Radioimmune Precipitation of 3-Hydroxy-3-methylglutaryl Coenzyme A Reductase from Chinese Hamster Fibroblasts EFFECT OF 25-HYDROXYCHOLESTEROL*

Antibody prepared against 3-hydroxy-3-methylglu- taryl coenzyme A (HMG-CoA) reductase of rat liver can be shown to inhibit this enzyme in extracts prepared from cultured Chinese hamster ovary (CHO-K1) cells. The molecular weight (53,000) of the HMG-CoA reductase subunits of rat liver and Chinese hamster liver is identical with a [36S]methionine-labeled polypeptide that can be precipitated from CHO-K1 lysates by this antibody used in conjunction with protein A Sepharose. It is shown, that 25-hydroxycholesterol which lowers HMG-CoA reductase activity in cultured fibroblasts blocks the incorporation of labeled methionine into this polypeptide. Furthermore, the antibody immune pre-cipitates two other polypeptides with molecular weights of 127,000 and 60,000. The latter polypeptide responds to 25-hydroxycholesterol in the same fashion as the 53,000-dalton polypeptide. In a dominant 25-hy-droxycholesterol-resistant mutant of the CHO-K1 cell, 25-hydroxycholesterol did not inhibit incorporation of labeled methionine into either the 53,000- or 60,000- dalton polypeptides.

The rate of cholesterol biosynthesis in cultured mammalian cells in the presence of serum is largely determined by the number of receptors for the low density lipoprotein. As has been shown by the studies of Brown and Goldstein (l), this protein, which is the major cholesterol carrier in serum, enters cells which possess receptors and after hydrolysis releases free sterol which in turn inhibits cholesterol biosynthesis. In tissue culture systems (2), it can be shown that even cells which lack the low density lipoprotein receptor will be inhibited in cholesterol biosynthesis if some other mechanism is provided to mediate the uptake of exogenous sterol. Studies in a number of laboratories have provided biochemical (3) and genetic (4-6) data to support the hypothesis, originally advanced by Kandutsch et al. (7), that an oxygenated sterol rather than cholesterol itself serves this key regulatory role in cholesterol biosynthesis in mammalian cells.
* This work was supported by National Institutes of Health Grants GM 24732, CA 15794, and HD 02080 to M. S. and grants from the National Institutes of Health (HL 19063) and the Greater Los Angeles affiliate of the American Heart Association to P. E. This is contribution No. 352 from the Eleanor Roosevelt Institute. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "aduertisernent" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
* Established Investigators of the American Heart Association.
Oxygenated sterols depress the activity of the enzyme 3hydroxy-3-methylglutaryl coenzyme A (HMG-CoA)' reductase, the enzyme responsible for the synthesis of mevalonic acid and believed to be the key regulatory enzyme in the cholesterol biosynthetic pathway (8). The mechanism by which oxygenated sterols affect HMG-CoA reductase activity in mammalian cells has not previously been directly demonstrated, although experiments with inhibitors of protein biosynthesis and RNA biosynthesis have been shown to prevent the full expression of HMG-CoA reductase when regulatory sterols are removed from culture medium (4), and immune titration experiments (9) have suggested that these sterols as well as other regulatory effectors (10, 11) can affect both the number and type of HMG-CoA reductase molecules. The object of this report is to present data obtained by direct radioimmune precipitation analysis of HMG-CoA reductase in a cultured fibroblast line, the CHO-K1 cell. These results show the effects of 25-hydroxycholest.ero1 on the biosynthesis of HMG-CoA reductase in these cells. We have chosen to examine a cultured Chinese hamster fibroblast line as a model because of the increasing interest in utilizing somatic cell mutants of such cells (4-6) in analyzing the mechanism of regulation of cholesterol biosynthesis.

EXPERIMENTAL PROCEDURES
Medium a n d Cells-The CHO-K1 cell (12) was used in these experiments. Cells were grown on Ham's F12 medium (13) supplemented with 8% newborn calf serum or 2% delipidized serum (14). Wilton,NH) was added to cultures in ethanol and made up immediately prior to use.
Assays a n d Antisera-The antibody (ammonium sulfate-fractionated (28) IgG) used in this study was prepared in rabbits against rat liver HMG-CoA reductase. This reductase antibody was prepared against enzyme with a specific activity of 21,000 nmol of NADPH/ min/mg of protein and monospecificity was demonstrated by Ouchterlony double diffusion and immunoelectrophoresis (IO). Nonspecific IgG was also prepared from normal rabbit serum by ammonium sulfate fractionation. Both specific and nonspecific IgG were kept at a concentration of 10 mg of protein/ml in 0.15 M NaCl, pH 8.4, and 0.01 M borate buffer at 4 "C. HMG-CoA reductase activity was assayed in CHO-K1 cell extracts prepared by freeze-thawing in Kyro EOB as has been previously described (4). Extracts prepared in this fashion are stable for at least 18 h at room temperature.
Radioimmune Precipitation Analysis of HMG-CoA Reductase from Chinese Hamster Ovary Cells-Cells (5 X lo6) are plated on to each of six 150-mm Petri plates and incubated for 24 h. The medium is then changed to F12 supplemented with 2% delipidated calf serum I The abbreviations used are: HMG-GOA, 3-hydroxy-3-methylglutaryl coenzyme A CHO, Chinese hamster ovary; SDS, sodium dodecyl sulfate.
(13) and incubated for 16 h. The cells are then starved for methionine in an identical medium missing methionine for 1 h. Fifty pCi of [35s1 methionine (New England Nuclear, Waltham, MA) (750 Ci/mmol, final concentration of 4 X lo-' M) is then added to each plate and the cells are incubated for a period of 3 h. Under these conditions, approximately 1 x IO7 dpm of radioactive methionine is incorporated per plate of cells treated into trichloroacetic acid-precipitable material. No inhibition of incorporation into total protein by 25-hydro~ycholesterol is observed (Table I). We have found, however, that incubation of cells in no and then low methionine medium lowers the activity of HMG-CoA reductase about 33% and raises the half-life of the enzyme activity in the presence of 25-hydroxycholesterol from 1.6-3 h. The consequences of these effects on enzyme activity are shown in Table II. preparation is then sonicated in a Branson Sonifier with three 2-s bursts and heated at 37 "C for 1 h. An aliquot is removed for enzyme assay and the sample is then cleared of insoluble material by ultracentrifugation at room temperature at 100,OOO x g for 1 h. An aliquot of the supernatant is assayed so that solubilization of enzyme activity can be determined on each sample. This varies from 25-50%. The supernatant is diluted 1 to 3 with pH 8.0, 0.1 M Tris, a necessary dilution of the detergent in order to permit recognition by antibody. Normal rabbit antibody (10 pl of a 10 mg/ml crude IgG preparation) is then added and incubated with shaking for 1 h at room temperature. A slurry of protein A beads (Sigma, St. Louis, MO) in phosphatebuffered saline (75 pl) is then added and the samples incubated for 30 min at room temperature with shaking. The beads are then removed by centrifugation in an Eppendorf centrifuge. The supernatant is removed and mixed with specific antibody (0.1 mg/ml final concentration) and incubated with shaking for another hour at room temperature. In controls, normal rabbit IgG (0.1 mg/ml) was used in this step instead of specific antibody. A slurry of protein A beads in phosphate-buffered saline (25 pl) was then added and the samples incubated with shaking at room temperature for another 30 min. The pelleted beads were transferred to a second Eppendorf centrifuge tube and washed as follows: two washes with 0.15 ml of NaCl in 0.1 M Tris, pH = 8.0; two washes in 0.5 M LiCl in 0.1 M Tris, pH = 8.0; one wash in 1% Triton X-100 and 1% sodium deoxycholate in 0.15 M NaCl and 0.05 M Tris, pH 7.4, and one wash in 0.15 M NaCl in 0.05 M Tris, pH =i 7.4. The pelleted, washed beads were then resuspended in 0.3 m l of phosphate-buffered saline and transferred to another Eppendorf mented with 8% newborn calf serum in 60-mm Petri dishes and incubated for 24 h. The medium was then changed to F12 supplemented with 2% delipidized serum and the cells incubated for another 16 h with or without 25-hydroxycholesterol (0.5 pg/ml). The cells were then incubated in methionine-free medium followed by labeling with [35S]methionine (8 pCi/dish, 4 X lo-' M) for 3 h. The cells were harvested by scraping into 600 p1 of 0.05 M Tris, pH 7.4, in 0.15 M NaCl and broken by sonication. Fifty pl of sample was set aside for a protein determination and 5 pl was added to 900 pl of cold 7% trichloroacetic acid. After 5 min on ice, the samples were collected on Millipore Fiters and washed one time with 2 ml of cold 7% trichloroacetic acid followed by two washes with 5 ml of 7% trichloroacetic acid. The filters were then counted bv liauid scintillation.  Results are shown f S. D. centrifuge tube. The beads were then pelleted and resuspended in 100 pl of SDS sample buffer and heated at 100 O C for 5 min. After a brief centrifugation to remove beads, the supernatant was analyzed by SDS-polyacrylamide gel electrophoresis.
Electrophoretic Analysis-SDS-polyacrylamide gel electrophoretic analysis was carried out according to Laemmli (25) in 0.1% SDS/ Tris/glycine buffer on 5-20% (w/v) gradient acrylamide slab gels using a 5% (w/v) stacking gel. The gels were poured so as to have a 5-10% glycerol gradient across the gel. Samples for electrophoresis were dissolved in 3% (w/v) SDS, 10% (v/v) glycerol, 60 mM Tris, pH 6.8,5% (v/v) 0-mercaptoethanol, and 0.001% bromphenol blue (SDS sample buffer) and heated at 100 "C for 5 min before layering onto the gel. Gels were run at 150 volts until the tracker dye had reached the bottom and then were stained with 0.01% Coomassie blue in 50% trichloroacetic acid and then soaked in 1 M sodium salicylate containing 5% glycerol for 30 min with shaking. The gel is then dried and the sample bands visualized by fluorography at -80 "C using Kodak AR fiim with a Dupont Cronex Intensifying Screen. In the experiments reported, bands were visualized as shown by a I-week exposure. Apparent molecular weights of radioactive bands were determined by reference to the following standardproteins: phosphorylase b (94,000), bovine serum albumin (67,000), ovalbumin (43,000), and carbonic anhydrase (30,000).
Purification of 3-Hydroxy-3-methylglutaryl Coenzyme A Reductase-HMG-CoA reductase was purified from rat liver as previously described (10). This enzyme was also prepared from 210 mg of microsomal protein prepared as for rat liver (10) from the livers of 20 adult male Chinese hamsters placed on cholestyramine-supplemented diets (5%, w/w) for 4 days before sacrifice. For 14 days prior to sacrifice, the animals were maintained on a 12-h light-dark cycle and sacrificed 6 h into the last dark cycle. The specific activity of HMG-CoA reductase in the microsomal suspension was 0.15 nmol of mevalonate/min/mg of protein. The enzyme was purified as has been described for rat liver (10) to a final specific activity of 4.7 X lo3 nmol of mevalonate/min/mg of protein. The liver enzyme was assayed as previously described (27).

RESULTS
Kinetics of the Appearance and Disappearance of Enzyme Activity-In order to define the type of experiments necessary to elucidate HMG-CoA reductase regulation, it is valuable to examine the time course of enzyme activity as it changes under various conditions. When such an experiment is performed on the changes in HMG-CoA reductase activity under standardized conditions (Fig. l), it is found that after 16 h of incubation in delipidized serum, the activity is substantially higher relative to controls incubated with 25-hydroxycholesterol and that a half-life for enzyme activity of 1.6 h in the presence of 25-hydroxycholesterol is observed (Fig. 2). These were incubated for 24 h after inoculation in normal growth medium. The medium was changed to F12 supplemented with 2% delipidized newborn calf serum and incubated for another 18 h. At this time, 25hydroxycholesterol (0.5 pg/ml) was added to each plate and cells harvested in duplicate and assayed for HMG-CoA reductase activity at the time shown. We have previously reported a half-life of HMG-CoA reductase activity determined with cycloheximide of 2.3 h in CHO-K1 cells (4,23). results are in accord with data that have been published by other workers (6, 15) and suggest that observation of the incorporation of radioactive amino acids into HMG-CoA reductase after 16 h of incubation of CHO-K1 cells in delipidized serum would be of interest.

Radwimmune Precipitation Analysis of the Incorporation of [%]Methionine into Cultured Chinese Hamster Fibroblast HMG-CoA Reductase-We have found that it is difficult to isolate large quantities of HMG-CoA reductase from
Chinese hamster liver by published procedures (10) although preparation of small amounts of purified enzyme by these methods is possible. This is due to a combination of the small size of Chinese hamsters and an order of magnitude lower specific activity (and, hence, presumably enzyme protein) in the livers of these animals as opposed to rat liver. However, purified enzyme from Chinese hamster liver has a similar subunit molecular weight to that of rat liver (Fig. 3) of 53,000, suggesting that the hamster enzyme might be cross-reactive with the rat enzyme antibody. This has in fact proven to be the case (Fig. 4) 0). Approximately 5 pg of material was analyzed by SDS-polyacrylamide gradient (5-20%) gel electrophoresis. Samples were visualized by means of the diamine silver stain of Switzer (24). The preparation shown in lane C was used in the preparation of antisera. Lane B shows a series of molecular weight standards 94,000 (phosphorylase b), 67,000 (bovine serum albumin), 43,000 (ovalbumin), 30,000 (carbonic anhydrase), 20,000 (trypsin inhibitor), and 14,000 (lactalbumin). The molecular weight of the prominent sample band is 53,000. FIG. 4. Inhibition of CHO-K1 HMGCoA reductase activity by antibody to the rat liver enzyme. HMG-CoA reductase activity was solubilized (100,000 X g supernatant) from le CHO-K1 cells by treatment with 0.5% Kyro EOB. Antibody (ammonium sulfate-fractionated IgG, 10 mg/ml) was added to various final concentrations in the standard enzyme assay 30 min prior to substrate to test for capacity to inhibit enzyme activity. Ammonium sulfate-fractionated IgG from nonimmunized rabbits produced no inhibition of enzyme activity at IgG concentrations as high as 2 mg/ml. (Control, 1.0  precipitation protocol to analyze the effect of 25-hydroxycholesterol on the incorporation of radioactive methionine into CHO-K1 cell HMG-CoA reductase. Cells were incubated for 16 h in delipidized serum in order to maximize HMG-CoA reductase activity. At this time, the cells were starved for methionine for 1 h and then labeled with [35S]methionine for 3 h. In samples treated with 25-hydroxycholestero1, this compound was added at the beginning of the incubation period with delipidized serum and was present at all times until the cells were harvested. Under these conditions, cells treated with 25-hydroxycholestero1 show a substantial decrease in measured HMG-CoA reductase activity of about an order of magnitude. The HMG-CoA reductase was solubilized by treating whole cells with the detergent Kyro EOB and membranes removed from the preparation by ultracentrifugation at 100,000 x g for 1 h. An aliquot of 100,000 x g supernatant was removed and assayed for enzyme activity to ensure comparable solubilization in treated and control samples. These extracts were immune precipitated with antibody to rat liver HMG-CoA reductase and protein A Sepharose. After washing the protein A Sepharose beads thoroughly to remove as much nonspecific binding material as possible, the immune-precipitated material was solubilized in SDS sample buffer and applied to a 5-20% SDS-polyacrylamide gel for electrophoresis. T h e results (Fig. 5) show that treatment with 25-hydroxycholesterol substantially decreases the incorporation of radioactive methionine into polypeptides with molecular weights of 53,000 and 60,000. We also observed that the antibody seems to precipitate a polypeptide with molecular weight of approximately 127,000.
A similar result was obtained (Fig. 6 ) A and B ) with 25-hydroxycholesterol were immune precipitated with specific antiserum (lanes B and D ) or preimmune serum (lanes A and 0. Lane E contains molecular weight markers (94,000, 67,000, 43,000, and 30,000). In this experiment, cells were incubated in medium supplemented with 2% delipidated serum for 16 h in the presence or absence of 25-hydroxycholestero1 (0.5 &ml) prior to and during methionine starvation (1 h) and labeling (3 h). The HMG-CoA reductase activity in the 25-hydroxycholesteroItreated samples (2 X 10' cells) was 1.4 nmol of mevalonate/h/mg of protein of which 0.7 nmol of mevalonate/h was obtained in soluble form and subjected to immune precipitation. The untreated samples (2 X 1 0 ' cells) had an HMG-CoA reductase activity of 20 nmol of mevalonate/h/mg of protein of which 10 nmol of mevalonate/h of activity was obtained in soluble form and subjected to immune precipitation. This is a fluorogram of an SDS gradient gel (5-201).

Lysates prepared from cells (2 X IO') treated (lanes C and D) or untreated (lanes
FIG. 6. Radioimmune precipitation of HMG-CoA reductase solubilized from CHO-K1 microsomes. Cells were treated as described in the legend to Fig. 5 and then broken by sonication followed by preparation of microsomes. HMG-CoA reductase activity was solubilized by freeze-thawing microsomes into the following solution: 0.1% Triton X-100, 25% glycerol, 0.1 M sucrose, 0.05 M KCI, 0.05 M KnPO,, 0.03 M EDTA, 10 m M dithiothreitol, and 1% stock aprotinin at pH = 9.0. This was then sonicated with three 2-s bursts and incubated at 4 "C with shaking for 1 h. The preparation was cleared of insoluble material by a 100,000 X g ultracentrifugation for 1 h and the supernatant diluted 1:3 with 0.1 M Tris, pH = 8.0 containing 1% aprotinin prior to immune precipitation. Extracts prepared from microsomes of cells treated (lanes D and E ) or untreated (lanes B and C) with 25hydroxycholesterol were immune precipitated with specific antibody (lanes C and E ) or nonimmune serum (lanes B and D). Lane A contains molecular weight markers (94,000,67,000, 43,000, and 30,000). The reductase activity in the 25-hydroxycholesteroI-treated samples was 16 nmol of mevalonate/h/mg of protein, whereas that obtained from an identical number of untreated cells was 26 nmol of mevalonate/h/mg of protein. Twenty-five% of the total activity was obtained in soluble form for each sample and subjected to immune precipitation.
tracts subjected to immune precipitation. Cytosols (100,000 X g supernatant) of cells broken by homogenization showed no specifically immune-precipitable bands. Because of these results, all subsequent experiments were performed on extracts prepared from whole cells.
The decrease of an immune-precipitable polypeptide of molecular weight identical with that of purified HMG-CoA reductase concomitant with a decrease in enzyme activity brought about by 25-hydroxycholestero1 supports the contention that this compound produces its effects on enzyme activity by decreasing the number of enzyme molecules present in the cell. We next sought to determine whether this was a t least in part due to an effect of 25-hydroxycholestero1 on HMG-CoA reductase synthesis.
To examine the effect of 25-hydroxycholestero1 on HMG-CoA reductase synthesis, cells were incubated with delipidized serum and labeled with methionine as described above except that 25-hydroxycholesterol was added to the cultures simultaneously with the labeled methionine. Under these conditions, the cells treated with 25-hydroxycholestero1 show little difference in activity of HMG-CoA reductase from untreated cells (see 'Experimental Procedures"). After solubilization of HMG-CoA reductase and removal of insoluble material by ultracentrifugation at 100,000 x g for 1 h, an aliquot of the soluble extract was assayed for enzyme activity. Matched amounts of enzyme activity could thus be immune precipitated and analyzed by SDS-polyacrylamide gel electrophoresis. The results of such experiments (Fig. 7) show that treatment with 25-hydroxycholestero1 again results in decreased incorporation of label into the bands at 53,000 and 60,000. In order to verify the physiological relevance of these results, we have performed an identical experiment on a dominant 25-hydroxycholesterol-resistant somatic cell mutant (CRl) which has been previously described in detail (7). Prior measurement of the half-life of enzyme activity in cyclohexi- FIG. 7. The effect of 25-hydroxycholesterol on the incorporation of labeled methionine into HMG-CoA reductase. Cells were incubated in medium supplemented with 2 8 delipidated serum for 16 h and then starved for methionine for 1 h. Then labeled methionine was added for 3 h with or without 25-hydroxycholesterol (0.5 pg/ml). Lysates prepared from cells treated (lane D) or untreated (lanes B and C ) with 25-hydroxycholesteroI were immune precipitated with specific antibody (lanes C and D) or preimmune serum (lane B ) . Lysates were assayed prior to immune precipitation and aliquots chosen from each sample so that identical amounts of enzyme activity were treated with antibody regardless of whether the preparation came from cells exposed to 25-hydroxycholestero1. Lane A contains molecular weight markers (94,000,67,000,43,000, and 30,000). The reductase activity in the 25-hydroxycholesteroI-treated sample was 15 nmol of mevalonate/h/mg of protein and that of the untreated sample was 24 nmol of mevalonate/h/mg of protein. Fifty 96 of the total activity was obtained in soluble form for both samples. The amount of enzyme activity subjected to immune precipitation for each sample was 8.1 nmol of mevalonate/h. to immune precipitation and aliquots chosen from each sample so that identical amounts of enzyme activity were treated with antibody regardless of whether the preparation came from cells exposed to 25hydroxycholesterol. Lane A contains molecular weight markers (94,000, 67,000, 43.000, and 30,000). The reductase activity in the 25hydroxycholesterol-treated sample was 22 nmol of mevalonate/h/mg of protein and that of the untreated sample was 29 nmol of mevalonate/h/mg of protein. The soluble activity was 31% of the total in the untreated sample and 40% in the cells treated with 25-hydroxycholesterol. The amount of enzyme activity subjected to immune precipitation was 7.7 nmol of mevalonate/h for each sample. mide-treated wild type and 25-hydroxycholesterol-resistant somatic cell mutant cells suggested that this mutant would be defective in the regulation of synthesis of HMG-CoA reductase by 25-hydroxycholesterol (4). The results of an "activitymatched" immune precipitation experiment (Fig. 8) show no effect of 25-hydroxycholesterol on incorporation of labeled methionine into either the 53,000-or 60,000-dalton polypeptides.

DISCUSSION
The antibody which was used in these studies has previously been demonstrated to be monospecific in Ouchterlony double diffusion and immunoelectrophoresis experiments (10) with crude microsomal extracts. Yet, quite clearly, at least three microsomal polypeptides (Fig. 6) can be recognized by this antibody, only one of which corresponds in molecular weight to the subunit of purified HMG-CoA reductase from animal liver. This polypeptide, with a molecular weight of 53,000, seems to be affected by 25-hydroxycholestero1 in a fashion which corresponds well with the effects of this compound on the activity of HMG-CoA reductase, and it is our observations on this polypeptide that lead us to conclude that the synthesis of HMG-CoA reductase is affected by oxygenated sterols. However, the polypeptide with a molecular weight of 60,000 is also affected by 25-hydroxycholesterol in a fashion consistent with the effects of this compound on HMG-CoA reductase.
No definite conclusions about the relationship between this polypeptide and HMG-CoA reductase can be drawn from the studies in this report. We have noted, however, when we have compared samples with variable solubilization of enzyme activity that the band at 53,000 has become relatively more intense with increasing solubilization of activity. The gel shown in Fig. 7 is an example with 50% solubilization of both 25-hydroxycholesterol-treated and untreated samples. The pattern shown in Fig. 6 (lane C) is an example of 25% solubilization. Furthermore, in a somatic cell mutant which we have recently isolated which has an order of magnitude greater HMG-CoA reductase activity than the wild type CHO-K1 cell, both the 60,000 and 53,000 bands are increased. This mutant will be the subject of a subsequent report. Likewise, as described in this report, neither the 60,000 nor the 53,000 bands are affected by 25-hydroxycholesterol in a dominant 25-hydroxycholesterol-resistant mutant.
The role of the polypeptide with a molecular weight of 127,000 in understanding the mechanism of regulation of HMG-CoA reductase is not clarified by this report, although the labeling of this band relative to untreated controls does not seem to be reduced by 25-hydroxycholesterol treatment of cells.
A variety of studies have been performed in a number of laboratories over the last few years on the mechanism of regulation of HMG-CoA reductase by oxygenated sterols. Prior to this report, all such studies have been indirect, utilizing either immune titration experiments (9) or the analysis of the rates of disappearance of enzyme activity in the absence of protein synthesis by means of treatment with cycloheximide (4, 16). These studies have been performed on several cell types and taken together support the contentions that oxygenated sterols affect both the number and type of HMG-CoA reductase molecules and enhance the rate of disappearance of enzyme activity in the absence of protein biosynthesis. A similar conclusion has been reached regarding the mechanism of other regulatory effectors of HMG-CoA reductase by immune titration experiments (10,ll). A considerable amount of literature on the effects of phosphorylation on HMG-CoA reductase (17, 18) and the physiological condi-tions which alter the degree of phosphorylation (19-21) provides a plausible mechanism for regulation based on changing the catalytic efficiency of this enzyme.
Because of the indirect nature of prior attempts to determine whether the number of HMG-CoA reductase molecules in cells is altered by treatment with oxygenated sterols, and, particularly, because of the potential difficulties in interpreting data obtained with cycloheximide (22), the role of enzyme turnover in the regulation of this enzyme has remained somewhat ambiguous. The direct demonstration of effects on synthesis reported above and current studies on degradation should help eliminate this ambiguity. Furthermore, radioimmune precipitation analysis is essential in carrying forward studies on the mechanism by which HMG-CoA reductase synthesis and degradation are regulated, particularly when performed on somatic cell mutants defective in these processes (23).