based on 11 assays of a single pool of cells was 5.5%. Two pools of cells were incubated for 3 weeks, one with and one without ACTW, and assayed by the fluorometric method and a double isotope method (Levine et ul. 1972). The latter method involves addition of [14-'TCcorticosterone, extraction with methylene chloride, acetylation with tritiated acetic anhydride of known specific activity, oxidation with chromium trioxide, and purification to constant specific activity through a series of thin-layer chromatograms. There was no significant difference in the results of the two methods. Aldosterone was measured by radioimmunoassay as described by Underwood and Williams (1972), with minor modifications. A small quantity of tritiated aldosterone was added to the sample. Steroids were extracted from the medium with methylene chloride and chromatographed on paper in the Bush 5 system. Aldosterone was located on the chromatogram with a chromatogram scanner (Packard Instrument Company model 20Q1 ) . The aldosterone standards and the eluted samples and blanks were dissolved in the assay medium prior to transfer to the reaction vessels. The final reaction mixtures consisted of sample or standard in 0.5 ml of bovine albumin solution and 0.5 mi of 1 : 500 000 dilution of antiserum in borate buffer containing 3.6 X IQ' d.p.m. of [1,2-W]aldosterone. Standard curves were constructed by plotting the reciprocal of the fractions of radioactivity bound against the \tandard weights. Blank "weights" were subtracted from sample weights before the latter were corrected for recovery. In 17 assays, the mean and standard deviation of the blank were 8.3 zk 6.2 pg. In the sample measurements reported, except for those at zero incubation time, the fraction of the sample used in the binding reaction contained at least 75 gg. The coefficient of
533
variation of replicate aliquots ( a = 4 ) from three cell suspensions measured in the same assay varied from 8.0% at the 3 ng level to 3.2% at 11.5 ng. The aldosterone content of a pool of cells incubated for 72 h in the presence of ACTH was assayed by the immunoassay and a double isotope assay (Levine ct (11. 1972). The results obtained were 18.2 -t- 1.2 ng/ml (mean -+ S.D., tr = 14) and 17.6 -C 0.6 ng/ml ( 1 2 = 3 ) , respectively.
Results Microscopic Appearance of Cells Cells have been examined only in unstained preparations, using the light microscope; two types of cells have been distinguished. The cells of one type contained a variable number of granules in the cytoplasm ("granule cells") and were of relatively uniform size (1 0-1 5 pm). Cells of the second type lacked granules ("pale cells") and varied in size from 3 to 15 pm. In the first 48 h of incubations, the cells had a somewhat indistinct and ragged outline, but subsequently became more uniformly rounded and developed sharp, clear outlines. The latter appearance was maintained throughout the remainder of incubation. The cells became adherent to the surface of the dish rather slowly; virtually 100% of the cells were adherent by the fourth or fifth day. With continued incubation, the cells became less adherent, so that by the third week a considerable number were free in the medium and many of the re~llaindercould be dislodged by shaking the vessel. Cells which had been harvested and re-incubated in fresh medium were firmly adherent within 24 h, but subsequently displayed the same pattern of decreasing adherence. In no case was evidence of cell replication observed and, for the most part, the cells were not in contact with one another. Addition of adrenocorticotropic hormone (ACTH) to incubations had no effect on the appearance of the cells. However, in incubations with ACTH the cells appeared to be less adherent to the dish and formed occasional small clumps in which cells and cell debris were not always distinguishable. The cells in these clumps were easily separated by gentle pipetting. Cell Counts Over I Month of Incubation Figure 1 shows the cell counts obtained in a representative experiment during 30 days of incubation without ACTH. Over the first 2
Can. J. Physiol. Pharmacol. Downloaded from www.nrcresearchpress.com by UNIV CHICAGO on 11/15/14 For personal use only.
534
CAN. J. PHYSIOL. PHAWACOL. VOL. 53, 1975
FIG.2. Corticosterone and aldosterone accumulation in incubations (1.65 X 10" ceIls/ml) without and with ACTH (100 mTJ/ml). Each point represents the steroid content of a single dish. Measurements were made on two dishes at each time point; the curves join the means of the two measurements. Csrticosterone, .--a : aldssterone. l - - - - - -@ . INCJSATW IN DAYS
FIG. 1 . Cell counts over 1 month of incubation. Each point represents the mean of four setxirate counts on each of two incubations. Total cells, 0-0; granule cells, A - - - - - A: pale cells, 0- - -0. I
days of incubation, the total cell count fell precipitously and by the second day about 50% of the cells initially incubated remained. After 2 days the fall in total cell count was more gradual and at 30 days of incubation, approximately 20% of the cells remained. In experiments in which ACTH (100 mU/ml) was added at zero time, a similar pattern was seen, but after day 2, cell counts were 35-45% lower than in non-ACTH controls. The initial precipitous fall in total cell count was due, for the most part, to a fall in the number of pale cells. By 8 days of incubation only granule cells remained. Corticc~steroneRespotzse to ACTH Added at Zero Time Figure 2 shows corticosterone accumulation in cell suspensions during 10 days of incubation, in the absence and presence of ACTH (200 mU/ml). The results shown, which are from a single experiment, resemble those obtained in three similar studies. In the unstimulated suspensions, after an initial lag phase, corticosterone increased until day five, then ceased to accumulate. In the presence of
ACTH, corticosterone accumulation was much increased. The rate of accumulation was virtually constant for the first 3 days then slowly decreased over the remainder of the incubation. The period of constant corticosterone accumulation coincided with the rapid fall in cell number (see Fig. 1 ) . In similar studies the corticosterone response (the difference between ACTH-treated and control incubations) to 10, 100, 500, and 1000 m u of ACTH per millilitre was determined daily for 14 days. The maximum corticosterone response over this time period was reached at 100 m u of ACTH per millilitre, and this concentration was used in all other experiments except those involving studies of sensitivity to ACTH. Cell suspensions prepared in the usual way did not respond to less than 25 pU of ACTH per millilitre. However, cells washed once with trypsin-free buffer or medium before incubation were more sensitive. In each of six experiments, in which cell numbers varied from 6.5 x l o 4 to 4.3 x 10" the corticosterone content of incubations containing 5pU of ACTH per millilitre was greater than that of incubations without ACTH. The mean difference (-t 1 S.E.M.) was 97 t 40 ng of corticesterone, which is significant at the 5% level. The high standard error reflects variation in activity of the individual cell preparations.
PRICE ET AL.: LONG TERM INCUBATION OF ADRENAL CELLS
TABLE 1. Metabolism of [4-14C]corticosterone by cell suspensions
Can. J. Physiol. Pharmacol. Downloaded from www.nrcresearchpress.com by UNIV CHICAGO on 11/15/14 For personal use only.
Incubation time 72 h
Recovered 14C as corticosterone
(x)"
Sample Medium Medium f cells Medium cells + ACTH Medium Medium + cells Medium + cells ACTM Medium Medium + cells Medium + cells ACTH
+
1 week
+
3 weeks
+
"14-L4C]Corticosterone. 1.07 s 105d.p.m., was added to each dish at the beginning of incubation. Each number represents the mean of two determinations.
The test of significance was based on tho signs of the differences.
Correlatiorz Between Corticosterone Accurnnlation and Number of Cells Incubated A cell suspension prepared in the usual manner was diluted serially to provide three suspensions, which, when counted, were found to Sontain 165 x io3, 88 x lo3, and 18 x lo3 cells per millilitre. Duplicate 1.0 ml aliquots of each suspension were incubated without and with ACTH ( 108 mU/ml) for 24, 48, 72, 96, 168, and 240 h, and the corticosterone response to ACTH was determined. The results are shown in Fig. 3. It can be seen that, within the range of 18 x i 0 3 - 165 x lo3 cells per
CELL COUNT
x
~O'!/INCUBATION
FIG.3. Relationship between the number of cells incubated and corticosterone response to ACTH (100 mUirnl added at zero time). Each point represents the mean of corticosterone measurements on two dishes.
incubation, the relationship between corticosterone response to ACTH and cell numbers at each tirne'period is reasonably well described Except where noted, in by a straight studies of corticosterone the concentration of cells incubated fell within the range described in this experiment.
Metabolism of Corticosterone The possibility that corticosterone was being transformed to other compounds during incubation was investigated. Cell suspensions were prepared in the usual manner, except that each incubation dish contained 1.07 x 105 d.p.m. of [4-14C]corticosterone. Cells without and with ACTH ( 100 mU/ml) were incubated for 3, 7, and 21 days. As controls, aliquots of medium containing labelled steroid but not cells were incubated for the same time periods. Tritiated corticosterone was added to each sanzgle before extraction. The extract was carried through the double isotope assay procedure (Levine et nl. 19721, with the substitution of unlabelled acetic anhydride for the usual tritiated reagent. It was then possible to calculate what fraction of the 14C-labelled material recovered from the dish was corticosterone. The results are shown in Table 1. It is apparent that in the cell suspension, but not in the medium alone, corticosterone was metabolized in some manner. For this reason we refer only to accumulation of corticosterone in our studies. It seems likely that the smaller fractional loss of I4C as corticosterone in the incubations with ACTH reflects the larger amount of corticosterone in these incubations, although one cannot exclude the possibility that ACTH had a direct effect on the process
Can. J. Physiol. Pharmacol. Downloaded from www.nrcresearchpress.com by UNIV CHICAGO on 11/15/14 For personal use only.
CAN. J. PHYSIOL. PHARMACOL. VOL. 53, 1975
cubation before ACTH treatment. Treatment with ACTH restored, in part at least, the steroidogenic capacity of the cells. In this same experiment cells were incubated for 7 and I6 days without ACTH, then harvested and re-incubated in fresh medium containing 100 mU of ACTW per millilitre. The results of cell counts and corticosterone measurements are shown in Fig. 5. Considering the average recovery of 75-9096 of cells through the harvesting procedure, the corticosterone response of these cells resembles closely that of unharvested cells treated with ACTH at the same time (see Fig. 4). It seems likely that the change of medium had little effect on the response of the cells to ACTH. Figure 5 also illustrates the relative stability of cell numbers at these tiines of incubation. 0
1
2
3
4
5
5
7
T l M t OF INCUBATION IN DAVS AFTER ACClTlON O t ACTH
FIG.4. Corticosterone response to ACTH added after various times of incubation. The original suspension contained 1.35 X 10" cells/ml. Each point represents the corticosterone content of a single incubation; duplicate incubations were assayed at each time point.
Aldosterone Accumulation in Cell Suspensiorzs A cell suspension was prepared and aliquots were incubated in the usual manner ( K + concentration 5.0 rnequiv./litre) for periods up to
of removal. The [4-1'1Cjcorticosteronewas not, of course, a suitable tracer for all of the corticosterone produced during incubation, so no statements about rates of disappearance of corticosterone can be made. We have not done studies of the metabolism of aldosterone in this system.
Respmzse to ACTH Addld after Varying Periods sf Incubation Cell suspensions prepared in the usual manner were incubated for 0, 7, 16: and 21 days before the addition of 100 m u of ACTH. The results are shown in Fig. 4. The response of the cells treated with ACTH at 21 days indicates the presence of viable cells through the fourth week of incubation. It can be seen that, with increasing time of incubation before addition of ACTH, the initial response, reflected in the measurement of corticosterone one day after addition s f ACTH, progressively diminished. It may also be seen that with eontinued incubation the cells treated at 7, 16, and 21 days increased their rate of steroid accumulation. The period before this increase became apparent lengthened with time of in-
U
z
INCUBATION TIME IN DAYS AFTER HARVESTING
FIG. 5 . Cell counts and costicostcrone response to A C T H after harvesting. The upper and lower panels show, respectively, the cell counts and corticosterone responses to ACTH (100 rnUirnl) of suspensions which were harvested and reincubated. The solid and broken lines refer to cells harvested at 7 and I6 days. Each point represents the mean of rneasurements on two dishes.
537
PRICE ET AL,.: LONG 'TERM INCUBATION OF ADRENAI, CELI,S
TABLE2. Aldosterone accumulation in response to changes sf potassiunl concentration in the incubation medium. in the presence and absence of ACTH
Can. J. Physiol. Pharmacol. Downloaded from www.nrcresearchpress.com by UNIV CHICAGO on 11/15/14 For personal use only.
Aldosterone accumulation (ngiinc~batisn)~ time (h)
K' (mequiv. /li trey 5 7.5 10
ACTH (tnU/ml> 0 100 0 100
0 10Q
---
0
24
48
79
111
144
0.5 0.4 0.4 0.7 0.4 0.7
2.6 13.8 4.1 22.7 3.1 12.6
2.5 15.8 5.4 38.8 14.5 43.9
2.3 15.5 7.3 56.7 35.2 101
2.5 15.8 7.2 56.6 46.7 135
2.2 15.0 6.5 68.0 50.8 145
aAdded as KCI. bEa& number is the mean of measurcmcnts o n duplicate incubations. The number of cells iricubated were 1.9 for the cexperiments with K + concentrations of 5.0, 7.5, and 10 rnequiv./litre, respectively
10 days in the presence and absence of ACTH. The latter was added in a concentration of 1QO mU/ml, a concentration which caused an easily detectable increase in aldosterone accumulation. The results of aldosterone measurements on these incubations are shown in Fig. 2. At 2 3 h, aldosterone levels in ACTHcontaining dishes were approximately 16-fold higher than at zero time, and 5-fold higher than in the corresponding unstimulated incubations. There was little or no accumulation of aldosterone in ACTH-stimulat~d or control incubations during the subsequent 9 days of incubation. The corticosterone measurements shown in Fig. 2 are from the same experiment. In a similar experiment with incubations containing 1.0 x 10Qells, the means of replicate measurements ( 1 1 = 4 ) of the aldosterone response to ACTH (100 mU) at 2, 8, 16, 24, and 48 h were 0.52, 1.7, 2.5, 3.5, and 4.4 ng, respectively. In a single experiment, adrenal glands were decapsulated, and separate suspensions prepared from the outer and inner portions. Aliquots of the two suspensions. containing 1.46 x 10,' and 2.07 x 10') cells, respectively, were incubated without and with ACTH (100 mU) for periods up to 24 h. Aldosterone was undetectable or very low (< I ng) in incubations of the inner portions of the glands. In the incubations derived from the outer portion of the glands, the means of duplicate measurements of aldosterone response to ACTH were 2.4, 23, and 31.5 ng at 2, 16, and 24 h, respectively. Table 2 shows the effect on aldosterone accumulation of varying the potassium con-
/
165 1.8
105, and 1.7 x 105
centration in the medium, in the presence and absence of ACTH. Cells obtained from a single preparation were incubated in medium containing K + concentratiolls of 5.0, 7.5. or 10 mequiv./litre for periods up to 134 he As in thc experiment illustrated in Fig. 2, incubations with K i , 5.0 mequiv./litre, showed little accumulation of aldosterone beyond 24 h, cither in the presence or absence of ACTH. In incubations with K ', 10 mequiv./litre, however, aldosterone was found in much greater amounts and continued to increasc through the sixth day, with or without ACTH. Similar, though less marked changes occurred in incubations with 7.5 mequiv. of 161 per litrc.
Discussion The system described permits long term survival of adrenal cells of the normal adult rat. The cclls were non-confluent and no evidence of cell replication was found. Cell numbers fell sharply in the first 24 to 48 h of incubation, but thereafter declined very slowly. Of the two cell types identified only the granular cells could be found after the seventh day of incubation. We cannot say whether the disappearance of pale cells was due to their death or to transformation into granular fonns, nor have we attempted to determine the function of these cells. The corticosterone response to adrenocorticotropic hormone (ACTH) added late in incubation demonstrated the presence of functioning cells through the fourth week of incubation. A corticosterone response was seen to as little as 5.0 pU (50 pg) of ACTH1-24 per millilitre, a concentration which in terms of international units is at the lower end of the
Can. J. Physiol. Pharmacol. Downloaded from www.nrcresearchpress.com by UNIV CHICAGO on 11/15/14 For personal use only.
538
CAN. J . PHYSIOL,. PHARMACOL. VOL. 53, 1975
rangc of concentration of ACTH,-39 found in normal rat serum (kfatsuyama et a[. 1970). Sayers r l al. ( 1 97 1) found that isolated adult rat adrenal cells incubated in a simple buffer solution for 2 h responded to as little as 0.5 pU of ACTH1-3s, per millilitre. Corticosterone accumulation in response to a single addition of 100 mU of ACTH per millilitre at the beginning of incubation was virtually constant for the first 3 days, and declined slowly over the next 5-7 days. Whether the decline in response was due to a change in medium or cells, or due to destruction of ACTH (Bullock et al. 1972), remains to be determined. The data presented in Fig. 4 illustrate a decline in corticosterone response to ACTH with increasing time of incubation of unstirnulated cells. Comparison of the responses to ACTH of unharvested and harvested cells (Figs. 4 and 5) indicate that prolonged incubation without change of medium is not responsible for the decline in responsiveness. Figures 4 and 5 also illustrate the partial restoration by ACTH of steroidogenic activity of the cells. The responses described are reminiscent of those of patients with hypopituitarism and secondary adrenal insufficiency, and the system should prove to be a useful model of such a state. The rat adrenal cell system of O'Hare and Neville ( 3973u, 1973b) did not include cells of the aona glomerulosa. Their system, which employed other enzymes for cell dispersion, a different incubation medium, confluent cultures, and frequent feeding with medium containing ACTH, displayed virtually constant corticosterone production for up to 40 days. Cells incubated without ACTH for up to 8 weeks showed a lag in corticosterone production in response to ACTH, but within 4 or 5 days produced corticosterone at a rate comparable to that of cells maintained with ACTH throughout incubation. Responses to ACTH at concentrations below 1 mU/ml were not detected. B'Hare and Neville give little information about cell numbers during the course of incubation. The data presented in Fig. 2 show that the aldosterone response to ACTH, in medium containing 5.0 mequiv. of K-+ per litre, terminated after 24-48 h, at a time when
the corticosterone response was still near maximum. A similar phenomenon has been described in the human subject given ACTH continuously, that is, ddosterone production increases for a day or so, then declines, at a time when cortisol production is still increasing (Tucci et al. 1967). The findings displayed in Fig. 2 raise the possibility that this in vivn phenomenon is due, at least in part, to an intra-adrenal regulatory mechanism. However, the data of Fig. 2 could equally well be due to the death in vidro of the aldosteroneproducing cells. The latter possibility was eliminated by the finding that addition of increased amounts of Kt- to the medium resulted in an accumulation of aldosterone, which continued through the sixth day of incubation. The effect of increasing the concentration of potassium indicates that aldosterone-secreting cells can survive in this system for several days and that the system should be useful for studies of the regulation of aldosterone production. A significant advantage of the system described, like that of O'Hare and Neville, is that the role of factors involved in the maintenance of adrenal function can be examined in vitrcr without the necessity of treatment of the tissue donor prior to preparation of cell suspensions. The latter consideration is of particular importance in the study of human adrenal cells.
Addendum Following submission of this paper, Hornsby et al. (1974) reported a study of aldosterone production by isolated rat aona glomerulosa cells, incubated for periods up to 20 days in culture medium containing 6 mhf Kt- and 11 mM K +. In their paper Hornsby et al. referred to an earlier communication (Hornsby et al. 1973) which the present authors had not seen. In these two studies, Hornsby et al. showed that aldosterone production was considerably greater in the medium with 11 mM K + . In the latter medium in the absence of ACTH, daily production reached a peak at day five, then declined to very low levels over the next 7-10 days. Addition of ACTH to cultures in medium containing 11 mM K + increased the maximum rate of aldosterone production but shortened the period during which aldosterone was produced.
I'KICE ET AL.: I,ONG TERM INCLJBATION OF ADRENAL CEI,I,S
Can. J. Physiol. Pharmacol. Downloaded from www.nrcresearchpress.com by UNIV CHICAGO on 11/15/14 For personal use only.
We are grateful to Miss H. Baechtold and Miss P. McFadzean for their technical assistance, and t o Drs. 6. Williams and R. H. Underwood for a gift of antiserum for the aldosterone assay. U., and NUSSDORFER, G. G . 1972. Tissue culture ARMATO, of rat adult decapsulated adrenal glands. A methodological, ultrastructural and morphometric investigation. Z.Zellforsch. Mikrask. Anat. 135,24-273. Bur-LOCK,G . R., GII.I,AM, E. A., LOWRY, P. J., MC:MAK.I.IN, C., and PETERS,J . 1972. Observations on the fate of adrenocorticotrophin in isolated rat adrenal cell suspensions and the separation of an enriched adrenocorticotrophin-sensitivesteroidogenic cell fraction. Biochem. J. 129,23 p.(Abstr.). HALKEHSTON, I . 11. K., and FEINSTEIN, M. 1%8. Preparation of ACTH responsive isolated cells from rat adr e n a l ~ Fed. . Proc. 27.626 (Abstr.). HORNSBY, P. J., O'WARE,M . J., and NEVILLE, A. M. 1973. Effect of ACTH on biosynthesis of aldosterone and corticosterone by monolayer culttires of rat adrenal zona glornerulosa cells. Riochern. Biophys. Kes. Coinmun. 54, 1554-1559. 1974. Functional and morphological observations on rat adrenal zona glolnerulosa cells in monolayer culture. Endoc~inology,95, 1240-1251. KAHKI,A. 1. 1973. Inhibition of ACTH-induced differentiation of cortical cells and their mitochondria by corticosterone in tissue culture of fetal rat adrenals. Anat. Rec. 176,253-272. KI.OPPENBORG. P. W. C., ISLAND. D. P.. LIDDLE,G . W., MICHEI-AKIS, A. M., and NICHOLSON, W. E. 1968. Method of preparing adrenal cell s~lspensionsand its applicability to the in vitro study of adrenal metabolism. Endocrinology, 82, 1053-1058. LEVINE,J . PI., LAIDI.AW,J . C.. and RUSE,J . L. 1972. EtTects of a sulfated mycopolysaccharide (KOI-8307) on zona glonlerulosa width and steroid production by the rat adrenal cortex. Can. J . Physiol. Pharmacol. 50, 270-275. MKTSIJYAMA, H., RUHMANN-WENNIIOLL>, A., and NELSON, D. H. 1970. Biologic and immunologic simi-
539
larities between rat and human adrenocorticotropin (ACTH). Endocrinology, 87,756-763. O'HARE,M. J . , and NFVILLE, A. M. 1 9 7 3 ~Morphological . responses to corticotrophin and cyclic AMP by ada~lt rat adrenocortical cells in monolayer culture. J . Endocrinol. 56,529-536. 19736. The steroidogenic response of adult rat adrenocortical cells in mono1:lyer culture. J. Endocrinol. 56,537-549. PEYTREMANN, A., NIC:HOLSON, W. E., BROWN,R. D., LIDDI.~;., G. W., and HARDMAN, J. C. 1973. Comparative effects of angiotensin and ACTH on cyclic AMP and steroidogenesis in isolated bovine adrenal cells. J. Clin. Invest. 5 2 , 8 3 5 4 4 2 . Roos, B. A. 1974. Effect of ACTH and c 4 M P on human adrenocortical growth and function in vitro. Endocrinology, 94,685-690. RUSE, J. L., PRICE:,C., STIF,FF.L, M., and L A I L X A W J.,C. 1972. The influence of glucocorticoids and heparin on aldosterono production. I n Hypertension '72. Edircd by J. Genest, and E. Koiw. Springer-Verlag, Berlin. pp. 326-333. SAYERS,G., SWALI.OW,R. I,., and GIORDANO, N . D. 1971. An improved technic for the preparation of isolated rat adrenal cells: a sensitive, accurate and specific method for the assay of ACTH. Endocrinology, 88, 1063-1068. S I L B E KR. , H., BUSCH,R. D., and OSLAPAS,K. 1958. Practical procedure for estimation of corticosterone or hydrc>cortisone.Clin. Chem. 4,278-285. S w ~ r ..ow. 1 K. I.,. , and SAYERS, 6 . 1969. A technic for the preparation of isolated rat adrenal cells. Proc. Soc. Exp. Biol. Med. 131, 14. 'Tucc~,J. R., ESPINEK, A . , JAGGER, P. I . , P A U K , GL., . and I,AUI.EK,13. P. 1967. ACTH stimulation of aldosterone secretion in normal subjects and in patients with chronic adrenocortical insufficiency. J . Clin. Endocrinol. Metab. 27,568-575, UNDF.RWOOD, R. H., and WILLIAMS, G . H. 1972. The simultaneous measurement of aldoster-one, cortisol, and co~ticosteronein human peripher*al plasma by displacementanalysis. J. Lab. Clin. Med. 79,848-862.
An Adrenal Cell System Suitable for Long Term Study of Factors Affecting Steroidogenesis CARMAN S. PRICE,^ JAMESL. RUSE, AND JOHN C. LAID LAW^ Department of Medicine and In.rtit~rteo f Medical Science, University of Toronto, Toronto, Cbntario M5G 1.48 Received August 2 1, 1974
PRICE,C. S., RUSE,J. I,., and I,AIDI~AW, J . C. 1975. A11 adrenal cell system suitable for long term study of factors affecting steroidogenesis. Can. J. Yhysiol. Pharmacol. 53,531-539. Cell suspensions of normal adult rat adrenals have been prepared by trypsinization and incubated in Ham's nutrient mixture F10 containing horse serum, fetal calf serum, and lima bean trypsin inhibitor. In most experiments culture medium was not changed during incubation. In this system the number of cells fell to 5W% after 2 days, then slowly declined to 2% after I month of incubation. A corticosterone (B) response was seen to as little as 5 pU of ACTH per millilitre, a concentration which is within the range found in normal rat serum. With maximal stimulation (100 mU ACTHIml) the rate of accumulation was highest during the first 24 h then slowly decreased over the fc>llowing9 days. When in separate experiments AC'TH was added after various times of incubation up to 3 weeks, there was a B response which continued for as long as 1 week after the ACTH was added; the later the time at which ACTH was added the lesser was the initial B response and the longer the lag period before a substantial response occurred. In cell suspensions in medium containing 5.0 mequiv. of KS per litre, aldosterone content increased for approximately 24 h, then showed little or no change over the next 9 days. With increased Kt concentration, aldosterone was found in greater amounts and accumulation continued for longer periods, both without and with ACTH. This adrenal cell system appears suitable for long term study of factors affecting steroidogenesis. PRICE, C. S., RUSE.J. I,. et L,AIDL.~\w, J . C. 1975. An adrenal cell system suitable for long term study of factors affecting steroidogenesis. Can. J. Physiol. Pharmacol. 53,53 1-539. Des suspensions cellulaires de surrenales de rat adulte sont preparees par trypsinisation, et incubees dans un milieu nutritif F,, de Ham contenant du serum de cheval, du siri~rnfoetal de veau et l'inhibiteur de la trypsine de la feve "lima". Dans la plupart des experiences, le milieu de culture n'est pas changi au cours de I'incubation. Dans ce systPme, le nomhre des cellules tombe a 50% apres 2 jours. puis diminue lentement jusqu'a 20% apres un mois d'incubation. Une secretion de corticosterone est constatee en reponse a la concentration rninime de 5 pUlml (reponse B), qui est dans les limites des taux seriques normaux chez le rat. Apres une stimulation maximale (100 m u d'ACTH/ml), le taux d'accumulation est le plus eleve durant les premieres 24 h, puis decline dans la periode des 9 jours suivants. Dans des experiences separees, on ajoute I'ACTH apres des temps d'incubation allant jusqu'i trois semaines. On constate alors une reponse B prolongie jusqu'8 une semaine aprks I'addition d' ACTH. Plus I'ACTH est ajoute tard, moins la reponse B initiale est elevie, et plus la periode de latence de cette reponse est longue. Dans les suspensions cellulaires en milieux contenant 5.0 inequiv. de potassium par litre, le contenu en aldosterone est augmente pendant environ 24 h, mais au cours des 9 jours suivants, il y a peu de changements. La concentration d'aldostirone dans le milieu augmente avec la concense fait sur de plus longues periodes, avec ou sans ACTH. tration de potassium, et I'acct~n~ulation Ce systeme de suspensions cellulaires snrrenaliennes semble utilisable pour I'etude a long terme des facteurs affectant la steroidogCntse. [Traduit par le journal]
Preparations of isolated normal adult adrenal cells (Halkerston and Feinstein 1968; Kloppenborg ct al. 1968; Swallow and Sayers 1969; 'Supported by a grant from the Medical Research Council of Canada. "resented in part at the 55th meeting of the Endocrine Society, June 1973. "older of a Fellowship from the Medical Research Council of Canada. 'Please send reprint requests to: Dr. 1. C . Laidlaw, Medical Sciences Building, University of Toronto. Toronto, Ontario. ,%
Saycrs et al. 1971; Peytremann ct al. 1973) have proved to be of great value in the study of the function of the adrenocortical cell. Systems such as these were designed for experirnents requiring only short term incubations of the cells and are not suitable for the in vitro study of events which in vivo require many hours or even days for their expression. Two such events are the suppression of ddosterone production (Ruse et ". 1972) and the fall of aldosterone secretion, after an initial rise, in the face of continuing adrenocortico-
Can. J. Physiol. Pharmacol. Downloaded from www.nrcresearchpress.com by UNIV CHICAGO on 11/15/14 For personal use only.
532
CAN. J. PHYSIOL. PHARMACOL. VOL. 53, 1975
tropic hor~none(ACTH) infusion (Tucci et al. 1967). To study the adrenal component of these phenomena, we have developed a system which, like that of Armato and Nussdorfer (1972), Kahri ( 1973), O'Hare and Neville (1973a, 1973h), and Roos (1974), permits survival in vitro of functioning adrenocortical cells for several weeks. Isolated cells were prepared from whole normal adrenal glands of adult rat by a modification of the method described by Swallow and Sayers (1969) and incubated in Ham's nutrient mixture Ff 0, supplemented with horse and fetal calf serum, for periods up to 30 days. In this paper we present our observations on cell survival, corticosterone and aldosterone accumulation. and the effects of ACTH and K + concentration on the accumulation of these steroids.
Materials and Methods 1-yophilized trypsin (TRL) and lima-bean trypsin inhibitor (LRI) were obtained from Worthington Biochemical Corporation, Freehold, N.J., and heatinactivated horse and fetal calf serum from Grand Island Biological Company, Grand Island, N.Y. Sterile disposable plastic tubes and petri dishes (Optilux) were obtained from Falcon Plastics, Oxnard, Calif. Sterile disposable pipettes and membrane filter units (0.45 p n ~ ) were purchased from Fisher Scientific. Toronto, Ont. Corticosterone and aldosterone were purchased from Tkapharm, Ramat-Gan, Israel and used as received. 14-"C]Corticosterone (S.A. 57.3 mCi/mmol) , [I ,2-"Hlcorticosterone (S.A. 50 Ci/mmol) , [4-'CCldosterone (S.A. 55 mCi/mmol ), and [I ,2-Wlaldosterone (S.A. 54 Ci/mmol) were supplied by New England Nuclear Canada, Dorval. Que. Purity checks and storage conditions were as described by Levine et ul. (1972). Male Wistsr rats weighing from 250 to 300 g were obtained from High Oak Ranch, Aurora, Ont. They were maintained on Purina rat chow and water ail lihitltnl for approximately 1 week before sacrifice. Solvents and chemicals were certified or reagent grade. All solvents except n-butyl acetate, petroleum ether, and propylene glycol were distilled prior to use. Bovine albumin (fraction V ) was purchased from .4rmour Pharmaceutical Co., Chicago. Ill. Synthetic ACTHI-., (Cortrosyn, 0.1 U.S.P. unit per microgram, Organon Tnc., Montreal, Que.) was used throughout. Radioactivity was measured as previously described (Levine et al. 1972). Aqueous samples were counted in 15 ml of a scintillation mixture consisting of 40 ml of ethanol, 80 ml of Liquifluor (New England Nuclear), 1880 ml toluene and 500 ml of Biosolv (BBS11 1, Beckman Instrument Co., Fullerton, Calif.). Incuburion Mediirrn Adrenal cells were incubated in Ham's nutrient
mixture F1 0 (Grand Island Biological Co. ) supplemented with 15% horse serum and 2.5% fetal calf serum. F10 was purchased in the powder form, without sodium and potassium chloride. Except where noted, sodium, potassium, and calcium concentrations were adjusted to 140 mequiv. /litre, 5.0 mequiv. /litre, and 2.5 mM, respectively. These values approximate the concentrations measured in the serum of the rats used in our studies. The final chloride concentration was 115 mequiv. /litre, about 12 mequiv. higher than the mean concentration in the serum of the rats. Immediately prior to use, lima-bean trypsin inhibitor (1 g/litre) was added and the solution sterilized by filtration through a 0.45 pm membrane filter. In the presence of 5% COz in air, the pH of the medium was 7.4.
Preyuration and Incubaticrrz o f CclI Suspension F Adrenal cell suspensions were prepared by a minor modification of the method described by Swallow and Sayers ( 1969). Rats were killed by cervical dislocation. The adrenals were removed and cleaned as quickly as possible, then minced and placed in 20 ml of 0.25% trypsin solution in Krebs-Ringer bicarbonate with glucose. The solution was agitated with a magnetic stirrer and stirring bar. After 20 min the supernatant was removed. The incubation procedure was performed a total of six times. The pooled supernatants were centrifuged for 30 min at 100 X g and the resultant pellets resuspended in the incubation medium. Approximately 5 X 1 0 cells were obtained per adrenal gland. Cells were plated in 35 X 10 mm petri dishes in a total volume of 1.0 ml: 0.9 ml of the cell suspension and either 0.1 ml of medium or 0.1 ml of medium eontaining ACTH. Two of these small dishes, together with a third dish containing water, were placed in a larger covered petri dish and incubated at 37 " C in a water-saturated atmosphere consisting of 5 % CO:! in air. Evaporative losses and pH changes were negligible throughout the incubations. No antibiotics were added to the incubations; about 10% of incubations showed visible signs of infection and were discarded. Except where noted, medium was not changed during the course of incubation. The entire contents of a given dish were used either for cell counting or for steroid measurements. Cell Counting Aliquots of the original unincubated cell suspensions were counted directly using a Fuchs-Rosenthal chamber. Cells which had been incubated were counted after treatment with trypsin (final concentration 0.7 % ). These cell counts, therefore, included both cells that were free in the medium and cells that had been released from the bottom of the dish by the trypsin treatment. The time of the exposure to trypsin required to release adherent cells varied from a few minutes to 4 h. Further incubations with trypsin for as long as 3 h did not affect cell counts. Only cells excluding Trypan Blue were included in the count. At each time point in the experiments reported in this paper, cell counts on duplicate incubations were obtained. The coefficient of variation of these duplicate counts was 7.1 %, (n = 39).
Can. J. Physiol. Pharmacol. Downloaded from www.nrcresearchpress.com by UNIV CHICAGO on 11/15/14 For personal use only.
PRICE E'T AL.: L,ONCi TERM INCUBATICIN O F AIIKENAL, CELLS
Harvesting oj Cells Cells were harvested after exposure to trypsin (final concentration (3.7%) for 1-3 h. The pooled cells were centrifuged at 4 " C and 100 X g 30 min, resuspended in fresh medium and re-incubated in the same way as the original suspension. Recovery of cells varied from 75-9096. Meusurernetzt of Steroids The contents of incubation vessels were transferred to 60 ml extraction tubes, together with 1 ml of medium and 3 ml of water (for rinsing the vessels), and extracted with 15 ml of dichloromethane. The latter was washed once with 1 ml of 0.1 N NaBH. Extraction of a mixture of 2 ml of medium and 3 ml of water provided material for blank measurements. Corticosterone was measured by the method of Silber et nl. (1958). Standard solutions, prepared by dissolving known amounts of corticosterone in medium, were extrhcted in the same manner as the samples and blanks. Fluorescence was measured in an Aminco-Bowman spectrophotofluorometer et excitation and emission wave lengths of 464 and 520 pm, respectively. All measurements were done on duplicate incubations. The coefficient of variation of these duplicate measurements was 1196. Inter-assay variation, based on 11 assays of a single pool of cells was 5.5%. Two pools of cells were incubated for 3 weeks, one with and one without ACTW, and assayed by the fluorometric method and a double isotope method (Levine et ul. 1972). The latter method involves addition of [14-'TCcorticosterone, extraction with methylene chloride, acetylation with tritiated acetic anhydride of known specific activity, oxidation with chromium trioxide, and purification to constant specific activity through a series of thin-layer chromatograms. There was no significant difference in the results of the two methods. Aldosterone was measured by radioimmunoassay as described by Underwood and Williams (1972), with minor modifications. A small quantity of tritiated aldosterone was added to the sample. Steroids were extracted from the medium with methylene chloride and chromatographed on paper in the Bush 5 system. Aldosterone was located on the chromatogram with a chromatogram scanner (Packard Instrument Company model 20Q1 ) . The aldosterone standards and the eluted samples and blanks were dissolved in the assay medium prior to transfer to the reaction vessels. The final reaction mixtures consisted of sample or standard in 0.5 ml of bovine albumin solution and 0.5 mi of 1 : 500 000 dilution of antiserum in borate buffer containing 3.6 X IQ' d.p.m. of [1,2-W]aldosterone. Standard curves were constructed by plotting the reciprocal of the fractions of radioactivity bound against the \tandard weights. Blank "weights" were subtracted from sample weights before the latter were corrected for recovery. In 17 assays, the mean and standard deviation of the blank were 8.3 zk 6.2 pg. In the sample measurements reported, except for those at zero incubation time, the fraction of the sample used in the binding reaction contained at least 75 gg. The coefficient of
533
variation of replicate aliquots ( a = 4 ) from three cell suspensions measured in the same assay varied from 8.0% at the 3 ng level to 3.2% at 11.5 ng. The aldosterone content of a pool of cells incubated for 72 h in the presence of ACTH was assayed by the immunoassay and a double isotope assay (Levine ct (11. 1972). The results obtained were 18.2 -t- 1.2 ng/ml (mean -+ S.D., tr = 14) and 17.6 -C 0.6 ng/ml ( 1 2 = 3 ) , respectively.
Results Microscopic Appearance of Cells Cells have been examined only in unstained preparations, using the light microscope; two types of cells have been distinguished. The cells of one type contained a variable number of granules in the cytoplasm ("granule cells") and were of relatively uniform size (1 0-1 5 pm). Cells of the second type lacked granules ("pale cells") and varied in size from 3 to 15 pm. In the first 48 h of incubations, the cells had a somewhat indistinct and ragged outline, but subsequently became more uniformly rounded and developed sharp, clear outlines. The latter appearance was maintained throughout the remainder of incubation. The cells became adherent to the surface of the dish rather slowly; virtually 100% of the cells were adherent by the fourth or fifth day. With continued incubation, the cells became less adherent, so that by the third week a considerable number were free in the medium and many of the re~llaindercould be dislodged by shaking the vessel. Cells which had been harvested and re-incubated in fresh medium were firmly adherent within 24 h, but subsequently displayed the same pattern of decreasing adherence. In no case was evidence of cell replication observed and, for the most part, the cells were not in contact with one another. Addition of adrenocorticotropic hormone (ACTH) to incubations had no effect on the appearance of the cells. However, in incubations with ACTH the cells appeared to be less adherent to the dish and formed occasional small clumps in which cells and cell debris were not always distinguishable. The cells in these clumps were easily separated by gentle pipetting. Cell Counts Over I Month of Incubation Figure 1 shows the cell counts obtained in a representative experiment during 30 days of incubation without ACTH. Over the first 2
Can. J. Physiol. Pharmacol. Downloaded from www.nrcresearchpress.com by UNIV CHICAGO on 11/15/14 For personal use only.
534
CAN. J. PHYSIOL. PHAWACOL. VOL. 53, 1975
FIG.2. Corticosterone and aldosterone accumulation in incubations (1.65 X 10" ceIls/ml) without and with ACTH (100 mTJ/ml). Each point represents the steroid content of a single dish. Measurements were made on two dishes at each time point; the curves join the means of the two measurements. Csrticosterone, .--a : aldssterone. l - - - - - -@ . INCJSATW IN DAYS
FIG. 1 . Cell counts over 1 month of incubation. Each point represents the mean of four setxirate counts on each of two incubations. Total cells, 0-0; granule cells, A - - - - - A: pale cells, 0- - -0. I
days of incubation, the total cell count fell precipitously and by the second day about 50% of the cells initially incubated remained. After 2 days the fall in total cell count was more gradual and at 30 days of incubation, approximately 20% of the cells remained. In experiments in which ACTH (100 mU/ml) was added at zero time, a similar pattern was seen, but after day 2, cell counts were 35-45% lower than in non-ACTH controls. The initial precipitous fall in total cell count was due, for the most part, to a fall in the number of pale cells. By 8 days of incubation only granule cells remained. Corticc~steroneRespotzse to ACTH Added at Zero Time Figure 2 shows corticosterone accumulation in cell suspensions during 10 days of incubation, in the absence and presence of ACTH (200 mU/ml). The results shown, which are from a single experiment, resemble those obtained in three similar studies. In the unstimulated suspensions, after an initial lag phase, corticosterone increased until day five, then ceased to accumulate. In the presence of
ACTH, corticosterone accumulation was much increased. The rate of accumulation was virtually constant for the first 3 days then slowly decreased over the remainder of the incubation. The period of constant corticosterone accumulation coincided with the rapid fall in cell number (see Fig. 1 ) . In similar studies the corticosterone response (the difference between ACTH-treated and control incubations) to 10, 100, 500, and 1000 m u of ACTH per millilitre was determined daily for 14 days. The maximum corticosterone response over this time period was reached at 100 m u of ACTH per millilitre, and this concentration was used in all other experiments except those involving studies of sensitivity to ACTH. Cell suspensions prepared in the usual way did not respond to less than 25 pU of ACTH per millilitre. However, cells washed once with trypsin-free buffer or medium before incubation were more sensitive. In each of six experiments, in which cell numbers varied from 6.5 x l o 4 to 4.3 x 10" the corticosterone content of incubations containing 5pU of ACTH per millilitre was greater than that of incubations without ACTH. The mean difference (-t 1 S.E.M.) was 97 t 40 ng of corticesterone, which is significant at the 5% level. The high standard error reflects variation in activity of the individual cell preparations.
PRICE ET AL.: LONG TERM INCUBATION OF ADRENAL CELLS
TABLE 1. Metabolism of [4-14C]corticosterone by cell suspensions
Can. J. Physiol. Pharmacol. Downloaded from www.nrcresearchpress.com by UNIV CHICAGO on 11/15/14 For personal use only.
Incubation time 72 h
Recovered 14C as corticosterone
(x)"
Sample Medium Medium f cells Medium cells + ACTH Medium Medium + cells Medium + cells ACTM Medium Medium + cells Medium + cells ACTH
+
1 week
+
3 weeks
+
"14-L4C]Corticosterone. 1.07 s 105d.p.m., was added to each dish at the beginning of incubation. Each number represents the mean of two determinations.
The test of significance was based on tho signs of the differences.
Correlatiorz Between Corticosterone Accurnnlation and Number of Cells Incubated A cell suspension prepared in the usual manner was diluted serially to provide three suspensions, which, when counted, were found to Sontain 165 x io3, 88 x lo3, and 18 x lo3 cells per millilitre. Duplicate 1.0 ml aliquots of each suspension were incubated without and with ACTH ( 108 mU/ml) for 24, 48, 72, 96, 168, and 240 h, and the corticosterone response to ACTH was determined. The results are shown in Fig. 3. It can be seen that, within the range of 18 x i 0 3 - 165 x lo3 cells per
CELL COUNT
x
~O'!/INCUBATION
FIG.3. Relationship between the number of cells incubated and corticosterone response to ACTH (100 mUirnl added at zero time). Each point represents the mean of corticosterone measurements on two dishes.
incubation, the relationship between corticosterone response to ACTH and cell numbers at each tirne'period is reasonably well described Except where noted, in by a straight studies of corticosterone the concentration of cells incubated fell within the range described in this experiment.
Metabolism of Corticosterone The possibility that corticosterone was being transformed to other compounds during incubation was investigated. Cell suspensions were prepared in the usual manner, except that each incubation dish contained 1.07 x 105 d.p.m. of [4-14C]corticosterone. Cells without and with ACTH ( 100 mU/ml) were incubated for 3, 7, and 21 days. As controls, aliquots of medium containing labelled steroid but not cells were incubated for the same time periods. Tritiated corticosterone was added to each sanzgle before extraction. The extract was carried through the double isotope assay procedure (Levine et nl. 19721, with the substitution of unlabelled acetic anhydride for the usual tritiated reagent. It was then possible to calculate what fraction of the 14C-labelled material recovered from the dish was corticosterone. The results are shown in Table 1. It is apparent that in the cell suspension, but not in the medium alone, corticosterone was metabolized in some manner. For this reason we refer only to accumulation of corticosterone in our studies. It seems likely that the smaller fractional loss of I4C as corticosterone in the incubations with ACTH reflects the larger amount of corticosterone in these incubations, although one cannot exclude the possibility that ACTH had a direct effect on the process
Can. J. Physiol. Pharmacol. Downloaded from www.nrcresearchpress.com by UNIV CHICAGO on 11/15/14 For personal use only.
CAN. J. PHYSIOL. PHARMACOL. VOL. 53, 1975
cubation before ACTH treatment. Treatment with ACTH restored, in part at least, the steroidogenic capacity of the cells. In this same experiment cells were incubated for 7 and I6 days without ACTH, then harvested and re-incubated in fresh medium containing 100 mU of ACTW per millilitre. The results of cell counts and corticosterone measurements are shown in Fig. 5. Considering the average recovery of 75-9096 of cells through the harvesting procedure, the corticosterone response of these cells resembles closely that of unharvested cells treated with ACTH at the same time (see Fig. 4). It seems likely that the change of medium had little effect on the response of the cells to ACTH. Figure 5 also illustrates the relative stability of cell numbers at these tiines of incubation. 0
1
2
3
4
5
5
7
T l M t OF INCUBATION IN DAVS AFTER ACClTlON O t ACTH
FIG.4. Corticosterone response to ACTH added after various times of incubation. The original suspension contained 1.35 X 10" cells/ml. Each point represents the corticosterone content of a single incubation; duplicate incubations were assayed at each time point.
Aldosterone Accumulation in Cell Suspensiorzs A cell suspension was prepared and aliquots were incubated in the usual manner ( K + concentration 5.0 rnequiv./litre) for periods up to
of removal. The [4-1'1Cjcorticosteronewas not, of course, a suitable tracer for all of the corticosterone produced during incubation, so no statements about rates of disappearance of corticosterone can be made. We have not done studies of the metabolism of aldosterone in this system.
Respmzse to ACTH Addld after Varying Periods sf Incubation Cell suspensions prepared in the usual manner were incubated for 0, 7, 16: and 21 days before the addition of 100 m u of ACTH. The results are shown in Fig. 4. The response of the cells treated with ACTH at 21 days indicates the presence of viable cells through the fourth week of incubation. It can be seen that, with increasing time of incubation before addition of ACTH, the initial response, reflected in the measurement of corticosterone one day after addition s f ACTH, progressively diminished. It may also be seen that with eontinued incubation the cells treated at 7, 16, and 21 days increased their rate of steroid accumulation. The period before this increase became apparent lengthened with time of in-
U
z
INCUBATION TIME IN DAYS AFTER HARVESTING
FIG. 5 . Cell counts and costicostcrone response to A C T H after harvesting. The upper and lower panels show, respectively, the cell counts and corticosterone responses to ACTH (100 rnUirnl) of suspensions which were harvested and reincubated. The solid and broken lines refer to cells harvested at 7 and I6 days. Each point represents the mean of rneasurements on two dishes.
537
PRICE ET AL,.: LONG 'TERM INCUBATION OF ADRENAI, CELI,S
TABLE2. Aldosterone accumulation in response to changes sf potassiunl concentration in the incubation medium. in the presence and absence of ACTH
Can. J. Physiol. Pharmacol. Downloaded from www.nrcresearchpress.com by UNIV CHICAGO on 11/15/14 For personal use only.
Aldosterone accumulation (ngiinc~batisn)~ time (h)
K' (mequiv. /li trey 5 7.5 10
ACTH (tnU/ml> 0 100 0 100
0 10Q
---
0
24
48
79
111
144
0.5 0.4 0.4 0.7 0.4 0.7
2.6 13.8 4.1 22.7 3.1 12.6
2.5 15.8 5.4 38.8 14.5 43.9
2.3 15.5 7.3 56.7 35.2 101
2.5 15.8 7.2 56.6 46.7 135
2.2 15.0 6.5 68.0 50.8 145
aAdded as KCI. bEa& number is the mean of measurcmcnts o n duplicate incubations. The number of cells iricubated were 1.9 for the cexperiments with K + concentrations of 5.0, 7.5, and 10 rnequiv./litre, respectively
10 days in the presence and absence of ACTH. The latter was added in a concentration of 1QO mU/ml, a concentration which caused an easily detectable increase in aldosterone accumulation. The results of aldosterone measurements on these incubations are shown in Fig. 2. At 2 3 h, aldosterone levels in ACTHcontaining dishes were approximately 16-fold higher than at zero time, and 5-fold higher than in the corresponding unstimulated incubations. There was little or no accumulation of aldosterone in ACTH-stimulat~d or control incubations during the subsequent 9 days of incubation. The corticosterone measurements shown in Fig. 2 are from the same experiment. In a similar experiment with incubations containing 1.0 x 10Qells, the means of replicate measurements ( 1 1 = 4 ) of the aldosterone response to ACTH (100 mU) at 2, 8, 16, 24, and 48 h were 0.52, 1.7, 2.5, 3.5, and 4.4 ng, respectively. In a single experiment, adrenal glands were decapsulated, and separate suspensions prepared from the outer and inner portions. Aliquots of the two suspensions. containing 1.46 x 10,' and 2.07 x 10') cells, respectively, were incubated without and with ACTH (100 mU) for periods up to 24 h. Aldosterone was undetectable or very low (< I ng) in incubations of the inner portions of the glands. In the incubations derived from the outer portion of the glands, the means of duplicate measurements of aldosterone response to ACTH were 2.4, 23, and 31.5 ng at 2, 16, and 24 h, respectively. Table 2 shows the effect on aldosterone accumulation of varying the potassium con-
/
165 1.8
105, and 1.7 x 105
centration in the medium, in the presence and absence of ACTH. Cells obtained from a single preparation were incubated in medium containing K + concentratiolls of 5.0, 7.5. or 10 mequiv./litre for periods up to 134 he As in thc experiment illustrated in Fig. 2, incubations with K i , 5.0 mequiv./litre, showed little accumulation of aldosterone beyond 24 h, cither in the presence or absence of ACTH. In incubations with K ', 10 mequiv./litre, however, aldosterone was found in much greater amounts and continued to increasc through the sixth day, with or without ACTH. Similar, though less marked changes occurred in incubations with 7.5 mequiv. of 161 per litrc.
Discussion The system described permits long term survival of adrenal cells of the normal adult rat. The cclls were non-confluent and no evidence of cell replication was found. Cell numbers fell sharply in the first 24 to 48 h of incubation, but thereafter declined very slowly. Of the two cell types identified only the granular cells could be found after the seventh day of incubation. We cannot say whether the disappearance of pale cells was due to their death or to transformation into granular fonns, nor have we attempted to determine the function of these cells. The corticosterone response to adrenocorticotropic hormone (ACTH) added late in incubation demonstrated the presence of functioning cells through the fourth week of incubation. A corticosterone response was seen to as little as 5.0 pU (50 pg) of ACTH1-24 per millilitre, a concentration which in terms of international units is at the lower end of the
Can. J. Physiol. Pharmacol. Downloaded from www.nrcresearchpress.com by UNIV CHICAGO on 11/15/14 For personal use only.
538
CAN. J . PHYSIOL,. PHARMACOL. VOL. 53, 1975
rangc of concentration of ACTH,-39 found in normal rat serum (kfatsuyama et a[. 1970). Sayers r l al. ( 1 97 1) found that isolated adult rat adrenal cells incubated in a simple buffer solution for 2 h responded to as little as 0.5 pU of ACTH1-3s, per millilitre. Corticosterone accumulation in response to a single addition of 100 mU of ACTH per millilitre at the beginning of incubation was virtually constant for the first 3 days, and declined slowly over the next 5-7 days. Whether the decline in response was due to a change in medium or cells, or due to destruction of ACTH (Bullock et al. 1972), remains to be determined. The data presented in Fig. 4 illustrate a decline in corticosterone response to ACTH with increasing time of incubation of unstirnulated cells. Comparison of the responses to ACTH of unharvested and harvested cells (Figs. 4 and 5) indicate that prolonged incubation without change of medium is not responsible for the decline in responsiveness. Figures 4 and 5 also illustrate the partial restoration by ACTH of steroidogenic activity of the cells. The responses described are reminiscent of those of patients with hypopituitarism and secondary adrenal insufficiency, and the system should prove to be a useful model of such a state. The rat adrenal cell system of O'Hare and Neville ( 3973u, 1973b) did not include cells of the aona glomerulosa. Their system, which employed other enzymes for cell dispersion, a different incubation medium, confluent cultures, and frequent feeding with medium containing ACTH, displayed virtually constant corticosterone production for up to 40 days. Cells incubated without ACTH for up to 8 weeks showed a lag in corticosterone production in response to ACTH, but within 4 or 5 days produced corticosterone at a rate comparable to that of cells maintained with ACTH throughout incubation. Responses to ACTH at concentrations below 1 mU/ml were not detected. B'Hare and Neville give little information about cell numbers during the course of incubation. The data presented in Fig. 2 show that the aldosterone response to ACTH, in medium containing 5.0 mequiv. of K-+ per litre, terminated after 24-48 h, at a time when
the corticosterone response was still near maximum. A similar phenomenon has been described in the human subject given ACTH continuously, that is, ddosterone production increases for a day or so, then declines, at a time when cortisol production is still increasing (Tucci et al. 1967). The findings displayed in Fig. 2 raise the possibility that this in vivn phenomenon is due, at least in part, to an intra-adrenal regulatory mechanism. However, the data of Fig. 2 could equally well be due to the death in vidro of the aldosteroneproducing cells. The latter possibility was eliminated by the finding that addition of increased amounts of Kt- to the medium resulted in an accumulation of aldosterone, which continued through the sixth day of incubation. The effect of increasing the concentration of potassium indicates that aldosterone-secreting cells can survive in this system for several days and that the system should be useful for studies of the regulation of aldosterone production. A significant advantage of the system described, like that of O'Hare and Neville, is that the role of factors involved in the maintenance of adrenal function can be examined in vitrcr without the necessity of treatment of the tissue donor prior to preparation of cell suspensions. The latter consideration is of particular importance in the study of human adrenal cells.
Addendum Following submission of this paper, Hornsby et al. (1974) reported a study of aldosterone production by isolated rat aona glomerulosa cells, incubated for periods up to 20 days in culture medium containing 6 mhf Kt- and 11 mM K +. In their paper Hornsby et al. referred to an earlier communication (Hornsby et al. 1973) which the present authors had not seen. In these two studies, Hornsby et al. showed that aldosterone production was considerably greater in the medium with 11 mM K + . In the latter medium in the absence of ACTH, daily production reached a peak at day five, then declined to very low levels over the next 7-10 days. Addition of ACTH to cultures in medium containing 11 mM K + increased the maximum rate of aldosterone production but shortened the period during which aldosterone was produced.
I'KICE ET AL.: I,ONG TERM INCLJBATION OF ADRENAL CEI,I,S
Can. J. Physiol. Pharmacol. Downloaded from www.nrcresearchpress.com by UNIV CHICAGO on 11/15/14 For personal use only.
We are grateful to Miss H. Baechtold and Miss P. McFadzean for their technical assistance, and t o Drs. 6. Williams and R. H. Underwood for a gift of antiserum for the aldosterone assay. U., and NUSSDORFER, G. G . 1972. Tissue culture ARMATO, of rat adult decapsulated adrenal glands. A methodological, ultrastructural and morphometric investigation. Z.Zellforsch. Mikrask. Anat. 135,24-273. Bur-LOCK,G . R., GII.I,AM, E. A., LOWRY, P. J., MC:MAK.I.IN, C., and PETERS,J . 1972. Observations on the fate of adrenocorticotrophin in isolated rat adrenal cell suspensions and the separation of an enriched adrenocorticotrophin-sensitivesteroidogenic cell fraction. Biochem. J. 129,23 p.(Abstr.). HALKEHSTON, I . 11. K., and FEINSTEIN, M. 1%8. Preparation of ACTH responsive isolated cells from rat adr e n a l ~ Fed. . Proc. 27.626 (Abstr.). HORNSBY, P. J., O'WARE,M . J., and NEVILLE, A. M. 1973. Effect of ACTH on biosynthesis of aldosterone and corticosterone by monolayer culttires of rat adrenal zona glornerulosa cells. Riochern. Biophys. Kes. Coinmun. 54, 1554-1559. 1974. Functional and morphological observations on rat adrenal zona glolnerulosa cells in monolayer culture. Endoc~inology,95, 1240-1251. KAHKI,A. 1. 1973. Inhibition of ACTH-induced differentiation of cortical cells and their mitochondria by corticosterone in tissue culture of fetal rat adrenals. Anat. Rec. 176,253-272. KI.OPPENBORG. P. W. C., ISLAND. D. P.. LIDDLE,G . W., MICHEI-AKIS, A. M., and NICHOLSON, W. E. 1968. Method of preparing adrenal cell s~lspensionsand its applicability to the in vitro study of adrenal metabolism. Endocrinology, 82, 1053-1058. LEVINE,J . PI., LAIDI.AW,J . C.. and RUSE,J . L. 1972. EtTects of a sulfated mycopolysaccharide (KOI-8307) on zona glonlerulosa width and steroid production by the rat adrenal cortex. Can. J . Physiol. Pharmacol. 50, 270-275. MKTSIJYAMA, H., RUHMANN-WENNIIOLL>, A., and NELSON, D. H. 1970. Biologic and immunologic simi-
539
larities between rat and human adrenocorticotropin (ACTH). Endocrinology, 87,756-763. O'HARE,M. J . , and NFVILLE, A. M. 1 9 7 3 ~Morphological . responses to corticotrophin and cyclic AMP by ada~lt rat adrenocortical cells in monolayer culture. J . Endocrinol. 56,529-536. 19736. The steroidogenic response of adult rat adrenocortical cells in mono1:lyer culture. J. Endocrinol. 56,537-549. PEYTREMANN, A., NIC:HOLSON, W. E., BROWN,R. D., LIDDI.~;., G. W., and HARDMAN, J. C. 1973. Comparative effects of angiotensin and ACTH on cyclic AMP and steroidogenesis in isolated bovine adrenal cells. J. Clin. Invest. 5 2 , 8 3 5 4 4 2 . Roos, B. A. 1974. Effect of ACTH and c 4 M P on human adrenocortical growth and function in vitro. Endocrinology, 94,685-690. RUSE, J. L., PRICE:,C., STIF,FF.L, M., and L A I L X A W J.,C. 1972. The influence of glucocorticoids and heparin on aldosterono production. I n Hypertension '72. Edircd by J. Genest, and E. Koiw. Springer-Verlag, Berlin. pp. 326-333. SAYERS,G., SWALI.OW,R. I,., and GIORDANO, N . D. 1971. An improved technic for the preparation of isolated rat adrenal cells: a sensitive, accurate and specific method for the assay of ACTH. Endocrinology, 88, 1063-1068. S I L B E KR. , H., BUSCH,R. D., and OSLAPAS,K. 1958. Practical procedure for estimation of corticosterone or hydrc>cortisone.Clin. Chem. 4,278-285. S w ~ r ..ow. 1 K. I.,. , and SAYERS, 6 . 1969. A technic for the preparation of isolated rat adrenal cells. Proc. Soc. Exp. Biol. Med. 131, 14. 'Tucc~,J. R., ESPINEK, A . , JAGGER, P. I . , P A U K , GL., . and I,AUI.EK,13. P. 1967. ACTH stimulation of aldosterone secretion in normal subjects and in patients with chronic adrenocortical insufficiency. J . Clin. Endocrinol. Metab. 27,568-575, UNDF.RWOOD, R. H., and WILLIAMS, G . H. 1972. The simultaneous measurement of aldoster-one, cortisol, and co~ticosteronein human peripher*al plasma by displacementanalysis. J. Lab. Clin. Med. 79,848-862.
