0013.7227/92/1311-0201$03.00/0 Endocrinology Copyright V; 1992 by The Endocrme
Vol. 131, No. 1 Prrnted in U S.A.
Society
Intermediate Filaments Y- 1 Cells: Acrylamide TIANA
M. SHIVER,
National Institute Bethesda, Maryland
of
DAN
L. SACKETT,
Diabetes, Digestive 20892
and Steroidogenesis in Adrenal Stimulation of Steroid Production
LESLIE
KNIPLING,
AND
and Kidney Diseases National
J. WOLFF
Institutes
of
Health,
ABSTRACT The possible role of intermediate filaments in steroidogenesis was investigated in Y-l mouse adrenal tumor cells by treatment with acrylamide, which is thought to disrupt intermediate filaments without directly affecting microtubules or microfilaments. Treatment of cells with 5 mM acrylamide increases steroidogenesis after a lag period of 46 h and induces rounding of the cells at approximately the same time. The effect of acrylamide on steroidogenesis is not CAMP mediated and occurs before pregnenolone formation. DNA synthesis is inhibited, while protein synthesis is not. Acrylamide does not affect polymerization/depolymerization of microtubules in uitro. Acrylamide stimulation of steroidogenesis is additive with that produced by either colchicine or ACTH, implying that acrylamide, ACTH,
and colchicine act at different rate-limiting steps in steroidogenesis. In addition, acrylamide stimulation is additive with that of forskolin. Pretreatment of cells with taxol, an agent that specifically promotes microtubule polymerization, decreases acrylamide-stimulated (as well as colchicine or ACTH-stimulated) steroidogenesis, implying that there must also be some shared elements in the stimulating pathways. We hypothesize that regulation of steroidogenesis in the Y-l cell depends on 1) disruption of a vimentin or tubulin coat surrounding lipid droplets and 2) possible functional shortening of the distance between cholesterol droplets and the mitochondrion. However, because of interactions between cytoplasmic fibers, it is currently impossible to say whether interruption of any one of them is a direct or indirect stimulus of steroidogenesis. (Endocrinology 131: 201-207, 1992)
T
HE CYTOSKELETON is involved in the coordination of a wide variety of cellular processes. Among these are the morphological and biochemical responses to hormone stimulation (1, 2). The response of steroidogenic cells to appropriate stimulation is an example of such a process, and the Y-l cell line provides a useful model system. This mouse adrenocortical tumor cell line responds to ACTH and other agents by characteristic shape changes (cell rounding), by increasing the production of CAMP, and by increasing the conversion of cholesterol into steroids, which are secreted into the culture medium (3). Agents that increase CAMP, such as forskolin, can mimic these responses, resulting in even more pronounced rounding and steroid production than those caused by ACTH. In addition, agents that target the filament systems of the cytoskeleton may have similar effects. The morphological changes following stimulation, involving transformation from a flat extended cell to a rounded one, clearly involve large changes in the cytoskeleton. This change is accompanied by increased steroidogenesis at a step preceding pregnenolone formation, and it is the movement of cholesterol from its storage droplet to the mitochondrion that constitutes this rate-limiting step. The shape changes and the increased production of steroids are clearly related events (4), but may not necessarily be coupled (5, 6). The involvement of microfilaments (MF) in these repsonses has been shown using agents that bind to actin (5, 7-9), and tubulin-directed agents have demonstrated the involvement of microtubules (MT) (10-15). At high concentrations of these agents, inhibition of steroidogenesis may occur.
Intermediate filaments (IF) comprise the third fiber system of the cytoskeleton. These filaments are composed of any of a family of related proteins and may be cell type specific (16). IF may rearrange upon dissociation of MT, suggesting that the two filament systems can be tightly associated (17, 18). In addition, it has been demonstrated that a variety of hydrophobic or lipophilic molecules, including cholesterol and cholesterol ester, can form tight associations with IF proteins, including vimentin (19-21). Vimentin has been shown to rearrange during the differentiation of preadipocytes, forming a cage that bounds the forming lipid droplets (18). Recently, it has been shown that in Y-l cells, which contain vimentin IF, lipid droplets that store cholesterol have a boundary containing vimentin and are tightly associated with vimentin IF (22). Thus, it was of interest to look for biochemical evidence of the involvement of IF in steroidogenesis. Investigating the role of MT and MF in steroidogenesis (and other cellular processes) is facilitated by the availability of agents that disrupt these filaments with high specificity: colchicine and cytochalasin. Acrylamide may prove to be such an agent for IF, since millimolar concentrations disrupt IF networks without significant effects on cell metabolism or the distribution of MT and MF (23-26). Accordingly, we have undertaken a study of the effects of acrylamide on Y-l cells and here report that this treatment results in cell rounding and an increase in steroid production at a step preceding pregnenolone formation, but no concomitant increase in CAMP.
Received February 26, 1992. Address requests for reprints to: Dr. Tiana Institute of Child Health and Human Development, of Health, Bethesda, Maryland 20892.
Cells
M.
Shiver, National
Materials
National Institutes
or
and Methods
Y-l cells were a gift from Dr. B. Schimmer were obtained from American Type
(Toronto, Culture
Ontario, Collection
201
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 12 November 2015. at 23:16 For personal use only. No other uses without permission. . All rights reserved.
Canada) (Rock-
202
IF AND
STEROIDOGENESIS
ville, MD), and were grown in six-well plates in Ham’s F-10 containing 12.5% horse serum and 2.5% fetal calf serum or in Dulbecco’s Modified Eagle’s Medium (DME)-F-12 (1:l) with 8% horse serum and 2% fetal calf serum, supplemented in both cases with 60 fig/ml penicillin, 100 &ml streptomycin, and 2 rnM glutamine in an atmosphere of 4.5% C02-95.5% air at 37 C. Media were changed two or three times weekly.
Endo. 1992 Vol 131 .No 1
protein was monitored by turbidity at 350 nm using a Cary 219 spectrophotometer (Varian Instruments, Sugar Land, TX). The sample chamber was maintained at 37 C with a thermostatically controlled Lamda K2-R circulator (Brinkman Instruments, Westbury, NY). Details of this procedure have been described previously (30).
Results
Steroids After experimental manipulations, the cell medium was collected, and steroids were measured fluorometrically, as previously described (lo), after extraction of steroids from the medium by a modification of the procedure of Ramirez et al. (27). Briefly, l-ml samples of medium were diluted with 6 ml 0.01 M acetate buffer, pH 4.5, and 3-ml aliquots were loaded onto each of two columns containing 200 mg Cl8 absorbant (PGC Scientific, Gaithersbure. MD). which had been orewashed with 3 ml methanol and 3 ml acetate buffer. The sample was allowed to flow through the column and was followed by two washes of 3 ml buffer and elution with 3 ml methanol. The methanol was evaporated, 1 ml 65% sulfuric acid-35% ethanol was added, and steroid fluorescence (excitation at 470 nm, emission at 530 nm) was measured, as previously described (lo), using 20n-dihydroprogesterone as standard.
CAMP The production of CAMP is reliably assayed in the medium of Y-l cells (3). Media for these experiments were supplemented with the phosphodiesterase inhibitor Ro 20-1724 at 0.1 mM. Media were collected and made 2 rnM in EDTA and 60 rnM in acetic acid. Measurement of CAMP was performed by RIA, essentially as described previously (15).
Incorporation
of thymidine
and leucine
Cells were grown in the usual manner and transfered to the appropriate media for uptake measurements. Thymidine uptake was measured in DME or DME-F-12 with the same results. Leucine uptake was measured in leucine-free DME. Growth media were removed and replaced with fresh media, supplemented with [3H]leucine (New England Nuclear, Boston, MA; 135 Ci/mmol) or [3H]thymidine (New England Nuclear; 6.76 Ci/mmol) with or without test agents. Plates were incubated overnight. The following day, the media were removed, and the cells were washed with cold PBS and resuspended in 10% trichloroacetic acid (TCA) on ice. TCA-insoluble material was collected on Whatman GF/C glass fiber filters (Clifton, NJ), which were washed five times with 5% TCA, dried, and counted in a scintillation counter. Parallel plates were used to determine cell protein per well, as described below.
Treatment
of cells
Acrylamide (electrophoresis grade, 161-0100, Bio-Rad, Richmond, CA) and ACTH (Sigma Chemical Co., St. Louis, MO) were added to media from aqueous solutions; taxol (a gift of Dr. Matthew Suffness, NCI) was added from a solution in dimethylsulfoxide. Appropriate solvent controls were performed. After treatment, cells were routinely scored for cell rounding before harvesting and assay of steroid and CAMP production. All results represent at least duplicate determinations on two separate dishes (wells) of cells; in most cases three separate dishes were used. All experiments were repeated at least twice. Cell mass was estimated by protein determination. After removal of growth medium, the cell monolayer was washed twice with PBS and extracted with 1 ml 0.1 N NaOH, and protein was determined with the bicinchoninic acid method (Pierce Chemical Co., Rockford, IL).
Preparation
of MT protein
Rat brain MT protein was purified by two cycles of temperaturedependent polymerization and depolymerization in Mes assembly buffer [O.l M 2(N-morpholino)ethanesulfonic acid (Mes), 1 rnM MgCl*, and 1 mM EGTA, pH 6.91 with 1 rnM GTP (28). Purified protein was stored in liquid nitrogen, as previously described (29). Polymerization of MT
Morphology
of the Y-l adrenal
cell culture
The typical Y-l cell grows in a flat epithelial-like monolayer (Fig. 1A). This morphology and the time course of rounding upon exposure to acrylamide are shown in Fig. 1 (A-E). The refractile rounded morphology typical of stimulated cells is shown by white arrows in Fig. lD, and the percent rounding in each panel is given in the legend. After treatment with acrylamide for 1.5 h (Fig. lB), little change was observed in the rounding of the cells; after 2 h, partial rounding was seen (Fig. lC), and after 2.5 h of treatment, about half of the cells were rounded (Fig. 1D). By 3.5 h of treatment, greater than 90% of the cells were rounded (Fig. 1E). This distinct morphological change of near-uniform rounding occurring between 2-3 h of treatment with acrylamide differs from that observed after 3 h of ACTH treatment, in which case patchy nonuniform rounding was seen (Fig. 1F). The degree of cell rounding in response to ACTH varies with different clones of Y-l cells. Effect of a&amide
on cell function
and growth
Treatment of Y-l cells with ACTH inhibited cell growth (31). Overnight treatment with acrylamide reduced thymidine incorporation beyond that observed with ACTH alone, indicating that acrylamide also inhibits cell growth (Table 1). Protein synthesis, however, was not inhibited by ACTH or acrylamide, since [3H]leucine incorporation during overnight incubation with either agent was the same as that in the controls. Effect of acrylamide
on steroid and CAMP production
Steroidogenesis induced by ACTH shows a rapid early increase, followed by a slower continuing rate of increase. This has been observed previously (10). By contrast, acrylamide enhanced steroidogenesis in a different temporal pattern, with a slow early phase, followed by a more rapid phase starting after about 5 h of exposure (Fig. 2), shortly after the time corresponding to the time of maximal rounding of the cells (Fig. 1E) and similar to the time of delayed stimulation seen after colchicine treatment (10). Like colchitine, the effects of acrylamide on steroid production occur at a step preceding pregnenolone formation; Table 2 illustrates that addition of acrylamide to cells treated with pregnenolone fails to produce a further increase in steroid production above that observed with the addition of pregnenolone alone. A striking difference between ACTH and acrylamide stimulations is depicted in Fig. 3. The rapid burst in CAMP production and subsequent maintenance of elevated levels after ACTH treatment were not seen after exposure of Y-l cells to 5 mM acrylamide. In fact, acrylamide values did not differ significantly from control values over a period of 24 h.
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 12 November 2015. at 23:16 For personal use only. No other uses without permission. . All rights reserved.
IF AND
STEROIDOGENESIS
203
FIG. 1. Time course of cell rounding in the presence of acrylamide. Phase micrographs of treated and untreated Y-l adrenal cortical cells are shown. Untreated cells are flat, and the typical rounded morphology seen after stimulation is indicated by the three white arrows in D; note the highly refractile character of these rounded cells. For each time point in the figure, a total of 400-500 cells from 3-4 fields were scored for rounding. The percent rounding (*SD) is given for each panel. A, Control: normal flattened appearance of Y-l cells. Rounding = 1.6 + 1%. B-E, Acrylamide treatment (5 mM). B, Incubated for 1.5 h, showing few morphological changes. Rounding = 6.3 t 2.2%. C, Two-hour incubation, showing the appearance of rounded cells. Rounding = 27 + 8%. D, Two and a half-hour incubation, showing significant rounding of cells. Rounding = 43.2 k 7%. E, Three and a half-hour incubation, showing rounding of essentially all cells. Rounding = essentially total. F, ACTH treatment (0.125 NM), 3-h incubation, showing typical rounding of scattered cells. The bar between B and E = 50 pm.
It is apparent, therefore, that the mechanismsof stimulation of steroidogenesisby ACTH and acrylamide differ. Moreover, acrylamide-induced steroidogenesis is unlikely to be basedon a CAMP-dependent mechanism. Effect of acrylamide on MT Previous studies have shown that treatment of PtK-1 cells with acrylamide disrupted IF while leaving MT intact (25, 26). The lack of a direct effect of acrylamide on MT is confirmed by in vitro studies on the polymerization of MT protein, as shown in Fig. 4. MT assembly, as measured by optical density at 350 nm at 37 C in the presence of GTP was not affected by very high (50 mM) concentrations of acrylamide (arrow in Fig. 4A). Moreover, MT in the presence of acrylamide exhibited temperature-sensitive depolymerization like untreated MT. Furthermore, as shown in Fig. 4B, MT formed in the presence of 50 mM acrylamide, polymerized normally, and repolymerized normally. There is, thus, every reason to believe that the effects seen with acrylamide in Y-l cells are not, primarily, exerted on MT. This is also shown by a comparison of acrylamide and colchicine in intact Y-l adrenal cells. The in viva effects of acrylamide on MT were tested by
comparing treatment of cellswith colchicine, the potent antiMT agent, in the presenceand absenceof acrylamide. Figure 5 illustrates that the effects of colchicine and acrylamide were additive, as were the effects of ACTH and acrylamide when used together in the Y-l cell. As shown by the dashed lines in Fig. 5, these additivities were essentially identical to those calculated from effects produced when the agentswere incubated separately. Finally, we found (Fig. 5C) that forskolin, which is frequently used to stimulate CAMP-dependent reactions without intervention of membrane receptors, promoted vigorous steroid production, and this stimultation was additive with acrylamide enhancement of steroidogenesis. Because simple additivity was noted, we have made the presumptive conclusion that these agents can act at steps different from those of acrylamide; the relative importance of the different stepsmay depend on the mode of stimulation of steroidogenesis.The results in Table 3 illustrate that pretreatment of cells with taxol results in decreasedacrylamidestimulated production of steroids. This is in keeping with previous observations (15) that pretreatment of Y-l cells with taxol, which specifically promotes MT polymerization, causes a reduction in basal and ACTH-stimulated steroid prodution. This suggeststhat despite a considerable independence of the stimulatory mechanisms mentioned above, these same
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 12 November 2015. at 23:16 For personal use only. No other uses without permission. . All rights reserved.
IF AND
204 TABLE 1. Effect incorporation
of acrylamide
on leucine
and thymidine
Acid-insoluble
radioactivity
Leucine kdpm
Control ACTH (0.1 PM) Acrvlamide (4.5
(12.6)
1.0 0.92
mM)
TABLE 2. Effect pregnenolone
Thymidine
AVW3~e” zk 0.09
0.92 -c 0.07
Endo. Voll31.
STEROIDOGENESIS
kdpm
AWK3ge”
(3.54)
1.0 0.89 k 0.02 0.58 f 0.12
Cells were grown to about half-confluency, and medium was removed and replaced with fresh medium containing [“Hlleucine or [“Hlthymidine plus the indicated agents. Leucine uptake was performed in leucine-free DME using cells grown for l-2 days in DME. Thymidine uptake was performed in DME or DME-F-12 using cells grown for l2 days in the appropriate medium. Thymidine uptake results were the same in both media. [“H]Leucine (135 Ci/mmol) was added to give a final concentration of 2.2 mM in the medium; [“Hlthymidine (6.76 Ci/ Fmol) was added to give a final concentration of 0.5 fiM in the medium. Agents were added to give the indicated concentrations, and cells were exposed to medium with agent and tracer at the same time and incubated overnight (-16 h). After incubation, the cells were washed, and TCA-insoluble radioactivity was determined as described in Muterials and Methods. Results presented are averages f SE from three separate experiments of two wells each. The disintegration per min/ mg protein is given for the control, and other results are normalized to the control. ’ Average value normalized to the control.
Basal Pregnenolone Prennenolone
of acrylamide
on steroid
production
1992 No 1
from
Steroid
produced
(udmr
urotein)
0.2 2.76 + 0.5 2.68 + 0.6
+ acrvlamide
Cells were grown to half-confluency. Media were removed and replaced with fresh media supplemented with 5 mM acrylamide alone or media plus 5 mM acrylamide and pregnenolone (10 fig/ml). Media were removed and assayed as described in Materials and Methods. Duplicate determinations were made on duplicate wells.
,
a i 5 0
200 I ys, 5
10
15 20 TIME, h
Acryl. Control
25
FIG. 3. Time course of CAMP production. Cells and conditions are described in Fig. 2. After removal of the medium, CAMP was determined by RIA, as described in Materials and Methods. As in Fig. 2, each point represents duplicate determinations on triplicate wells of cells. Total CAMP is plotted as a function of time. Representative SEs are shown. Acryl., Acrylamide.
TIME,
h
FIG. 2. Time course of steroid production. Cells were grown to about half-confluency, and at time zero, media were removed and replaced with fresh media containing 0.1 mM Ro 20-1724 supplemented with acrylamide (Acryl, 5 mM), ACTH (0.125 PM) or no addition, as indicated. At the indicated times, media were removed, and steroid production was assayed. Each point represents duplicate determinations on triplicate wells of cells. Total steroid is plotted as a function of time. Representative SES are shown.
filamentous other.
elements
are not entirely
independent
of each
Discussion The cytoplasm is a dense complex network of interlaced fibers (MT, IF, and MF) in,whose interstices reside the various organelles that may be free floating, near one or another fiber, or actually attached to one or more of these. The cholesterol ester-rich lipid droplet of adrenal cortical cells is one of these organelles. The fiber network might either
promote movement of lipid droplets along one of the fibers with the aid of a motor, such as kinesin, or it might act as a barrier to migration of droplets toward the mitochondrion. Colchicine, cytochalasin-D, and acrylamide are thought to cause their intracellular effects by functional disruption or redistribution of the cytoskeletal fibers: MT, MF, and IF, respectively. Usually the particular fiber-mediated process is inhibited, which is then interpreted as a requirement for that fiber to translocate a particular organelle. It was, therefore, a surprise that secretion of steroids (which are not stored in granules in the adrenal cell) was enhanced upon incubation of Y-l cells with colchicine or its congeners in the absence of ACTH (10, 11, 14). Since the cholesterol ester precursor of steroids is packaged in lipid droplets, the possibility existed that MT-disrupting agents might act at this earlier stage in steroid secretion, and we proposed that the MT was a barrier to the access of cholesterol ester droplets to the mitochondrion (10). Similar explanations, but implicating MF, might apply to the enhancement of basal steroidogenesis by cytochalasin-B or -D (5, 9, 32). The results of the present paper show that manipulation of IF networks by acrylamide enhances basal steroidogenesis and causes cell rounding, but does not change cellular CAMP concentrations. These phenomena are similar to those produced by disruption of MT, but acrylamide does not cause gross changes in MT or actin filaments. Thus, the effects of
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 12 November 2015. at 23:16 For personal use only. No other uses without permission. . All rights reserved.
IF AND
STEROIDOGENESIS
205
IA
I
6 F B E zi 8 E'
250-
400
200-
T.
L 150-
300
loo-
200
2 8E P i
100
50-
U-J
T 1
2
3
1
2
3
5. Additivity of acrylamide effect with that of ACTH, colchicine, and forskolin. Cells were grown and treated, and steroid was assayed as described in Fig. 2. Plates were incubated overnight (-16 h) at 37 C. In A and B, 1 represents no addition, 2 represents 0.1 FM ACTH, and 3 represents 5 pM colchicine. A, The results of these treatments alone. B, The results of these treatments in the presence of 2.5 mM acrylamide, added at the same time as the other agents. C, The results of treatment with 1 pM forskolin (no. 1) and 1 pM forskolin plus 2.5 mM acrylamide added at the same time (no. 2). The ordinate scale for A and B is on the left, and that for C is on the right. The dotted lines in B and C represent the results expected if the effects of acrylamide and the treatments are additive: B, 2, ACTH plus acrylamide; B, 3, colchicine plus acrylamide; C, 2, forskolin plus acrylamide. All results are expressed as the percent increase over the control steroid production (0.524 -C 0.089 pg/mg protein) found in the absence of any added agent (A, 1). Data represent the mean + SE of duplicate determinations on a total of 11 replicates from 4 separate experiments.
FIG.
TIME
(seconds)
FIG. 4. Lack of effect of acrylamide on in uitro polymerization of MT. A, Rat brain MT protein (1.5 mg/ml) in Mes assembly buffer plus 1 mM GTP was polymerized at 37 C. At the time indicated by the arrow, acrylamide (in Mes assembly buffer) was added to 50 mM. The optical density was monitored for several minutes further, after which the cuvet was placed on ice for 5 min (broken line) to measure cold-induced depolymerization. B, Rat brain MT protein (1.2 mg/ml) in Mes assembly buffer plus 1 mM GTP was polymerized at 37 C in the absence (upper curue) or presence (lower curue) of 50 mM acrylamide. After polymerization plateaued, the cuvets were transferred to ice for 5 min to depolymerize the protein (broken lines). After warming, the repolymerization was monitored, followed by another depolymerization on ice (second set of broken lines).
the three fiber types show a degree of independence from each other, and this is confirmed by the additivity of colchitine- and acrylamide-mediated steroidogenesis. Several observations point to more complicated mechanisms for stimulation of steroidogenesis by disruption of cytoskeletal elements, and two additional mechanisms come to mind: 1) disruption of tubulin or vimentin coats surrounding cholesterol droplets and 2) redistribution of mitochondria. The coat hypothesis enjoys some experimental support. While coats around lipid droplets had been identified in the past, the nature of the coating material was not. Clark and Shay (33, 34) showed that cholesterol droplets were coated with tubulin in an unspecified polymeric form that could be stained with antibodies. ACTH and colchicine removed this layer (33, 34). More recently, it has been shown that neutral lipid droplets in adipocytes were encased in IF networks (18). In Y-l cells it has been shown that approximately one third of the cholesterol droplets is firmly attached to vimentin fibers (22). Another third is present as (apparently) free droplets, while the last third of the droplets is attached less firmly than to IF, presumably to other members of the cytoskeleton, i.e. MT and MF. It is conceivable that these differently attached cholesterol droplets may well respond
TABLE
3. Taxol steroidogenesis
inhibition
Treatment Control Acrylamide Taxol + acrylamide
of acrylamide-stimulated
Steroid
(&mg
mot&n)
Steroid (normalized)
1.24 (kO.07) 2.83 + 0.11
2.28 f 0.09
2.10 & 0.04
1.69 + 0.03
1
Cells were grown as described in Materials and Methods. Agents were added from concentrated stocks to fresh medium to the desired final concentration, 5 mM for acrylamide and 0.01 mM for taxol. The cells exposed to taxol plus acrylamide were exposed to taxol alone during a l-h preincubation before being exposed to taxol and acrylamide together for the remainder of the exposure period. After overnight incubation, steroids and proteins were determined, as described in Materials and Methods, on duplicate samples from duplicate wells; the values reported are the mean and SD of these four values. Similar results were obtained in replicate experiments.
differently to various stimuli and may account for partial responses and for additivity. The affinity of vimentin for lipids is shown by binding of a variety of phospholipid vesicles and cholesterol to isolated vimentin filaments (19, 20); this occurs by electrostatic interactions of arginine residues of the N-terminal vimentin domain with the phosphates of acidic phospholipid and hydrophobic interactions between nonpolar amino acids of the N-terminus and uncharged lipid moieties, including cholesterol (35). The response of the vimentin coat to acrylamide has not yet been studied, but the findings presented here would be consistent with the
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 12 November 2015. at 23:16 For personal use only. No other uses without permission. . All rights reserved.
IF AND
206
STEROIDOGENESIS
uncoating hypothesis applied to vimentin. Finally, it has often been noted that mitochondria may be near or attached to MT and/or IF (36-38). Disruption of MT or IF leads to a marked redistribution of mitochondria (38, 39). Other organelles may also be attached (36), and such associations are critical for intracellular transport promoted by motors such as kinesin or cytoplasmic dynein. In principle, interruption of these associations could shorten the distance between droplet and mitochondrion. We suggest that both uncoating of lipid droplets and mitochondrial or droplet movement may together markedly facilitate access of cholesterol to the side-chain cleavage site in the mitochondrion. It is improbable, however, that the above events can be unambiguously ascribed to action on a single type of cytoskeletal fiber, because there is clear evidence that these three classes of intracellular fibers can interact with each other, e.g. through MT-associated proteins, synapsin, etc., that can bind to MT, actin filaments, and/or IF, or through the binding of two different fibers to the same organelle (38, 40, 41). It is important, therefore, to ascertain whether the disruption of one class of filaments exerts its steroidogenic stimulation indirectly through an effect on another filament. Generally, addition of cytochalasin to cells does not grossly change MT distribution, and conversely, colcemid has little effect on stress fibers or actin networks (23). On the other hand, disruption of MT led to a marked rearrangement of the IF network (18, 42). Thus, colchicine might stimulate through a loss of MT restraints on IF distribution. The fact that taxol inhibits both colchicine-stimulated (12, 15) and acrylamidestimulated steroidogenesis supports such a mechanism, provided that taxol has no direct effect on IF. Normal steroidogenesis would then depend in part on the dynamic instability of MT (43), which could effect temporary changes in the vimentin filaments. Whether such instabilities also occur in the tubulin that coats cholesterol droplets is not known at present. Similarly, factors that remove the vimentin coat have yet to be directly demonstrated. It is clear that the proteins of the major classes of cytoplasmic filaments regulate steroidogenesis in Y-l adrenal tumor cells. Whether they do so as the usually identified fibrous polymers or as coats around the lipid droplets (in the case of tubulin and vimentin) remains to be resolved. Because the filaments interact with each other through associated proteins or organelles, it is currently not possible to say whether interruption of any one of them is a direct or indirect stimulus to steroidogenesis. It is not clear to what extent the ACTH/cAMP pathway overlaps the cytoskeletal pathway, except that some portions of it are different since they are additive and only one uses CAMP. References 1. Zor U 1983 Role of cytoskeletal organization in the regulation of adenylate cyclase-cyclic adenosine monophosphate by hormones. Endocr Rev 4:1-21 2. Hall PF 1984 The role of the cytoskeleton in hormone action. Can J Biochem Cell Biol 62:653-665 3. Schimmer BP 1979 Adrenocortical Y-I cells. Methods Enzymol 58:570-574 4. Betz G, Hall PF 1987 Steroidogenesis in adrenal tumor cells: influ-
Endo. 1992 Vol 131 *No 1
ence of cell shape. Endocrinology 120:2547-2554 5. Cortese F, Wolff I 1978 Cvtochalasin-stimulated steroidoeenesis from high’density fipoprotei&. J Cell Biol 77:507-516 ” 6. Ichikawa Y 1989 Composition of culture media for steroid hormone secretion by murine adrenal tumor cells, Y-l clone. Acta Med Okayama 43:97-103 7. Mrotek JJ, Hall PF 1975 The influence of cytochalasin B on the response of adrenal tumor cells to ACTH and cyclic AMP. Biochem Biophys Res Commun 64:891-896 8. Hall PF, Charpponnier C, Nakamura M, Gabbiani G 1979 The role of microfilaments in the response of adrenal tumor cells to adrenocorticotropic hormone. J Biol Chem 254:9080-9084 9. Rainey WE, Shay JW, Mason JI 1984 The effect of cytochalasin D on steroid production and stress fiber organization in cultured bovine adrenocortical cells. Mol Cell Endocrinol 35:189-197 10. Temple R, Wolff J 1973 Stimulation of steroid secretion by antimicrotubular agents. J Biol Chem 248:2691-2698 11. Ray P, Strott CA 1978 Stimulation of steroid synthesis by normal rat adrenocrotical cells in response to antimicrotubular agents. Endocrinology 103:1281-1288 12. Rainey WE, Kramer RE, Mason JI, Shay JW 1985 The effects of taxol, a microtubule-stabilizing drug, on steroidogenic cells. J Cell Physiol 123:17-24 13. Carnegie JA, Dardick I, Tsang BK 1987 Microtubules and the gonadotropic regulation of granulosa cell steroidogenesis. Endocrinology 120:819-828 14. Kotani S, Murofushi H, Sakai H 1988 Dual effect of antimitotic drugs on steroid secretion in mouse adrenocortical Y-l tumor cells. Cell Struct Funct 13:171-177 15. Sackett DL, Wolff J 1986 Cyclic AMP-independent stimulation of steroidogenesis in Y-l adrenal tumor cells by antimitotic agents. Biochim Biophys Acta 888: 163-l 70 16. Stewart M 1990 Intermediate filaments: structure, assembly and molecular interactions. Curr Opin Cell Biol 2:91-100 17. Franke WW, Schmid E, Osborn M, Weber K 1978 Different intermediate-sized filaments distinguished by immunofluorescence microscopy. Proc Nat1 Acad Sci USA 75:5034-5038 18. Franke WW, Hergt M, Grund C 1987 Rearrangement of the vimentin cytoskeleton during adipose conversion: formation of an intermediate filament cage around lipid globules. Cell 49:131-141 19. Traub P, Perides G, Scherbarth A, Traub U 1985 Tenacious binding of lipids to vimentin during its isolation and purification from Ehrlich ascites tumor cells. FEBS Lett 193:217-221 20. Traub P, Perides G, Kuhn S, Scherbarth A 1987 Efficient interaction of nonpolar lipids with intermediate filaments of the vimentin type. Eur J Cell Biol 43:55-64 21. Asch HL, Mayhew E, Lazo RO, Asch BB 1990 Lipids noncovalently associated with keratins and other cytoskeletal proteins of mouse mammary epithelial cells in primary culture. Biochim Biophys Acta 1034:303-308 22. Almahbobi G, Hall PF 1990 The role of intermediate filaments in adrenal steroidogenesis. J Cell Sci 97:679-687 23. Durham HD, Pena S, Carpenter S 1983 The neurotoxins 2,5 hexanedione and acrylamide promote aggregation of intermediate filaments in cultured fibroblasts. Muscle Nerve 6:631-637 24. Eckert BS 1985 Alteration of intermediate filament distribution in PtKl cells by acrylamide. Eur J Cell Biol 37:169-174 25. Eckert BS 1986 Alteration of the- distribution of intermediate filaments in PtKl cells by acrylamide. II. Effect on the organization of cytoplasmic organelles. Cell Motil Cytoskel 6:15-24 26. Eckert BS, Yeagle PL 1988 Acrylamide treatment of PtKl cells causes dephosphorylation of keratin polypeptides. Cell Motil Cytoskel 11:24-30 27. Ramirez LC, Millot C, Maume BF 1982 Sample purification using a C(18)-bonded reversed phase cartridge for the quantitative analysis of corticosteriods in adrenal cell cultures by high-performance liquid chromatography or gas chromatography/mass spectrometry. J Chromatogr 229:267-281 28. Shelanski ML, Gaskin F, Cantor CR 1973 Microtubule assembly in the absence of added nucleotides. Proc Nat1 Acad Sci USA 70:765-768 29. Sackett DL, Lippoldt R 1991 Thermodynamics of reversible mon-
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 12 November 2015. at 23:16 For personal use only. No other uses without permission. . All rights reserved.
IF AND omer-dimer
association
30. Bhattacharyya
of tubulin.
Biochemistry
30:3511-3517 hybrid
8, Sackett DL, Wolff J 1985 Tubulin,
and tubulin
STEROIDOGENESIS 37. Traub P 1985 Intermediate
Filaments:
A Review.
Springer-Verlag,
Heidelberg
dimers,
S. I Biol Chem 260:10208-10216 In: Sato G (ed) Functionally Differentiated Cell Lines. Liss, New York, p 61 32. Mattson P, Kowal J 1982 Effects of cytochalasin B on unstimulated and ACTH-stimulated adrenocortical tumor cells in vitvo. Endocrinology 111:1632-1647 33. Clark MA, Shay JW 1979 The response of whole and enucleated adrenocortical tumor cells (Y-l cells) to ACTH treatment. Scanning Electron Microsc 3:527-535 34. Clark MA, Shay JW 1981 The role of tubulin in the steroidogenic response of murine adrenal and rat Leydig cells. Endocrinology 109:2261-2263 35. Perides G, Harter C, Traub P 1987 Electrostatic and hydrophobic interactions of the intermediate filament protein vimentin and its amino terminus with lipid bilayers. J Biol Chem 262:13742-13749 36. Rousset BAT 1988 Functions of cytoskeletal elements in organelle distribution. In: Rousset BAF (ed) Structures and Functions of the Cytoskeleton. Colloque INSERM. Libby, London, vol 171:275-400
207
38. Bereiter-Hahn
31. Schimmer BP i9Sl
J 1990 Behavior of mitochondria in the living cell. 122:1-63 Summerhayes IC, Wong D, Chen LB 1983 Effect of microtubules and intermediate filaments on mitochondrial distribution. J Cell Sci 61:87-105 Griffith L, Pollard TD 1978 Evidence for actin filament-microtubule interaction mediated by microtubule-associated protein. J Cell Biol 78:958-965 Sattilaro RF, Dentler WL, LeCluyse EL 1981 Microtubule-associated proteins (MAPS) and the organization of actin filaments in vitro. J Cell Biol 90:467-473 lnt Rev Cytol
39. 40.
41.
42. Bershadsky AD, Ivanoya OY, Lyass LA, Pletyushkina OY, Vasiliev JM, Gelfand IM 1990 Cytoskeletal reorganizations responsible
43.
for the phorbol ester-induced formation of cytoplasmic processes: possible involvement of intermediate filaments. Proc Nat1 Acad Sci USA 87:1884-1888 Mitchison T, Kirschner M 1984 Dynamic instability of microtubule growth. Nature 312:237-242
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 12 November 2015. at 23:16 For personal use only. No other uses without permission. . All rights reserved.
Vol. 131, No. 1 Prrnted in U S.A.
Society
Intermediate Filaments Y- 1 Cells: Acrylamide TIANA
M. SHIVER,
National Institute Bethesda, Maryland
of
DAN
L. SACKETT,
Diabetes, Digestive 20892
and Steroidogenesis in Adrenal Stimulation of Steroid Production
LESLIE
KNIPLING,
AND
and Kidney Diseases National
J. WOLFF
Institutes
of
Health,
ABSTRACT The possible role of intermediate filaments in steroidogenesis was investigated in Y-l mouse adrenal tumor cells by treatment with acrylamide, which is thought to disrupt intermediate filaments without directly affecting microtubules or microfilaments. Treatment of cells with 5 mM acrylamide increases steroidogenesis after a lag period of 46 h and induces rounding of the cells at approximately the same time. The effect of acrylamide on steroidogenesis is not CAMP mediated and occurs before pregnenolone formation. DNA synthesis is inhibited, while protein synthesis is not. Acrylamide does not affect polymerization/depolymerization of microtubules in uitro. Acrylamide stimulation of steroidogenesis is additive with that produced by either colchicine or ACTH, implying that acrylamide, ACTH,
and colchicine act at different rate-limiting steps in steroidogenesis. In addition, acrylamide stimulation is additive with that of forskolin. Pretreatment of cells with taxol, an agent that specifically promotes microtubule polymerization, decreases acrylamide-stimulated (as well as colchicine or ACTH-stimulated) steroidogenesis, implying that there must also be some shared elements in the stimulating pathways. We hypothesize that regulation of steroidogenesis in the Y-l cell depends on 1) disruption of a vimentin or tubulin coat surrounding lipid droplets and 2) possible functional shortening of the distance between cholesterol droplets and the mitochondrion. However, because of interactions between cytoplasmic fibers, it is currently impossible to say whether interruption of any one of them is a direct or indirect stimulus of steroidogenesis. (Endocrinology 131: 201-207, 1992)
T
HE CYTOSKELETON is involved in the coordination of a wide variety of cellular processes. Among these are the morphological and biochemical responses to hormone stimulation (1, 2). The response of steroidogenic cells to appropriate stimulation is an example of such a process, and the Y-l cell line provides a useful model system. This mouse adrenocortical tumor cell line responds to ACTH and other agents by characteristic shape changes (cell rounding), by increasing the production of CAMP, and by increasing the conversion of cholesterol into steroids, which are secreted into the culture medium (3). Agents that increase CAMP, such as forskolin, can mimic these responses, resulting in even more pronounced rounding and steroid production than those caused by ACTH. In addition, agents that target the filament systems of the cytoskeleton may have similar effects. The morphological changes following stimulation, involving transformation from a flat extended cell to a rounded one, clearly involve large changes in the cytoskeleton. This change is accompanied by increased steroidogenesis at a step preceding pregnenolone formation, and it is the movement of cholesterol from its storage droplet to the mitochondrion that constitutes this rate-limiting step. The shape changes and the increased production of steroids are clearly related events (4), but may not necessarily be coupled (5, 6). The involvement of microfilaments (MF) in these repsonses has been shown using agents that bind to actin (5, 7-9), and tubulin-directed agents have demonstrated the involvement of microtubules (MT) (10-15). At high concentrations of these agents, inhibition of steroidogenesis may occur.
Intermediate filaments (IF) comprise the third fiber system of the cytoskeleton. These filaments are composed of any of a family of related proteins and may be cell type specific (16). IF may rearrange upon dissociation of MT, suggesting that the two filament systems can be tightly associated (17, 18). In addition, it has been demonstrated that a variety of hydrophobic or lipophilic molecules, including cholesterol and cholesterol ester, can form tight associations with IF proteins, including vimentin (19-21). Vimentin has been shown to rearrange during the differentiation of preadipocytes, forming a cage that bounds the forming lipid droplets (18). Recently, it has been shown that in Y-l cells, which contain vimentin IF, lipid droplets that store cholesterol have a boundary containing vimentin and are tightly associated with vimentin IF (22). Thus, it was of interest to look for biochemical evidence of the involvement of IF in steroidogenesis. Investigating the role of MT and MF in steroidogenesis (and other cellular processes) is facilitated by the availability of agents that disrupt these filaments with high specificity: colchicine and cytochalasin. Acrylamide may prove to be such an agent for IF, since millimolar concentrations disrupt IF networks without significant effects on cell metabolism or the distribution of MT and MF (23-26). Accordingly, we have undertaken a study of the effects of acrylamide on Y-l cells and here report that this treatment results in cell rounding and an increase in steroid production at a step preceding pregnenolone formation, but no concomitant increase in CAMP.
Received February 26, 1992. Address requests for reprints to: Dr. Tiana Institute of Child Health and Human Development, of Health, Bethesda, Maryland 20892.
Cells
M.
Shiver, National
Materials
National Institutes
or
and Methods
Y-l cells were a gift from Dr. B. Schimmer were obtained from American Type
(Toronto, Culture
Ontario, Collection
201
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 12 November 2015. at 23:16 For personal use only. No other uses without permission. . All rights reserved.
Canada) (Rock-
202
IF AND
STEROIDOGENESIS
ville, MD), and were grown in six-well plates in Ham’s F-10 containing 12.5% horse serum and 2.5% fetal calf serum or in Dulbecco’s Modified Eagle’s Medium (DME)-F-12 (1:l) with 8% horse serum and 2% fetal calf serum, supplemented in both cases with 60 fig/ml penicillin, 100 &ml streptomycin, and 2 rnM glutamine in an atmosphere of 4.5% C02-95.5% air at 37 C. Media were changed two or three times weekly.
Endo. 1992 Vol 131 .No 1
protein was monitored by turbidity at 350 nm using a Cary 219 spectrophotometer (Varian Instruments, Sugar Land, TX). The sample chamber was maintained at 37 C with a thermostatically controlled Lamda K2-R circulator (Brinkman Instruments, Westbury, NY). Details of this procedure have been described previously (30).
Results
Steroids After experimental manipulations, the cell medium was collected, and steroids were measured fluorometrically, as previously described (lo), after extraction of steroids from the medium by a modification of the procedure of Ramirez et al. (27). Briefly, l-ml samples of medium were diluted with 6 ml 0.01 M acetate buffer, pH 4.5, and 3-ml aliquots were loaded onto each of two columns containing 200 mg Cl8 absorbant (PGC Scientific, Gaithersbure. MD). which had been orewashed with 3 ml methanol and 3 ml acetate buffer. The sample was allowed to flow through the column and was followed by two washes of 3 ml buffer and elution with 3 ml methanol. The methanol was evaporated, 1 ml 65% sulfuric acid-35% ethanol was added, and steroid fluorescence (excitation at 470 nm, emission at 530 nm) was measured, as previously described (lo), using 20n-dihydroprogesterone as standard.
CAMP The production of CAMP is reliably assayed in the medium of Y-l cells (3). Media for these experiments were supplemented with the phosphodiesterase inhibitor Ro 20-1724 at 0.1 mM. Media were collected and made 2 rnM in EDTA and 60 rnM in acetic acid. Measurement of CAMP was performed by RIA, essentially as described previously (15).
Incorporation
of thymidine
and leucine
Cells were grown in the usual manner and transfered to the appropriate media for uptake measurements. Thymidine uptake was measured in DME or DME-F-12 with the same results. Leucine uptake was measured in leucine-free DME. Growth media were removed and replaced with fresh media, supplemented with [3H]leucine (New England Nuclear, Boston, MA; 135 Ci/mmol) or [3H]thymidine (New England Nuclear; 6.76 Ci/mmol) with or without test agents. Plates were incubated overnight. The following day, the media were removed, and the cells were washed with cold PBS and resuspended in 10% trichloroacetic acid (TCA) on ice. TCA-insoluble material was collected on Whatman GF/C glass fiber filters (Clifton, NJ), which were washed five times with 5% TCA, dried, and counted in a scintillation counter. Parallel plates were used to determine cell protein per well, as described below.
Treatment
of cells
Acrylamide (electrophoresis grade, 161-0100, Bio-Rad, Richmond, CA) and ACTH (Sigma Chemical Co., St. Louis, MO) were added to media from aqueous solutions; taxol (a gift of Dr. Matthew Suffness, NCI) was added from a solution in dimethylsulfoxide. Appropriate solvent controls were performed. After treatment, cells were routinely scored for cell rounding before harvesting and assay of steroid and CAMP production. All results represent at least duplicate determinations on two separate dishes (wells) of cells; in most cases three separate dishes were used. All experiments were repeated at least twice. Cell mass was estimated by protein determination. After removal of growth medium, the cell monolayer was washed twice with PBS and extracted with 1 ml 0.1 N NaOH, and protein was determined with the bicinchoninic acid method (Pierce Chemical Co., Rockford, IL).
Preparation
of MT protein
Rat brain MT protein was purified by two cycles of temperaturedependent polymerization and depolymerization in Mes assembly buffer [O.l M 2(N-morpholino)ethanesulfonic acid (Mes), 1 rnM MgCl*, and 1 mM EGTA, pH 6.91 with 1 rnM GTP (28). Purified protein was stored in liquid nitrogen, as previously described (29). Polymerization of MT
Morphology
of the Y-l adrenal
cell culture
The typical Y-l cell grows in a flat epithelial-like monolayer (Fig. 1A). This morphology and the time course of rounding upon exposure to acrylamide are shown in Fig. 1 (A-E). The refractile rounded morphology typical of stimulated cells is shown by white arrows in Fig. lD, and the percent rounding in each panel is given in the legend. After treatment with acrylamide for 1.5 h (Fig. lB), little change was observed in the rounding of the cells; after 2 h, partial rounding was seen (Fig. lC), and after 2.5 h of treatment, about half of the cells were rounded (Fig. 1D). By 3.5 h of treatment, greater than 90% of the cells were rounded (Fig. 1E). This distinct morphological change of near-uniform rounding occurring between 2-3 h of treatment with acrylamide differs from that observed after 3 h of ACTH treatment, in which case patchy nonuniform rounding was seen (Fig. 1F). The degree of cell rounding in response to ACTH varies with different clones of Y-l cells. Effect of a&amide
on cell function
and growth
Treatment of Y-l cells with ACTH inhibited cell growth (31). Overnight treatment with acrylamide reduced thymidine incorporation beyond that observed with ACTH alone, indicating that acrylamide also inhibits cell growth (Table 1). Protein synthesis, however, was not inhibited by ACTH or acrylamide, since [3H]leucine incorporation during overnight incubation with either agent was the same as that in the controls. Effect of acrylamide
on steroid and CAMP production
Steroidogenesis induced by ACTH shows a rapid early increase, followed by a slower continuing rate of increase. This has been observed previously (10). By contrast, acrylamide enhanced steroidogenesis in a different temporal pattern, with a slow early phase, followed by a more rapid phase starting after about 5 h of exposure (Fig. 2), shortly after the time corresponding to the time of maximal rounding of the cells (Fig. 1E) and similar to the time of delayed stimulation seen after colchicine treatment (10). Like colchitine, the effects of acrylamide on steroid production occur at a step preceding pregnenolone formation; Table 2 illustrates that addition of acrylamide to cells treated with pregnenolone fails to produce a further increase in steroid production above that observed with the addition of pregnenolone alone. A striking difference between ACTH and acrylamide stimulations is depicted in Fig. 3. The rapid burst in CAMP production and subsequent maintenance of elevated levels after ACTH treatment were not seen after exposure of Y-l cells to 5 mM acrylamide. In fact, acrylamide values did not differ significantly from control values over a period of 24 h.
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 12 November 2015. at 23:16 For personal use only. No other uses without permission. . All rights reserved.
IF AND
STEROIDOGENESIS
203
FIG. 1. Time course of cell rounding in the presence of acrylamide. Phase micrographs of treated and untreated Y-l adrenal cortical cells are shown. Untreated cells are flat, and the typical rounded morphology seen after stimulation is indicated by the three white arrows in D; note the highly refractile character of these rounded cells. For each time point in the figure, a total of 400-500 cells from 3-4 fields were scored for rounding. The percent rounding (*SD) is given for each panel. A, Control: normal flattened appearance of Y-l cells. Rounding = 1.6 + 1%. B-E, Acrylamide treatment (5 mM). B, Incubated for 1.5 h, showing few morphological changes. Rounding = 6.3 t 2.2%. C, Two-hour incubation, showing the appearance of rounded cells. Rounding = 27 + 8%. D, Two and a half-hour incubation, showing significant rounding of cells. Rounding = 43.2 k 7%. E, Three and a half-hour incubation, showing rounding of essentially all cells. Rounding = essentially total. F, ACTH treatment (0.125 NM), 3-h incubation, showing typical rounding of scattered cells. The bar between B and E = 50 pm.
It is apparent, therefore, that the mechanismsof stimulation of steroidogenesisby ACTH and acrylamide differ. Moreover, acrylamide-induced steroidogenesis is unlikely to be basedon a CAMP-dependent mechanism. Effect of acrylamide on MT Previous studies have shown that treatment of PtK-1 cells with acrylamide disrupted IF while leaving MT intact (25, 26). The lack of a direct effect of acrylamide on MT is confirmed by in vitro studies on the polymerization of MT protein, as shown in Fig. 4. MT assembly, as measured by optical density at 350 nm at 37 C in the presence of GTP was not affected by very high (50 mM) concentrations of acrylamide (arrow in Fig. 4A). Moreover, MT in the presence of acrylamide exhibited temperature-sensitive depolymerization like untreated MT. Furthermore, as shown in Fig. 4B, MT formed in the presence of 50 mM acrylamide, polymerized normally, and repolymerized normally. There is, thus, every reason to believe that the effects seen with acrylamide in Y-l cells are not, primarily, exerted on MT. This is also shown by a comparison of acrylamide and colchicine in intact Y-l adrenal cells. The in viva effects of acrylamide on MT were tested by
comparing treatment of cellswith colchicine, the potent antiMT agent, in the presenceand absenceof acrylamide. Figure 5 illustrates that the effects of colchicine and acrylamide were additive, as were the effects of ACTH and acrylamide when used together in the Y-l cell. As shown by the dashed lines in Fig. 5, these additivities were essentially identical to those calculated from effects produced when the agentswere incubated separately. Finally, we found (Fig. 5C) that forskolin, which is frequently used to stimulate CAMP-dependent reactions without intervention of membrane receptors, promoted vigorous steroid production, and this stimultation was additive with acrylamide enhancement of steroidogenesis. Because simple additivity was noted, we have made the presumptive conclusion that these agents can act at steps different from those of acrylamide; the relative importance of the different stepsmay depend on the mode of stimulation of steroidogenesis.The results in Table 3 illustrate that pretreatment of cells with taxol results in decreasedacrylamidestimulated production of steroids. This is in keeping with previous observations (15) that pretreatment of Y-l cells with taxol, which specifically promotes MT polymerization, causes a reduction in basal and ACTH-stimulated steroid prodution. This suggeststhat despite a considerable independence of the stimulatory mechanisms mentioned above, these same
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 12 November 2015. at 23:16 For personal use only. No other uses without permission. . All rights reserved.
IF AND
204 TABLE 1. Effect incorporation
of acrylamide
on leucine
and thymidine
Acid-insoluble
radioactivity
Leucine kdpm
Control ACTH (0.1 PM) Acrvlamide (4.5
(12.6)
1.0 0.92
mM)
TABLE 2. Effect pregnenolone
Thymidine
AVW3~e” zk 0.09
0.92 -c 0.07
Endo. Voll31.
STEROIDOGENESIS
kdpm
AWK3ge”
(3.54)
1.0 0.89 k 0.02 0.58 f 0.12
Cells were grown to about half-confluency, and medium was removed and replaced with fresh medium containing [“Hlleucine or [“Hlthymidine plus the indicated agents. Leucine uptake was performed in leucine-free DME using cells grown for l-2 days in DME. Thymidine uptake was performed in DME or DME-F-12 using cells grown for l2 days in the appropriate medium. Thymidine uptake results were the same in both media. [“H]Leucine (135 Ci/mmol) was added to give a final concentration of 2.2 mM in the medium; [“Hlthymidine (6.76 Ci/ Fmol) was added to give a final concentration of 0.5 fiM in the medium. Agents were added to give the indicated concentrations, and cells were exposed to medium with agent and tracer at the same time and incubated overnight (-16 h). After incubation, the cells were washed, and TCA-insoluble radioactivity was determined as described in Muterials and Methods. Results presented are averages f SE from three separate experiments of two wells each. The disintegration per min/ mg protein is given for the control, and other results are normalized to the control. ’ Average value normalized to the control.
Basal Pregnenolone Prennenolone
of acrylamide
on steroid
production
1992 No 1
from
Steroid
produced
(udmr
urotein)
0.2 2.76 + 0.5 2.68 + 0.6
+ acrvlamide
Cells were grown to half-confluency. Media were removed and replaced with fresh media supplemented with 5 mM acrylamide alone or media plus 5 mM acrylamide and pregnenolone (10 fig/ml). Media were removed and assayed as described in Materials and Methods. Duplicate determinations were made on duplicate wells.
,
a i 5 0
200 I ys, 5
10
15 20 TIME, h
Acryl. Control
25
FIG. 3. Time course of CAMP production. Cells and conditions are described in Fig. 2. After removal of the medium, CAMP was determined by RIA, as described in Materials and Methods. As in Fig. 2, each point represents duplicate determinations on triplicate wells of cells. Total CAMP is plotted as a function of time. Representative SEs are shown. Acryl., Acrylamide.
TIME,
h
FIG. 2. Time course of steroid production. Cells were grown to about half-confluency, and at time zero, media were removed and replaced with fresh media containing 0.1 mM Ro 20-1724 supplemented with acrylamide (Acryl, 5 mM), ACTH (0.125 PM) or no addition, as indicated. At the indicated times, media were removed, and steroid production was assayed. Each point represents duplicate determinations on triplicate wells of cells. Total steroid is plotted as a function of time. Representative SES are shown.
filamentous other.
elements
are not entirely
independent
of each
Discussion The cytoplasm is a dense complex network of interlaced fibers (MT, IF, and MF) in,whose interstices reside the various organelles that may be free floating, near one or another fiber, or actually attached to one or more of these. The cholesterol ester-rich lipid droplet of adrenal cortical cells is one of these organelles. The fiber network might either
promote movement of lipid droplets along one of the fibers with the aid of a motor, such as kinesin, or it might act as a barrier to migration of droplets toward the mitochondrion. Colchicine, cytochalasin-D, and acrylamide are thought to cause their intracellular effects by functional disruption or redistribution of the cytoskeletal fibers: MT, MF, and IF, respectively. Usually the particular fiber-mediated process is inhibited, which is then interpreted as a requirement for that fiber to translocate a particular organelle. It was, therefore, a surprise that secretion of steroids (which are not stored in granules in the adrenal cell) was enhanced upon incubation of Y-l cells with colchicine or its congeners in the absence of ACTH (10, 11, 14). Since the cholesterol ester precursor of steroids is packaged in lipid droplets, the possibility existed that MT-disrupting agents might act at this earlier stage in steroid secretion, and we proposed that the MT was a barrier to the access of cholesterol ester droplets to the mitochondrion (10). Similar explanations, but implicating MF, might apply to the enhancement of basal steroidogenesis by cytochalasin-B or -D (5, 9, 32). The results of the present paper show that manipulation of IF networks by acrylamide enhances basal steroidogenesis and causes cell rounding, but does not change cellular CAMP concentrations. These phenomena are similar to those produced by disruption of MT, but acrylamide does not cause gross changes in MT or actin filaments. Thus, the effects of
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 12 November 2015. at 23:16 For personal use only. No other uses without permission. . All rights reserved.
IF AND
STEROIDOGENESIS
205
IA
I
6 F B E zi 8 E'
250-
400
200-
T.
L 150-
300
loo-
200
2 8E P i
100
50-
U-J
T 1
2
3
1
2
3
5. Additivity of acrylamide effect with that of ACTH, colchicine, and forskolin. Cells were grown and treated, and steroid was assayed as described in Fig. 2. Plates were incubated overnight (-16 h) at 37 C. In A and B, 1 represents no addition, 2 represents 0.1 FM ACTH, and 3 represents 5 pM colchicine. A, The results of these treatments alone. B, The results of these treatments in the presence of 2.5 mM acrylamide, added at the same time as the other agents. C, The results of treatment with 1 pM forskolin (no. 1) and 1 pM forskolin plus 2.5 mM acrylamide added at the same time (no. 2). The ordinate scale for A and B is on the left, and that for C is on the right. The dotted lines in B and C represent the results expected if the effects of acrylamide and the treatments are additive: B, 2, ACTH plus acrylamide; B, 3, colchicine plus acrylamide; C, 2, forskolin plus acrylamide. All results are expressed as the percent increase over the control steroid production (0.524 -C 0.089 pg/mg protein) found in the absence of any added agent (A, 1). Data represent the mean + SE of duplicate determinations on a total of 11 replicates from 4 separate experiments.
FIG.
TIME
(seconds)
FIG. 4. Lack of effect of acrylamide on in uitro polymerization of MT. A, Rat brain MT protein (1.5 mg/ml) in Mes assembly buffer plus 1 mM GTP was polymerized at 37 C. At the time indicated by the arrow, acrylamide (in Mes assembly buffer) was added to 50 mM. The optical density was monitored for several minutes further, after which the cuvet was placed on ice for 5 min (broken line) to measure cold-induced depolymerization. B, Rat brain MT protein (1.2 mg/ml) in Mes assembly buffer plus 1 mM GTP was polymerized at 37 C in the absence (upper curue) or presence (lower curue) of 50 mM acrylamide. After polymerization plateaued, the cuvets were transferred to ice for 5 min to depolymerize the protein (broken lines). After warming, the repolymerization was monitored, followed by another depolymerization on ice (second set of broken lines).
the three fiber types show a degree of independence from each other, and this is confirmed by the additivity of colchitine- and acrylamide-mediated steroidogenesis. Several observations point to more complicated mechanisms for stimulation of steroidogenesis by disruption of cytoskeletal elements, and two additional mechanisms come to mind: 1) disruption of tubulin or vimentin coats surrounding cholesterol droplets and 2) redistribution of mitochondria. The coat hypothesis enjoys some experimental support. While coats around lipid droplets had been identified in the past, the nature of the coating material was not. Clark and Shay (33, 34) showed that cholesterol droplets were coated with tubulin in an unspecified polymeric form that could be stained with antibodies. ACTH and colchicine removed this layer (33, 34). More recently, it has been shown that neutral lipid droplets in adipocytes were encased in IF networks (18). In Y-l cells it has been shown that approximately one third of the cholesterol droplets is firmly attached to vimentin fibers (22). Another third is present as (apparently) free droplets, while the last third of the droplets is attached less firmly than to IF, presumably to other members of the cytoskeleton, i.e. MT and MF. It is conceivable that these differently attached cholesterol droplets may well respond
TABLE
3. Taxol steroidogenesis
inhibition
Treatment Control Acrylamide Taxol + acrylamide
of acrylamide-stimulated
Steroid
(&mg
mot&n)
Steroid (normalized)
1.24 (kO.07) 2.83 + 0.11
2.28 f 0.09
2.10 & 0.04
1.69 + 0.03
1
Cells were grown as described in Materials and Methods. Agents were added from concentrated stocks to fresh medium to the desired final concentration, 5 mM for acrylamide and 0.01 mM for taxol. The cells exposed to taxol plus acrylamide were exposed to taxol alone during a l-h preincubation before being exposed to taxol and acrylamide together for the remainder of the exposure period. After overnight incubation, steroids and proteins were determined, as described in Materials and Methods, on duplicate samples from duplicate wells; the values reported are the mean and SD of these four values. Similar results were obtained in replicate experiments.
differently to various stimuli and may account for partial responses and for additivity. The affinity of vimentin for lipids is shown by binding of a variety of phospholipid vesicles and cholesterol to isolated vimentin filaments (19, 20); this occurs by electrostatic interactions of arginine residues of the N-terminal vimentin domain with the phosphates of acidic phospholipid and hydrophobic interactions between nonpolar amino acids of the N-terminus and uncharged lipid moieties, including cholesterol (35). The response of the vimentin coat to acrylamide has not yet been studied, but the findings presented here would be consistent with the
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 12 November 2015. at 23:16 For personal use only. No other uses without permission. . All rights reserved.
IF AND
206
STEROIDOGENESIS
uncoating hypothesis applied to vimentin. Finally, it has often been noted that mitochondria may be near or attached to MT and/or IF (36-38). Disruption of MT or IF leads to a marked redistribution of mitochondria (38, 39). Other organelles may also be attached (36), and such associations are critical for intracellular transport promoted by motors such as kinesin or cytoplasmic dynein. In principle, interruption of these associations could shorten the distance between droplet and mitochondrion. We suggest that both uncoating of lipid droplets and mitochondrial or droplet movement may together markedly facilitate access of cholesterol to the side-chain cleavage site in the mitochondrion. It is improbable, however, that the above events can be unambiguously ascribed to action on a single type of cytoskeletal fiber, because there is clear evidence that these three classes of intracellular fibers can interact with each other, e.g. through MT-associated proteins, synapsin, etc., that can bind to MT, actin filaments, and/or IF, or through the binding of two different fibers to the same organelle (38, 40, 41). It is important, therefore, to ascertain whether the disruption of one class of filaments exerts its steroidogenic stimulation indirectly through an effect on another filament. Generally, addition of cytochalasin to cells does not grossly change MT distribution, and conversely, colcemid has little effect on stress fibers or actin networks (23). On the other hand, disruption of MT led to a marked rearrangement of the IF network (18, 42). Thus, colchicine might stimulate through a loss of MT restraints on IF distribution. The fact that taxol inhibits both colchicine-stimulated (12, 15) and acrylamidestimulated steroidogenesis supports such a mechanism, provided that taxol has no direct effect on IF. Normal steroidogenesis would then depend in part on the dynamic instability of MT (43), which could effect temporary changes in the vimentin filaments. Whether such instabilities also occur in the tubulin that coats cholesterol droplets is not known at present. Similarly, factors that remove the vimentin coat have yet to be directly demonstrated. It is clear that the proteins of the major classes of cytoplasmic filaments regulate steroidogenesis in Y-l adrenal tumor cells. Whether they do so as the usually identified fibrous polymers or as coats around the lipid droplets (in the case of tubulin and vimentin) remains to be resolved. Because the filaments interact with each other through associated proteins or organelles, it is currently not possible to say whether interruption of any one of them is a direct or indirect stimulus to steroidogenesis. It is not clear to what extent the ACTH/cAMP pathway overlaps the cytoskeletal pathway, except that some portions of it are different since they are additive and only one uses CAMP. References 1. Zor U 1983 Role of cytoskeletal organization in the regulation of adenylate cyclase-cyclic adenosine monophosphate by hormones. Endocr Rev 4:1-21 2. Hall PF 1984 The role of the cytoskeleton in hormone action. Can J Biochem Cell Biol 62:653-665 3. Schimmer BP 1979 Adrenocortical Y-I cells. Methods Enzymol 58:570-574 4. Betz G, Hall PF 1987 Steroidogenesis in adrenal tumor cells: influ-
Endo. 1992 Vol 131 *No 1
ence of cell shape. Endocrinology 120:2547-2554 5. Cortese F, Wolff I 1978 Cvtochalasin-stimulated steroidoeenesis from high’density fipoprotei&. J Cell Biol 77:507-516 ” 6. Ichikawa Y 1989 Composition of culture media for steroid hormone secretion by murine adrenal tumor cells, Y-l clone. Acta Med Okayama 43:97-103 7. Mrotek JJ, Hall PF 1975 The influence of cytochalasin B on the response of adrenal tumor cells to ACTH and cyclic AMP. Biochem Biophys Res Commun 64:891-896 8. Hall PF, Charpponnier C, Nakamura M, Gabbiani G 1979 The role of microfilaments in the response of adrenal tumor cells to adrenocorticotropic hormone. J Biol Chem 254:9080-9084 9. Rainey WE, Shay JW, Mason JI 1984 The effect of cytochalasin D on steroid production and stress fiber organization in cultured bovine adrenocortical cells. Mol Cell Endocrinol 35:189-197 10. Temple R, Wolff J 1973 Stimulation of steroid secretion by antimicrotubular agents. J Biol Chem 248:2691-2698 11. Ray P, Strott CA 1978 Stimulation of steroid synthesis by normal rat adrenocrotical cells in response to antimicrotubular agents. Endocrinology 103:1281-1288 12. Rainey WE, Kramer RE, Mason JI, Shay JW 1985 The effects of taxol, a microtubule-stabilizing drug, on steroidogenic cells. J Cell Physiol 123:17-24 13. Carnegie JA, Dardick I, Tsang BK 1987 Microtubules and the gonadotropic regulation of granulosa cell steroidogenesis. Endocrinology 120:819-828 14. Kotani S, Murofushi H, Sakai H 1988 Dual effect of antimitotic drugs on steroid secretion in mouse adrenocortical Y-l tumor cells. Cell Struct Funct 13:171-177 15. Sackett DL, Wolff J 1986 Cyclic AMP-independent stimulation of steroidogenesis in Y-l adrenal tumor cells by antimitotic agents. Biochim Biophys Acta 888: 163-l 70 16. Stewart M 1990 Intermediate filaments: structure, assembly and molecular interactions. Curr Opin Cell Biol 2:91-100 17. Franke WW, Schmid E, Osborn M, Weber K 1978 Different intermediate-sized filaments distinguished by immunofluorescence microscopy. Proc Nat1 Acad Sci USA 75:5034-5038 18. Franke WW, Hergt M, Grund C 1987 Rearrangement of the vimentin cytoskeleton during adipose conversion: formation of an intermediate filament cage around lipid globules. Cell 49:131-141 19. Traub P, Perides G, Scherbarth A, Traub U 1985 Tenacious binding of lipids to vimentin during its isolation and purification from Ehrlich ascites tumor cells. FEBS Lett 193:217-221 20. Traub P, Perides G, Kuhn S, Scherbarth A 1987 Efficient interaction of nonpolar lipids with intermediate filaments of the vimentin type. Eur J Cell Biol 43:55-64 21. Asch HL, Mayhew E, Lazo RO, Asch BB 1990 Lipids noncovalently associated with keratins and other cytoskeletal proteins of mouse mammary epithelial cells in primary culture. Biochim Biophys Acta 1034:303-308 22. Almahbobi G, Hall PF 1990 The role of intermediate filaments in adrenal steroidogenesis. J Cell Sci 97:679-687 23. Durham HD, Pena S, Carpenter S 1983 The neurotoxins 2,5 hexanedione and acrylamide promote aggregation of intermediate filaments in cultured fibroblasts. Muscle Nerve 6:631-637 24. Eckert BS 1985 Alteration of intermediate filament distribution in PtKl cells by acrylamide. Eur J Cell Biol 37:169-174 25. Eckert BS 1986 Alteration of the- distribution of intermediate filaments in PtKl cells by acrylamide. II. Effect on the organization of cytoplasmic organelles. Cell Motil Cytoskel 6:15-24 26. Eckert BS, Yeagle PL 1988 Acrylamide treatment of PtKl cells causes dephosphorylation of keratin polypeptides. Cell Motil Cytoskel 11:24-30 27. Ramirez LC, Millot C, Maume BF 1982 Sample purification using a C(18)-bonded reversed phase cartridge for the quantitative analysis of corticosteriods in adrenal cell cultures by high-performance liquid chromatography or gas chromatography/mass spectrometry. J Chromatogr 229:267-281 28. Shelanski ML, Gaskin F, Cantor CR 1973 Microtubule assembly in the absence of added nucleotides. Proc Nat1 Acad Sci USA 70:765-768 29. Sackett DL, Lippoldt R 1991 Thermodynamics of reversible mon-
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 12 November 2015. at 23:16 For personal use only. No other uses without permission. . All rights reserved.
IF AND omer-dimer
association
30. Bhattacharyya
of tubulin.
Biochemistry
30:3511-3517 hybrid
8, Sackett DL, Wolff J 1985 Tubulin,
and tubulin
STEROIDOGENESIS 37. Traub P 1985 Intermediate
Filaments:
A Review.
Springer-Verlag,
Heidelberg
dimers,
S. I Biol Chem 260:10208-10216 In: Sato G (ed) Functionally Differentiated Cell Lines. Liss, New York, p 61 32. Mattson P, Kowal J 1982 Effects of cytochalasin B on unstimulated and ACTH-stimulated adrenocortical tumor cells in vitvo. Endocrinology 111:1632-1647 33. Clark MA, Shay JW 1979 The response of whole and enucleated adrenocortical tumor cells (Y-l cells) to ACTH treatment. Scanning Electron Microsc 3:527-535 34. Clark MA, Shay JW 1981 The role of tubulin in the steroidogenic response of murine adrenal and rat Leydig cells. Endocrinology 109:2261-2263 35. Perides G, Harter C, Traub P 1987 Electrostatic and hydrophobic interactions of the intermediate filament protein vimentin and its amino terminus with lipid bilayers. J Biol Chem 262:13742-13749 36. Rousset BAT 1988 Functions of cytoskeletal elements in organelle distribution. In: Rousset BAF (ed) Structures and Functions of the Cytoskeleton. Colloque INSERM. Libby, London, vol 171:275-400
207
38. Bereiter-Hahn
31. Schimmer BP i9Sl
J 1990 Behavior of mitochondria in the living cell. 122:1-63 Summerhayes IC, Wong D, Chen LB 1983 Effect of microtubules and intermediate filaments on mitochondrial distribution. J Cell Sci 61:87-105 Griffith L, Pollard TD 1978 Evidence for actin filament-microtubule interaction mediated by microtubule-associated protein. J Cell Biol 78:958-965 Sattilaro RF, Dentler WL, LeCluyse EL 1981 Microtubule-associated proteins (MAPS) and the organization of actin filaments in vitro. J Cell Biol 90:467-473 lnt Rev Cytol
39. 40.
41.
42. Bershadsky AD, Ivanoya OY, Lyass LA, Pletyushkina OY, Vasiliev JM, Gelfand IM 1990 Cytoskeletal reorganizations responsible
43.
for the phorbol ester-induced formation of cytoplasmic processes: possible involvement of intermediate filaments. Proc Nat1 Acad Sci USA 87:1884-1888 Mitchison T, Kirschner M 1984 Dynamic instability of microtubule growth. Nature 312:237-242
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 12 November 2015. at 23:16 For personal use only. No other uses without permission. . All rights reserved.
