Inhibition of 3-Hydroxy-3-methylglutaryl Coenzyme A Reductase Activity in Hepatoma Tissue Culture Cells by Pure Cholesterol and Several Cholesterol Derivatives EVIDENCE DISTINCT activity of HMG-CoA’ reductase, the enzyme which catalyzes

Pure cholesterol associated in complexes with lipoproteins (whole serum and human low density lipoproteins) or esterified with succinic acid (cholesteryl succinate) and bound to albumin effectively suppresses 3-hydroxy-3-methylglutaryl coenzyme A (HMG-CoA) reductase activity in hepatoma tissue culture (HTC) cells grown in lipoprotein-poor serum medium during (4-hour) incubation periods. Simultaneous measurements of the kinetics of uptake of radioactive unesterified cholesterol of whole serum and cholesteryl succinate, their conversion to lipid products, and the decay in enzyme activity, suggest that the cholesterol-induced suppression is mediated by the sterol itself rather than by inhibitory lipid products derived from its metabolism. Several cholesterol derivatives such as cholestenone, and and 25-hydroxycholes-terol

used are: HMG-CoA, 3.hydroxy-3-methylglutacholest-4-ene-3-one. Synthesis in HTC Cells but has no effect on the production of mevalonic acid in any of the primary or transplanted hepatomas thus far studied (6-9). In contrast, previous studies from this laboratory (10-13) have demonstrated that sterol synthesis and reductase activity in cultured hepatoma cells are decreased severalfold by specific serum cholesterol fractions.
Kandutsch and Chen (14,15) have reported that high concentrations of highly purified cholesterol added to cultures of mouse fetal liver and fibroblast cells fail to cause significant inhibition of sterol synthesis and reductase activity, whereas impure choleste+ol and cholesterol derivatives oxygenated at positions C,, Cpo, CpB, or C,, are strongly inhibitory.
Substantial differences between the inhibitory potency of cholesterol and oxygenated cholesterol derivatives on the rate of sterol synthesis in cultured human fibroblast cells have also been reported ( 16).
Kandutsch and Chen (14), based on their findings, and on the fact that several of the inhibitory oxygenated cholesterol derivatives can be produced by the in vim oxidation of cholesterol, have questioned the role of cholesterol per se as a modulator of sterol synthesis in cultured cells.
To learn more about the structural requirements necessary for regulation of sterol synthesis in malignant hepatic cells, and to determine whether this regulation can be mediated by cholesterol per se, we have compared the effects of pure low density lipoprotein, pure cholesterol.albumin complex, and several cholesterol derivatives on the rate of 3@-hydroxysterol production and loss of reductase activity in HTC cells. The relationship between the rate of uptake of several of these sterogenic inhibitors by HTC cells, their metabolism to products, and the inhibition of reductase activity has been examined.
In addition, the mechanism by which some of the inhibitors suppress HMG-CoA reductase activity in HTC cells has also been investigated.

Biochemicals and Reagents-DL-3-Hydroxy-
[3-L'C]methylglutaryl coenzyme A, DL-[5-'Hlmevalonic acid (dibenzylethylenediamine salt) [l-Y k&ate, [2-'Clacetate, and DL-[2-"Clmevalonate were purchased from New England Nuclear. Glucose 6-phosphate, glucose-6.phosphate dehydrogenase, NADP, and NADPH were obtained from Boehringer Mannheim Corp. Cholesterol was obtained from Calbiothem, and the other steroids from Steraloids. Fetal calf serum, calf serum, and Swim's 77 media powder were obtained from Grand Island Biological Co. All other reagents used were of the highest analytical Purification-Steroids available in quantities greater than 100 mg were routinely purified by repeated crystallization from 95% ethanol. The first crop of crystals was used, and the mother liquor was discarded.
However, thin layer chromatography on 0.25 mm and 5 mm thick commercial silica gel plates (Brinkman Inc., New York) was used to characterize and purify most of the steroids studied. The primary solvent system used was n-heptane:ethyl acetate (l:l, v/v); this system resolved cholesterol from its characteristic autooxidation products, which were also well separated from one another (2'2). If additional purification was needed, either diethyl ether (23), benzene:acetone (4:1, v/v), or n-heptane:acetone (l:l, v/v) were used as solvent systems. Steroids were visualized by minimal exposure to iodine vapors or sprayed with a l:l, 50% (v/v) sulfuric acid:water solution and incubated at 110" for 15 min (22). When a compound was to be recovered from the chromatographic plate, its location was determined from the migration of reference compounds. The sample band was scraped off and put into a glass wool-plugged column (8 x 125 mm) with a 50-ml solvent reservoir.
The steroid samples were eluted with two 4.ml additions of a chloroform:methanol (4:1, v/v) solution. All eluted samples were evaporated in uacuo, without heat, and the lipid residue was dissolved in 1 to 5 ml of 95% ethanol. This solution was stored at -20" until use.
The sulfuric acid spray detected 0.2 to 0.5 pg (22)  twice washed 900 x g pellet, whereas the total activity recovered from the whole homogenate is distributed in a bimodal pattern; the pellet contains 40% of the activity and the supernatant 60%. This pattern of enzyme activity distribution is a consistent finding. In addition, it is independent of the cell breakage procedure (sonication, Dounce homogenization, or nitrogen decavitation) and homogenization media (hypo-or isotonic salt, or isotonic sucrose) employed.
in one subcellular fraction, the possibility of using whole cell When a more complete fractionation of HTC cell homogenates is done (data not shown), the total activity associated with the 900 i g supernatant fraction in Experiment B sediments in the 100,000 x g pellet so that the highest specific activity of reductase is associated with the 900 x g and 100,000 x g pellets. Since these pellets have not been characterized electron microscopically or enzymatically, it is not clear whether the apparent bimodal distribution of HMG-CoA reductase activity is native to HTC cells, or is reflective of our inability to isolate pure subcellular fractions from them.
Reductase Activity in HTC Cell Homogenates--Since reductase activity, using our experimental procedure, is not localized homogenates of HTC cells for enzyme assays was explored. The enzymatic properties (K, for DL-HMG-CoA and NADPH, pH optima, and requirement for thiol reagent) of reductase from HTC cell homogenates are essentially the same (data not shown) as those reported previously for the microsomal activity of HTC cells (13) and for the rat liver enzyme (36-38).
As shown in Fig. 1, the reduction of DL-HMG-CoA to mevalonic acid by HTC cell homogenates proceeds linearly with time and is dependent on protein concentration over a broad range. The mevalonate formed is not converted to additional products under the conditions of our assay (data not presented). Table II demonstrates the reproducibility of reductase measurements using homogenates of HTC cells grown in media containing whole or lipoprotein-poor sera. As reported previously by this laboratory (13), the activity of reductase is increased in HTC cells transferred from a medium which contains whole serum to a medium which contains lipoproteinpoor serum. The increase in enzyme activity in HTC cell homogenates is approximately 7-fold, and the actual rate (27 nmol of mevalonate/hour/mg of protein) is comparable to that reported for the rat liver enzyme at the peak of its circadian rhythm (39-41). In addition, reductase activity decreases 45 to 50% in HTC cells which are grown in lipoprotein-poor serum medium, transferred to medium which contains intact serum, and incubated for 4 hours. 60   suppress HMG-CoA reductase activity and 3@-hydroxysterol production in several cultured cell systems-except at high concentrations and after prolonged incubation periods-raises the possibility that the seruminduced inhibition of reductase activity in HTC cells may be caused by autooxidation products of cholesterol rather than by the sterol itself. To determine whether cholesterol per se can suppress the activity of reductase, HTC cells were incubated with varying concentrations of whole serum and human LDL (both were free of cholesterol autooxidation products; see "Methods"), and with pure unesterified cholesterol and cholesteryl succinate bound to albumin.
The results of these experiments are shown in Fig. 2.
In agreement with the results obtained with cultured mouse cells (14), pure, unesterified cholesterol (bound to albumin) inhibits reductase activity minimally (20% in 4 hours), whereas impure cholesterol-albumin causes an 80% decrease in activity in the same time period, Cholesterol presented in the form of whole serum inhibits HMG-CoA reductase activity 60% in 4 hours, and is therefore significantly more effective than pure cholesterol added as an albumin emulsion.
Since high density lipoproteins are ineffective in inhibiting sterol synthesis in HTC cells (11,12), and they represent 40 to 50% of the total cholesterol content, of our calfifetal calf serum mix, ' a more accurate assessment of the inhibitory potency of serum cholesterol is obtained by testing the effects of the LDL fraction (density 1.019 to 1.063 g/ml), Although the maximum inhibition (60 to 65%) produced by whole serum and LDL are the same, the concentrations of unesterified cholesterol required to produce this effect are different (4 to 6 lg/ml for LDL and 35 to 40 pg/ml for whole serum).
The inhibitory potency of pure cholesterol-albumin can be substantially increased (from 20 to 95% maximum inhibition) when the sterol nucleus is esterified as a monoester of succinic acid (Fig. 2). Similar results are obtained with cholesteryl phthalate.
Effects of Cholesterol Derivatives on Reductase Actiuity- Fig. 3 shows the suppression of reductase activity in HTC cells grown in lipoprotein-poor serum and incubated for 4 hours with varying concentrations of non-lipoprotein-bound steroids. As indicated, specific structural modifications of the cholesterol molecule can greatly enhance its inhibitory potency. Of the cholesterol analogs tested, those oxygenated at position 25 (25hydroxycholesterol) and at position 7 (7~ hydroxycholesterol and 7-ketocholesterol) are the most inhibitory (panel C), while the 3-keto compounds (cholest-5-ene-3one, cholestenone, cholestan-3-one (panel B)) are significantly more effective in suppressing HMG-CoA reductase activity than their 3-hydroxy analogs (cholesterol, allocholesterol, and cholestanol (panel A)). In addition, the A' compounds (allocholesterol and cholestenone) are slightly more inhibitory than the A" isomers (cholesterol and cholest-5-ene-3-one), and the 3-hydroxy group is somewhat more effective in the alpha position (epicholesterol) compared to the fi position (cholesterol). Similar, but not identical, results have been reported for mouse fibroblast and fetal liver cells (14, 15) and human fibroblast cells (16). The effects of the various steroids on the incorporation of (2."Clacetate into BP-hydroxysterol production (closed symbols in Fig. 3)  control activity was 27 nmol/hour/mg of protein.
In Table III, the relative inhibitory potencies of the steroids tested are summarized in terms of the maximum inhibition achieved in 4 hours and the concentration of steroid required to obtain 40 to 50% of the maximum inhibition.
As indicated, LDL-cholesterol is the most potent of the cholesterol preparations (whole serum, cholesterol-albumin, cholesteryl succinatealbumin, and human LDL) tested. Relative to it, only the multifunctional steroids (25-and 7a-hydroxycholesterol and 7-ketocholesterol) and cholestenone are significantly more inhibitory.
Cholesteryl succinate and cholest-4,6-diene-3-one are slightly less effective than LDL-cholesterol, and the remaining steroids shown in Table III, including unesterified cholesterol bound to albumin, are significantly less inhibitory. The steroid-induced inhibition of cholesterol synthesis in HTC cells appears to be specific, since (Table IV) steroid concentrations which result in a 50% decrease in 38-hydroxysterol production from acetate have no significant effect on protein and fatty acid synthesis. Consistent with the fact that the rate of cholesterol synthesis in HTC cells is determined by the rate of reduction of HMG-CoA mevalonic acid, 3&hydroxysterol product,ion from mevalonate is also not affected by the inhibitory steroids. High concentrations of the inhibitory steroids added directly to HTC cell homogenates do not decrease the activity of reductase (data not shown).
Lack of Significant Product Formation by Inhibitory Steroids-Since many of the inhibitors in Table III may be  products of cholesterol metabolism, or may converge on a common metabolic effector which in turn suppresses cholesterogenesis, the relationship between the rate of uptake of several inhibitors, their metabolism to lipid products, and the rate of decrease in reductase activity were examined.
The results are shown in Fig. 4 Table V. Cholesterol ester synthesis (or that of any other lipid class) is not enhanced in HTC cells exposed for 5 hours to intact serum and several cholesterol derivatives. A more detailed study by Wiegand and Wood (42) also shows that HTC cells have a minimal capacity to synthesize cholesterol esters. Additional studies with prelabeled cells ([2-"Clacetate for 24 hours) which were later incubated with inhibitory steroids for 5 hours did not show any change in the percentage of radioactive cholesterol ester found in the total cellular lipid fraction. Radioactive sterol esters were not found in the incubation medium. Therefore, cholesterol esterification and the suppression of reductase activity in HTC cells are not as closely related as they appear to be in cultured human fibroblast cells. Effect of Steroid Uptake on Cholesterol Efflux by HTC Cells-To determine whether the inhibition of reductase activity by unesterified whole serum cholesterol and several steroids is related to their capacity to exchange with cellular cholesterol, the kinetics of their uptake was compared with the loss of cellular cholesterol.
HTC cells were prelabeled with tracer levels of "C or SH cholesterol and incubated for varying time periods with the indicated labeled steroids (Fig. 6).
Panel A of Fig. 6 shows the decrease in cellular radioactivity (cholesterol release) from prelabeled cells incubated with lipoprotein-poor serum alone, or with added albumin or ethanol, in the absence of exogenous steroids. This control rate of cholesterol loss was subtracted from the experimental results in panels B through E. As the difference between the filled symbols (steroid uptake) and open symbols (cholesterol release) in Fig. 6 indicates, the rate of uptake of the various steroids by HTC cells far exceeds the rate of loss of cholesterol from the cell and suggests that a net uptake of the steroids occurs.
Steroid  5. Comparison of effect of steroid concentration on initial rate discarded after each wash. After the last wash the cells were resusof uptake and inhibition of reductase activity. HTC cells grown in pended in 0.2 ml of water, and the amount of cell-associated radioacmedium which contained lipoprotein-poor serum was used in these tivity was determined as described under "Methods." Initial rates of studies.
The steroids were added to 'O-ml cultures and incubated at uptake were extrapolated from the linear portion of the kinetic curve :G", and the kinetics of uptake was determined over a period of 0.5 for each steroid concentration tested. All of the steroids except 25 to 90 min. Aliquots (0.5 or 1.0 ml) ot the cultures were removed and hydroxycholesterol gave linear rates tot-the first 30 to 90 min. 25-hyrapidly discharged into 2 to 4 ml ot cold (4') Swim's 77 medium which droxycholesterol uptake was linear for only 0.5 to 1.5 min at 37". The contained defatted albumin (10 mg/ml). These samples were centri-effect of steroid concentration on the inhibition of HMG-CoA reducfuged at 4' for 2 min in a cell-washing centrifuge (Sorvall model W-l), tase activity in 4 hours was determined as described in Fig. :1. Stippled and the supernatant fluid was discarded. The cell pellets were washed area represents the normal concentration range of unesterified cholestwice with 2.0 ml of cold wash medium and the supernatant fluid was t,erol in growth medium which contained whole serum.  (1 x lo-' M), or with cycloheximide alone, and the kinetics of decay in HMG-CoA reductase activity were measured as described in Fig. 8. Half-life values were calculated from first order decay curves similar to the ones shown for 25-hydroxycholesterol (Fig. 8) Consistent with data reported previously from this laboratory (13), and as indicated by the tl,* values shown in Table  VI, reductase activity decays at approximately the same rate when either cycloheximide alone, maximal concentrations of whole serum, or both of these agents are added to the culture medium. These data suggest that whole serum, in contrast to Shydroxycholesterol, suppresses HIVlG-CoA reductase activity by decreasing the enzyme's rate of synthesis rather than by increasing its rate of inactivation.
Incubation of HTC cells with cholesteryl succinate gave results similar to those obtained with whole serum, suggesting, as did the results in Fig. 7, that both forms of cholesterol might inhibit reductase activity by the same mechanism.
The decreased tl,* values for the reductase of HTC cells incubated with cycloheximide plus 7-ketocholesterol or cholestenone, compared to cycloheximide alone, suggests that the inhibition produced by these steroids is similar to that produced by 25hydroxycholesterol and may involve a post-translational process.
Effects of Steroid Inhibitors on Apparent Rate of Synthesis and Rate Constant of Degradation of Reductase-The potentiation of the inhibitory effects of 25hydroxycholestero1, 7ketocholesterol, and cholestenone by cycloheximide suggests that reductase synthesis is not completely suppressed even at concentrations of these steroids which inhibit enzyme activity by 90% in 4 hours (Fig. 3). However, the results in Table VI do not allow a distinction to be made between an inhibitory mechanism which involves a partial suppression of enzyme synthesis combined with an increased rate of enzyme inactivation (degradation) and one which is exclusively the latter. To investigate the effects of the inhibitors on the components of reductase turnover, the apparent rate of synthesis (K,) and rate constant of degradation (Kd) of the enzyme were determined in the presence of varying concentrations of the inhibitors using the steady state kinetic analysis reported by Price et al. (45). In this analysis it is assumed that the rate of change in the amount (catalytic activity) of an enzyme can be described by the equation: where E is the amount of active enzyme (units x mass-'), K. is the zero order rate constant of synthesis of active enzyme (units x mass-' x time-'), and Ko is the first order rate constant for degradation of active enzyme (time-').
In the steady state, when dE/dt = 0, then K'. = ZCo E',, It can be shown that for the transition from one steady state to another steady state where E, is the amount of enzyme at any time, E,, is the amount of enzyme in the initial steady state, E',, is the amount of enzyme in the final steady state, and K'a is the final steady state rate constant for breakdown of active enzymes. When In (E, -E',J is plotted as a function of time, K'a can be calculated from the slope of the line, and K', is calculated from Equation 2.
The steady state kinetics of inhibition of reductase activity by varying concentrations of 25-hydroxycholesterol are shown in panel A of Fig. 9. As indicated, each concentration of steroid suppresses enzyme activity to a different steady state levelsuggesting that the rate of synthesis and rate constant of degradation of reductase are differentially affected by 25 hydroxycholesterol.
In panels B and C, In (E, -E',,) uersus time is plotted for 25-hydroxycholesterol and whole serum, respectively. Since the slopes of these lines reflect changes in K'o, it is apparent that increasing concentrations of 25-hydroxycholesterol enhance this rate constant. However, increasing whole serum has a minimal effect on the K'o value. The apparent K'D and K', values calculated from the data in Fig. 9 are shown in Table VII, along with the rate constants determined from similar studies with cholesteryl succinate, 7-ketocholesterol, and cholestenone. Also shown in Table VII  indicated, general cellular proteins are degraded at a rate of 1% per hour, and this rate is increased only 10 to 15% in the presence of 7-ketocholesterol. This is contrasted to a 300 to 900% enhancement in the apparent rate constant for degradation of reductase by the oxygenated steroids. Therefore, bulk protein turnover is not affected by the cholesterogenic inhibitors.

Recovery
of Reductase Activity in HTC Cells after 2 Hours Exposure to Cholesterogenic Inhibitors-To investigate whether the suppression of reductase activity by cholesteryl succinate and the other cholesterol derivatives persists after their removal from the culture medium, HTC cells were incubated with the various derivatives and whole serum for 2 hours, washed, transferred to fresh medium without the inhibitors, and assayed for reductase activity periodically over the next 24 hours.
As shown in Fig. 10, reductase activity returns to 100% of the control activity in cells transferred from a culture medium which contained whole serum or cholesteryl succinate, and to 90 to 95% of control in cells previously incubated with 25-hydroxycholesterol.
In contrast, reductase activity increases slowly and is still significantly suppressed relative to the control after 24 hours in HTC cells transferred from medium These data suggest that the inhibitory effect of whole serum, cholesteryl succinate, and 25.hydroxycholesterol are reversible. Whether the lack of continued suppression of enzyme activity is due to cellular efflux of the steroids after they are transferred to a fresh medium, or to some other process, has not been determined. The incomplete reversal of the inhibitory effects of 7-ketocholesterol once it has been removed from the extracellular medium suggest that at least some of this steroid remains in the cell in the form that is capable of inactivating newly synthesized reductase molecules.
Of interest in Fig.  11 is the fact that reductase activity continues to decrease for 6 hours after HTC cells are removed from a culture medium which contained whole serum, and remained depressed for an additional 3 hours before enzyme activity began to return to the control level. These results are different from those obtained with the other inhibitors tested and suggest that the continued suppression of HMG-CoA reductase activity may result from the slow degradation of specific serum lipoproteins, which bind tightly to receptors on the cell surface during the initial incubation period (47). Aliquots of 1.0 ml were removed from each flask at 0, 1, 2, and 4 hours after the addition of the steroid, and the cellular trichloroacetic acid (TCA)-soluble and insoluble radioactivity were determined as described hg Hershko and Tomkins (46). The total radioactivity in the cells at zero time was 400,000 cpm/lO* cells. are severalfold less effective suppressors of reductase activity in mouse hepatic cells as compared to mouse L-and human fibroblast cells, HTC cells are as sensitive to inhibition by these compounds as the latter two cell types. 25Hydroxycholesterol is several times more effective than 7-ketocholesterol as an inhibitor of cholesterol synthesis in HTC cells, whereas the two derivatives appear to be equally inhibitory in cultured mouse hepatic cells (14). Furthermore, cholestenone is a weak effector for mouse hepatic cells and a strong inhibitor for HTC cells. Whether these contrasting results are related to species and/or cell type differences, or to the fact that the mouse hepatic cells studied by Kandutsch and Chen (14) are cultured in serum-free media, is not clear.

DISCUSSION
In addition to the above differences we have found that cholesterol ester formation is apparently not associated with the inhibition of reductase activity by cholesterol and several of its derivatives in HTC cells. This is in contrast to the results obtained with human fibroblast cells (35). Although our studies on cholesterol ester formation were limited to isotope incorporation experiments, Wood and Folch (51) have quantitatively measured the change in total unesterified and esterified cholesterol in HTC cells maintained in a wide range of serum environments, and their results indicate that the sterol ester content changes maximally 2-fold. This is in contrast to the B-fold changes observed with human fibroblasts (52). In addition, Wiegand and Wood (42), monitored the long term incorporation of a number of C ,,-C ,1 acids into HTC cells and found that cholesterol ester synthesis was low.
The inhibition of reductase activity in HTC cells by cholesterol derivatives is concentration-dependent and specific ( Fig.  3 and Table V). In addition, high concentrations of the inhibitors added to homogenates of HTC cells do not alter the K, (for DL-HMG-CoA) or the V,,,., of reductase (data not presented). Analogous to our findings with HTC cells (3). the steroidmediated suppression of HMG-CoA reductase activity in other cultured cells (14, 16, 23) does not appear to involve a direct effect on the enzyme. It is possible that the steroids or a metabolite interact with a protein which is the proximal modulator of reductase activity and their effectiveness is lost in whole homogenates. In addition, our data do not allow us to assess whether a lipid or nonlipid second messenger are the effecters responsible for the suppression of reductase activity in HTC cells.
The short half-life (3.4 hours) of reductase in HTC cells is consistent with the fact that significant changes in its level can be produced by alterations in the rate at which it is synthesized and degraded.
Although it has been demonstrated that changes in the synthetic rate of reductase account for the diurnal variation in the hepatic enzyme activity observed in uiuo (12) and the large increases in activity measured in cultured cells transferred from a medium which contained cholesterol to one which is lipid-free (20), few examples of variations in catalytic activity which result from specific alterations in the rate of degradation (inactivation) or reductase have been reported. Edwards and Gould (41) and Higgins et al. (43,44) have demonstrated an enhanced loss of HMG-CoA reductase activity in whole animals after acute cholesterol feeding, however, no analyses were done to assess whether the effect was on the rate constant for degradation or the rate of synthesis or both.
Based on the rapid suppression of enzyme activity in human fibroblast cells by 7-ketocholesterol, Brown and Goldstein (16) have postulated, in agreement with our findings, that the suppression is unlike that induced by cholesterol associated in complexes with lipoproteins which inhibit reductase synthesis in their cells. In contrast, Kandutsch and Chen (14,15) have concluded that 7-ketocholesterol and 25hydroxycholesterol inhibit the rate of synthesis of HMG-CoA reductase in cultured mouse cells. Their conclusions are based on half-life values of reductase determined in the presence of 25-hydroxycholesterol and 7-ketocholesterol in the absence of an inhibitor of protein synthesis. Turnover data obtained in this manner are insufficient to differentiate between inhibitory effects on synthesis, degradation, or both of these processes. Kandutsch and Chen (15) report a till value for reductase in the presence of maximal concentrations of 25-hydroxycholesterol (in the absence of cycloheximide) of 1 hour, which is similar to our value of 0.8 hour obtained under the same conditions (Table VI). However, our studies with cycloheximide gave a till of 0.4 hour and suggest that 25-hydroxycholesterol enhances the degradation and/or inactivation of reductase. Data presented in this report suggest that cholesterol (serum cholesterol and cholesteryl succinate) and its oxidized derivatives (25-hydroxycholesterol, 7-ketocholesterol, and cholestenone) suppress HMG-CoA reductase activity in HTC cells by two different mechanisms. Based on the half-life of reductase in the presence of steroid inhibitors plus and minus cycloheximide, and on the effects of steroid concentration on its apparent rate of synthesis and rate constant of degradation, it appears that whole serum and cholesteryl succinate inhibit reductase by specifically decreasing its rate of synthesis; cholestenone, 7-ketocholesterol, and 25-hydroxycholesterol increase the rate of inactivation of reductase. Consistent with this data, cholestenone, 7-ketocholesterol, and 25-hydroxycholesterol are able to further suppress reductase activity in HTC cells grown in whole serum. High concentrations of cholesteryl succinate have essentially no additional effect (Fig.  7), but this may be due to its reduced uptake (one-tenth that observed for cells incubated in lipoprotein-poor serum). Steroid structure appears to play an important role in the inhibition of HMG-CoA reductase synthesis. The increased potency of oxygenated steroids relative to cholesterol does not appear to be related strictly to differences in steroid polarity (16), since cholestenone is less polar than cholesterol, and 25-hydroxycholesterol is several times more inhibitory, but less polar, than 7Lu-hydroxycholesterol.
In addition, the presence of an unesterified 3-hydroxy group at the 3 position of sterol nucleus is apparently not essential (16), since the succinate ester of cholesterol did not diminish, but significantly increased the inhibitory potency of this sterol. Whole serum cholesterol and cholesteryl succinate appear to cause a suppression in the rate of synthesis of reductase, and the oxygenated steroids cause an enhanced loss of activity. Therefore, our results would suggest that steroid structure is also important for the mechanism used by the cell to respond to these effecters.
Since HMG-CoA reductase has not been isolated and purified from HTC cells, the effects of the various steroid inhibitors on the apparent rate of synthesis and rate constant of degradation of the enzyme have been determined indirectly by an analysis of steady state kinetic data using loss of catalytic activity as an indication of enzyme turnover. This method involves numerous assumptions (53) which have not been critically evaluated in this study. Therefore, firm conclusions regarding the mechanism of suppression of HMG-CoA reduc-