Regulation of Cytosolic Acetoacetyl Coenzyme A Thiolase, 3-Hydroxy-3-methylglutaryl Coenzyme A Synthase, 3-Hydroxy-3-methylglutaryl Coenzyme A Reductase, and Mevalonate Kinase by Low Density Lipoprotein and by 25-Hydroxycholesterol in Chinese Hamster Ovary Cells*

Removal of lipids from growth media of Chinese ham- ster ovary cells for 48 h resulted in significant increases in activities of the first four enzymes in the cholesterol biosynthetic pathway. In three experiments, the average maximal detectable increase in activity was 3.6- fold for cytosolic acetoacetyl-CoA thiolase, 16-fold for 3-hydroxy-3-methylglutaryl coenzyme A (HMG-CoA) synthase, 11-fold for HMG-CoA reductase, and 2.7-fold for mevalonate kinase. The activity increase of HMG- CoA reductase was shown to occur sooner than those of the other three enzymes, which is consistent with the concept that HMG-CoA reductase is the rate-limiting enzyme for sterol synthesis. Activity increases of all four enzymes were prevented largely by including low density lipoprotein (LDL) or 25-hydroxycholestero1 in lipid-depleted serum medium. Moreover, increased activities of these four enzymes were all shown to be suppressed by adding either LDL or 25-hydroxycholes-terol in the growth medium, implicating that HMG-CoA reductase is not the only enzyme under regulatory control by LDL. These results are consistent with the concept that 25-hydroxycholestero1 mimics the intra- cellular action(s) of LDL in tissue culture cells. Two putative mutants


Regulation of Cytosolic Acetoacetyl Coenzyme A Thiolase, 3-Hydroxy-3-methylglutaryl Coenzyme A Synthase, 3-Hydroxy-3-methylglutaryl Coenzyme A Reductase, and Mevalonate Kinase by Low Density
Lipoprotein and by 25-Hydroxycholesterol in Chinese Hamster Ovary Cells* (Received for publication, December 13, 1979, and in revised form, April 14, 1980) Ta-Yuan Chang and James S. Limanek Removal of lipids from growth media of Chinese hamster ovary cells for 48 h resulted in significant increases in activities of the first four enzymes in the cholesterol biosynthetic pathway. In three experiments, the average maximal detectable increase in activity was 3.6fold for cytosolic acetoacetyl-CoA thiolase, 16-fold for 3-hydroxy-3-methylglutaryl coenzyme A (HMG-CoA) synthase, 11-fold for HMG-CoA reductase, and 2.7-fold for mevalonate kinase. The activity increase of HMG-CoA reductase was shown to occur sooner than those of the other three enzymes, which is consistent with the concept that HMG-CoA reductase is the rate-limiting enzyme for sterol synthesis. Activity increases of all four enzymes were prevented largely by including low density lipoprotein (LDL) or 25-hydroxycholestero1 in lipid-depleted serum medium. Moreover, increased activities of these four enzymes were all shown to be suppressed by adding either LDL or 25-hydroxycholesterol in the growth medium, implicating that HMG-CoA reductase is not the only enzyme under regulatory control by LDL. These results are consistent with the concept that 25-hydroxycholestero1 mimics the intracellular action(s) of LDL in tissue culture cells. Two putative mutants (clones 25-RA and 25-RB) have been isolated for 25-hydroxycholesterol resistance from mutagenized Chinese hamster ovary cells. It was shown that 25-RA and 25-RB cells grown in 10% fetal calf serum medium had elevated levels of the first four cholesterogenic enzymes as compared with those found in the wild type cells; moreover, activities of these enzymes were all shown to be more resistant to suppression by 25-hydroxycholestero1 in 25-RA and 25-RB cells than those in wild type cells. These findings suggest the existence of a common cellular factor controlling the intracellular action(s) of 25-hydroxycholesterol on activities of the first four cholesterogenic enzymes; the function of this common controlling factor is defective or abnormal in clones 25-RA and 25-RB. These results, coupled with the fact that 25-hydroxycholesterol mimics the intracellular action(s) of LDL, imply that this putative common controlling factor may also be involved in mediating the intracellular action(s) of LDL-bound cholesterol (or its metabolite) on various cholesterogenic enzymes.
* This work was supported by National Institutes of Health Grants HL 21246 and IP30 CA 23108-01 Al. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "aduertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. Goldstein, Brown, and their co-workers discovered and delineated the low density lipoprotein (LDL)' receptor-mediated regulation of cholesterol synthesis in various nonhepatic cells (for a review, see Ref. 1). One of the regulatory signals elicited by the LDL-bound cholesterol (or its metabolite) causes suppression of 3-hydroxy-3-methylglutaryl coenzyme A (HMG-CoA) reductase, the putative rate-limiting enzyme in the cholesterol biosynthetic pathway (2)(3)(4). The same investigators have also reported the coordinate regulation of HMG-CoA reductase and HMG-CoA synthase by LDL in rat adrenal gland (5). Under certain conditions, HMG-CoA reductase and 4-methyl sterol oxidase in rat liver also exhibit coordinate feedback regulation as shown by Spence and Gaylor (6). Using cholesterol feeding experiments, Lane's laboratory has demonstrated that cytosolic acetoacetyl-CoA thiolase and cytosolic HMG-CoA synthase activities are under feedback control in rat and chicken liver ( 7 ) . The increase of cytosolic HMG-CoA synthase activity in tissue culture cells after removal of serum lipids from the growth medium has been demonstrated by Ramachandran et al. (8).
Kandutsch and Chen were the first to show that certain oxygenated analogs of cholesterol, especially 25-hydroxycholesterol (25-OH cholesterol), when present in the culture medium, cause strong suppression of HMG-CoA reductase and reduction of cholesterol synthesis in various cell types (4, 9). The cytotoxic effect(s) of 25-OH cholesterol and other oxygenated sterols eventually cause cell death (4,9). The mechanism for oxygenated sterol-mediated suppression is not clear, but was shown to bypass the cell surface LDL receptor (10). In cultured human fibroblast cells, these oxygenated sterols were shown to mimic the intracellular action(s) of LDL-bound cholesterol (10).
Previously, this laboratory reported the isolation of a Chinese hamster ovary (CHO) cell mutant defective in derepression of HMG-CoA reductase, lanosterol demethylation activity, and unsaturated fatty acid biosynthesis as three of its various phenotypic abnormalities (11). The isolation of such a mutant and biochemical analysis of its spontaneous revertant indicates that activities of certain cholesterogenic enzymes as well as activity for unsaturated fatty acid biosynthesis may be coordinately regulated by a common control mechanism (11). In experiments reported in this paper, we first validate a specific procedure (12) for preparing the cytosolic fraction which contains optimal cytosolic cholesterogenic enzymes from intact CHO cell monolayers; we then demonstrate that in CHO cells, the activities of the first four enzymes in the sterol biosynthetic pathway (for a review, see Ref. 13) are all regulated by LDL and by 25-OH cholesterol. Moreover, in two putative CHO cell mutants isolated for resistance to cytotoxic action($ of 25-OH cholesterol (designated as clone 25-RA and 25-RB), the activities of these four enzymes are all found to be more resistant to suppression by 25-OH cholesterol than those found in wild type cells. These findings suggest the existence of a common cellular factor controlling the intracellular action(s) of 25-OH cholesterol on activities of the first four cholesterogenic enzymes; the function of this cellular factor is defective or abnormal in clones 25-RA and 25-RB.

MATERIALS AND METHODS
Lipids were from Sigma, except 25-OH cholesterol was from Steraloids. 25-OH [24-3H]cholesterol was synthesized from desmosterol by Dr. Albert F. T. Chen at Beckman Microbia Division of Beckman Instruments using selective oxymercuration followed by NaBTd treatment according to the published procedure (14). Both the unlabeled and labeled 25-OH cholesterol and their acetate derivatives were shown to be at least 98% pure by four different thin layer chromatographic (TLC) systems. Both the unlabeled and labeled free sterols were shown further to contain no more than 6% detectable impurity by gas-liquid chromatographic (GLC) analysis or by radio-GLC analysis as described (15). Other radioactive chemicals were from New England Nuclear. Various coenzyme A derivatives were from Sigma except acetoacetyl coenzyme A (99% pure) was from P-L Biochemicals. All other chemicals were of analytical grade. Human low density lipoprotein (LDL), d 1.019 to 1.063 g/ml, and high density lipoprotein (HDL), d 1.063 to 1.215 g/&, were prepared according to published procedures (16, 17). The composition of LDL and the relative distribution of its lipid components agreed with published data (18). The HDL contained 0.5 mg of cholesterol and 0.3 mg of fatty acid/mg of protein.
Cells-Cultures of CHO cells (11,15) were grown as monolayers in 75-cm' Corning flasks in F-12 medium (linoleic acid-deleted) plus either 10% fetal calf serum, or 6% human lipoprotein-deficient serum (LP(-), 4 mg of protein/&), or 10% delipidated fetal calf serum (De-S, 4 mg of protein/&). LP(-) and De-S sera were prepared according to published procedures (16, 19). 25-OH cholesterol was added to the growth medium from stock dimethyl sulfoxide (Me'SO) or stock ethanolic solutions. Control experiments indicated that the presence of 0.3% Me2SO for 24 h in wild type cell cultures grown in F-12 + 10% De-S had no detectable effect on any of the fmt four cholesterogenic enzyme activities isolated from these cells; a separate control experiment indicated that the presence of 1% M e 8 0 for 8 h in both wild type and 25-RA cell cultures grown in F-12 + 10% De-S medium had no detectable effect on any of the four cholesterogenic enzyme activities isolated from these cells. The 25-OH cholesterol-resistant cells were isolated by growing mutagenized CHO cells in F-12 + 10% De-S + 1 pg/ml of 25-OH cholesterol + 0.3% ethanol according to the enrichment method previously described (20). Candidates for 25-OH cholesterol resistance were cloned twice before biochemical analyses. A total of eight independently isolated resistant clones were obtained from 5 X 10" mutagenized cells. The resistant phenotypes of each of these eight clones were found to be stable after continuous passages in F-12 + 10% fetal calf medium for at least 6 months. Two of these clones, designated as 25-RA and 25-RB, are studied in detail in this report. The cell-doubling time of clone 25-RA and 25-RB cells was found to be 14 to 15 h which was very similar to that of wild type cells (15 to 16 h). Growth rates of these three cell types in F-12 + 10% De-S were also found to be very similar (see Fig. 3 under "Results"). All experiments reported in this paper involving these three cell types were carried out using the following procedure: stock flasks of monolayers grown at confluency in F-12 + 10% fetal calf serum were dissociated by 0.05% trypsin; cells were seeded at various cell densities (as indicated in each figure or table under "Results") and allowed to grow in F-12 + 10% fetal calf serum for at least 48 h to assure steady and optimal cell growth before they were switched to any other type of medium. For all experiments reported here (except for some experimental points in Figs. 3 and 4 and Table V), cells were harvested during log-phase growth by visual examination. All three cell types were plated, grown, harvested, and assayed at the same time. At any given time point, the differences in cellular protein content per flask between the three different cell types were found to be within 30%. For most of the experiments reported here, a single source of cell culture was used for each experimental point; whenever duplicate cultures were used, variation between the duplicates was within 5%.
Lipid Extraction Procedure-Lipids were extracted as reported previously (11) except that cells were dissolved in 2 ml of 1.5 M KOH; [1,2-~"H]cholesterol (106 X IO1 dpm/tube) and [9-"H]oleic acid (43 X 10' dpm/tube) were added as internal standards for recovery purposes. After an aliquot was taken for protein determination, the aqueous slurry was evaporated under NB to near dryness. Then 3 ml of ethanol/benzene (4:l) was added and the solution was saponified at 80°C for 1 h. A control experiment indicated that this procedure hydrolyzed over 95% of a 200-pg cholesterol oleate sample into free cholesterol and oleate.
Lipid and Protein Determination-GLC and TLC of sterols and fatty acids were done as described (11, 21) except that Whatman LK5DF plates (20 X 20 cm) were used. Protein was determined by either microbiuret (22) or Lowry (23).
Enzyme Assays-Acetoacetyl-CoA thiolase, HMG-CoA synthase, and mevalonate kinase activities were assayed by published procedures (5, 24, 25) with slight modifications as described below. HMG-CoA reductase was assayed as reported (11). For HMG-CoA synthase and HMG-CoA reductase assays, cell extracts were prepared by Dounce homogenization after hypotonic shock treatment (11). Whole cell homogenate was used for reductase assay; the cytosolic fraction, prepared and dialyzed for 24 h in appropriate buffer as described (5), was used for synthase assay. The reaction rate for the synthase assay was found to be linear for 14 min with no more than 6% substrate consumption. For acetoacetyl-CoA thiolase and mevalonate kinase assays, cell monolayers were rinsed five times with 10 ml of phosphatebuffered saline at 4OC, cell extracts were prepared by sequentially exposing one to four Corning flasks (75-cm' size) of intact monolayer cells to 1.0 ml of buffered digitonin solution at room temperature (exposure time: 2 min/flask) as described (12,26). Each flask was

Regulation of Cholesterogenic Enzymes in CHO Cells 7789
tilted back and forth continuously by hand during exposure. Before exposure, cell monolayers were covered with 10 ml of phosphatebuffered saline at 4°C. After the fist exposure, the 1.0 ml of digitonin solution (containing cytosolic fraction from the first flask) was transferred to the second flask for the second exposure, and so on. The pooled cytosolic fraction was centrifuged at 10, OOO X g for 10 min at 4°C to pellet the insoluble digitonin suspension; the supernatant was stored at -7O'C before being used for assays. Control experiments indicated that, stored at -7O"C, the acetoacetyl-CoA thiolase activity and mevalonate kinase activity were stable for at least 7 days. The freshly prepared supernatant fraction, if dialyzed as described (5), was also used for HMG-CoA synthase activity in some experiments reported in this paper. The specific activity of mevalonate kinase and HMG-CoA synthase prepared by hypotonic shock treatment described above or by digitonin exposure treatment agreed with each other (see Tables I and I1 under "Results"). Control experiments showed that the HMG-CoA synthase activity prepared by either method was not stable upon storage, and needed to be assayed immediately after dialysis. For acetoacetyl-CoA thiolase assay, the reaction was performed at pH 8.0; the reaction mixture was preincubated at 25°C for 2 min, after which the reaction was started by addition of cell extract. The reaction was continued for at least 2 min. Reaction rate was found to be linear for at least 4 min. A stoichiometry of 2 mol of acetyl-coA formed/mol of acetoacetyl-CoA cleaved was confirmed by coupling acetyl-coA formation to NAD' reduction via the combined citrate synthase-malate dehydrogenase system as described (24). For the mevalonate kinase assay, the reaction rate was found to be linear for 12 rnin with no more than 10% substrate consumption. For all enzyme assays reported in this paper, two enzyme levels were used for each assay point to assure linearity between the two enzyme levels.

RESULTS
Validity of Sequential Digitonin Exposure Treatment in Preparing Cytosolic Cholesterogenic Enzymes in CHO Cells-We sought to develop an efficient and rapid procedure to prepare cytoplasmic fractions from monolayers of CHO cells with minimal contamination from mitochondrial leakage. This was necessary since Clinkenbeard et al.
Validity of sequential digitonin exposure treatment for releasing HMG-CoA synthase activity from CHO cells Wild type and 25-RA cells were plated at 0.2 X 106/75-cm' flask in 10 ml of F-12 + 10% fetal calf serum for 48 h. Medium was then renewed once and celh were grown for an additional 24 h. Afterwards, cytosolic fractions were prepared from cells by two different methods as described under "Materials and Methods." Cells from two flasks were pooled and used for Method 1. The cell extracts were either assayed right away, or dialyzed for 24 h at 4°C as described (5) before being used for enzyme assay. HMG-CoA synthase activity was determined as described under "Materials and Methods." Values shown represent the mean f variation from the mean of two enzyme assays using a single source of cell extract.   Table I indicate that, when a single flask is used, the digitonin exposure treatment causes rapid and efficient release of cytoplasmic enzymes with minimal release of enzymes from mitochondria. The specific activity of mevalonate kinase prepared by this method was shown to be similar to that prepared by the method of hypotonic shock followed by Dounce homogenization and ultracentrifugation. We have also found that this method can be used to extract sequentiauy cytosolic fractions from up to four separate 75-cm2 flasks of monolayer cells into 1 ml of buffered digitonin solution with efficient recovery in total cytosolic protein contents, and with efficient and constant recoveries in mevalonate kinase and acetoacetyl-CoA thiolase activities (Table I). In Table 11, we have validated this procedure further for releasing HMG-CoA synthase activities from both the wild type CHO cells and from a specific clone of CHO cells (designated as clone 25-RA) which has been found in this laboratory to contain much elevated activity of HMG-CoA synthase (the isolation and characteristics of 25-RA cells will be described later in this report). Data in Table   I1 also point out that, even with the digitonin exposure method, the dialysis step (5, 13) is still needed in order to obtain optimal activity measurements for HMG-CoA synthases in both cell types. Previously, Clinkenbeard et al. (13) have shown that the dialysis step is needed to inactivate the powerful mitochondrial HMG-CoA lyase which interferes with the HMG-CoA synthase enzyme assay used in this report.

HMG-CoA synthase specific activity
Regulation of HMG-CoA Reductase Activity by LDL in CHO Cells-Removal of LDL from the growth medium causes rapid increase in activity of HMG-CoA reductase in CHO cells as shown in Fig. 1; the increased activity is rapidly suppressible in a dose-dependent manner by adding back LDL to the growth medium. A very high level of HDL caused only a slight suppression of HMG-CoA reductase activity. This suppression could occur via nonspecific endocytosis of HDL by these cells. This experiment confirms the reports by Brown and Goldstein that CHO cells express LDL receptors at the cell surfaceL (1, 2 7 ) .
' In experiments not shown here, we have found that when 0.1 mM chloroquine was added to the culture medium concomitantly with LDL (at 10 to 200 pg/ml protein concentration) in a 6-h incubation, the suppression effect of LDL on HMG-CoA reductase activity was inhibited by a t least 80% in every single case, suggesting that lysosomal hydrolysis of LDL is required for LDL-mediated regulation of cholesterol metabolism in CHO cells; these results are similar to what has been demonstrated in human fibroblasts (28) (J. Chin and T. Y. Chang, results to be published elsewhere). Each experimental point shown in c was confirmed by using a different thiolase assay by measuring the acetyl-coA formation via the combined citrate synthase-malate dehydrogenase system as described (24).
Regulation of HMG-CoA Reductase, HMG-CoA Synthase, Cytosolic Acetoacetyl-CoA Thiolase, andMevalonate Kinase Activities by LDL a n d by 25-OH Cholesterol-In three different experiments, switching growth medium from 10% fetal calf serum to 10% delipidated serum for 48 h invariably causes an increase in the activities of the fiist four cholesterogenic enzymes in wild type CHO cells. In these experiments, the average maximal detectable activity increase was found to be 11-fold for HMG-CoA reductase, 16-fold for HMG-CoA synthase, 3.6-fold for acetoacetyl-CoA thiolase, and 2.7-fold for mevalonate kinase. A typical experiment was shown in Fig. 2, a to d. The increases in activities of all four enzymes invariably were prevented largely by including either LDL or 25-OH cholesterol in the delipidated serum medium. Moreover, increased enzyme activities were all invariably suppressed by adding either LDL or 25-OH cholesterol in the growth medium. The prevention effects and suppression effects of 25-OH cholesterol on activities of all four enzymes persisted regardless of whether this agent was added to the growth medium from a stock ethanolic solution or from a stock dimethyl sulfoxide (Me2SO) solution (data not shown). Data in Fig. 2  this is consistent with the concept that HMG-CoA reductase is the "rate-limiting" enzyme for sterol biosynthesis (2)(3)(4); however, these data also stress the fact that HMG-CoA reductase is not the only enzyme under regulatory control by LDL. Furthermore, these results are consistent with the concept (10) that 25-OH cholesterol mimics the intracellular action(s) of LDL-bound cholesterol (or its metabolite). The maximal detectable increases in activities of cytosolic acetoacetyl-coA thiolase and mevalonate kinase were not as large as those of HMG-CoA synthase and HMG-CoA reductase; the rates of activity increase of the former two enzymes were shown to be slower than the latter two enzymes. The less dramatic and slower increase of thiolase and kinase activities may explain why enhancement of their activities was not demonstrated in rat adrenal gland after cholesterol deprivation (5).
Characterization of 25-OH Cholesterol-resistant Clones 25-RA and 25-RB-Two different clones, designated as 25-RA and 25-RB, have been isolated for 25-OH cholesterol resistance from mutagenized CHO cells (see "Materials and Methods" for isolation procedure). Unlike wild type cells, these cells continued to grow (Fig. 3) and synthesize cholesterol (Fig. 4) in 25-OH cholesterol supplement medium ( 1 pg/ ml). GLC-sterol analyses indicated similar steady state incorporations of 25-OH cholesterol by wild type, 25-RA, and 25-RE cells (Fig. 4). We have also employed 25-OHVH]cholesterol to examine the uptake of cell-bound 25-OH cholesterol in three cell types. The results shown in Table I11 along /ml) (8, A, 0 ) . Media were renewed every day. Growth was followed by dissolving the cells in 1 ml of 2 M KOH and measuring protein. Each value represents the mean of duplicate cultures. Variation between the duplicates was within 5% of the mean. Cells were grown and harvested as described in Fig. 3. Sterols were analyzed by GLC as described (1 1 calf serum were allowed to grow for 63 h. Afterwards, cells were rinsed Medium was renewed and cells were grown for an additional 14 h. Afterwards, the medium was removed and cells were exposed to   Table IV indicate that 25-RA and 25-RB cells grown resistant to suppression by 25-OH cholesterol and by serum in 10% fetal calf serum possess elevated rates of sterol synthe-lipids (29,30). The data presented in Table IV extend these sis compared with that of wild type cells. The elevated rates of sterol synthesis in clones 25-RA and 25-RB were found not to correlate linearly with the elevated activities of HMG-CoA reductase found in these cells as compared with those of wild type cells; instead, these elevated rates can be explained much better by the finding that activities of the first four enzymes in cholesterol biosynthetic pathway in 25-RA and 25-RB cells were all elevated in comparison with those found in wild type cells (Table IV). Rates of "C-labeled fatty acid synthesis from [''Clacetate in these cells were shown to be only slightly elevated in comparison to that in the wild type cells (Table  IV, Column 3 ) . Previously, Chen et al. (29) and Sinensky et al. (30) reported the isolation of similar Chinese hamster mutant cells resistant to 25-OH cholesterol, and showed that the HMG-CoA reductase activities in these mutant cells were findings. These data also indicate that, under certain conditions, large changes in activities of several enzymes catalyzing early steps in cholesterol biosynthetic pathway can significantly alter the rate of sterol synthesis in intact CHO cells without substantial changes in activities of HMG-CoA reductase.

Increase and Resistance to Suppression by 25-OH Cholesterol of Cholesterogenic Enzyme Activities in Clones 25-RA and 25-RB Grown in 10% Delipidated
Serum-When wild type, 25-RA, and 25-RB cells grown in 10% fetal calf serum were switched to grow in 10% delipidated serum, it was found that, as in wild type cells, activities of all four cholesterogenic enzymes measured were increased in 25-RA and 25-RB (Fig.  5, a to d ) , indicating that the resistance to suppression by serum lipids of these four enzyme activities was not absolute TABLE IV Elevated rates of sterol synthesis, a n d elevated cholesterogenic enzyme activities in clone 25-RA a n d 25-RB cells as compared with those in wild type cells groum in F-12 + 10% fetal calf serum Cells plated a t 0.1 x 10" to 0.3 X 10"/75-cm2 flask in 15 ml of F-12 plated a t identical cell density, grown, harvested, and assayed at the + 10% fetal calf serum were allowed to grow for 48 to 72 h. Afterwards, same time. A single source of cell culture per cell type was used for cells were harvested for various enzyme assays in vitro. Cell extracts each experimental point. Specific enzyme assays were as described used for thiolase, synthase, and kinase assays were prepared by under "Materials and Methods." Enzyme activities were expressed as digitonin exposure treatment as described under "Materials and nanomoles. min". mg". Methods." For each individual experiment, all three cell tvpes were  I , 21). After the pulse, cells were dissolved in 2 ml of enzyme activities rt S.E. found in these 11 experiments. Cells were grown and harvested for enzyme assays as described in Fig. 2.   0, A, H, enzyme activities of c e b grown in 10% fetal calf serum. 0, A, 0, enzyme activities of cells grown in 10% De-S. Each point shown represents the mean of duplicate enzyme assays from a single source of cell extract; variations between the duplicates were within 10% of the mean for reductase, thiolase, and kinase. For the synthase assays, variations were within 15% of the mean. in 25-RA and 25-RB cells. The rates of activity increase of these enzymes were found to be similar in all three cell types with one exception: the HMG-CoA reductases of 25-RA and 25-RB cells never reached the highly elevated states reached by wild type cells during the entire time course (Fig. 5a). Also, the time course of activity increase of HMG-CoA reductase was shown to be slower in 25-RA and 25-RB cells than in the TABLE V Cellular cholesterol content of cells grown in F-12 -k 10% D e 3 Cells were grown as described in Fig. 3 except 25-OH cholesterol was deleted from the medium. Cellular cholesterol content was analyzed from duplicate cultures as described (11). Variation between duplicates was within 5% of the mean.  Table V). These results suggest that the derepression of HMG-CoA reductases of 25-RA and 25-RB cells may be more sensitive to inhibition by stored cellular cholesterol than are the other three cholesterogenic enzymes measured. Other possibilities are not excluded at present. We are currently pursuing this observation further in this laboratory.
After all three cell types were grown in 10% delipidated serum for 24 or 48 h, resistance to suppression by 25-OH cholesterol of the first four cholesterogenic enzymes in 25-RA and 25-RB cells were demonstrated (Fig. 6, a to d). While these data do not permit us to conclude that the degrees of resistance to suppression were different in 25-RA and 25-RB  have been consistently seen in two separate experiments. It is important to point out that the results presented in Fig. 6, a to d, cannot be consequences of nonspecific cytotoxic effects of 25-OH cholesterol on cell growth, since earlier experiments in this laboratory have shown that treatment of %-OH cholesterol at 50.3 pg/ml for 24 h has minimal inhibitory effect on cell growth of all three cell types (data not shown). Moreover, in a different experiment, when we treated wild type and 25-RA cells with various concentrations of 25-OH cholesterol in the same manner as what was described in Fig. 6, a to d, but for only 8 h, we have found that there were lower per cent suppressions of each of the four enzyme activities in 25-RA cells in comparison with those found in wild type cells. We do not know whether the defects manifested in clones 25-RA and 25-RB are due to a single gene mutation or not; however, the fact that these abnormalities were found in two independently isolated clones suggests that these phenotypic expressions were caused by a single gene mutation. The data presented in this report do not permit us to determine whether these two clones possess the same biochemical lesion (with different degrees of impairment at the mutated genetic locus).