Defective Elongation of Fatty Acids in a Recessive 250Hydroxycholesterol-resistant Mutant Cell Line*

counts/min in reductase to those in total proteins determined as trichloroacetic acid-precipitable material. The results are the mean + S.E. of triplicate cultures.

The Chinese hamster ovary recessive mutant, crB, has been selected for its resistance to the cytotoxic effects of 25-hydroxycholesterol in sterol-free media (Sinensky, M., Logel, J., and Torget, R. (1982)  malian cells by exogenous sterols (l-4). Mutants resistant to the killing effects of 25-hydroxycholesterol are, when selected under the appropriate conditions, also resistant to the downregulatory effects of 25hydroxycholesterol on the enzymes of cholesterol biosynthesis.
Analysis of the genotypes of these mutations by somatic cell hybridization techniques have shown that such mutations can be either dominant or recessive in at least two different complementation groups (5). It has been well documented that 25-hydroxycholesterol can act as a transcriptional regulator of 3-hydroxy-3-methylglutaryl-coenzyme A (HMG-CoA)' reductase in mammalian cells (6)(7)(8). The mRNA levels of at least two other enzymes of cholesterol biosynthesis, HMG-CoA synthase and farnesyl pyrophosphate synthetase, are also regulated by 25hydroxycholesterol, presumably also by a transcriptional control mechanism (9)(10)(11). We have previously described a dominant somatic cell mutant (CRl) that is pleiotropically defective in transcriptional regulation of enzyme activities of sterol biosynthesis by exogenous sterols (12, 13). However, translational control of HMG-CoA reductase appears to be intact in CR1 (13). The rate of degradation of HMG-CoA reductase is constitutively rapid in this mutant and is not further enhanced by exogenous sterols.
We now report that the recessive somatic cell mutant, crB, has a molecular biological phenotype similar to the one previously described for the dominant mutant, CRl. Surprisingly, in contrast to CRl, crB is auxotrophic for certain fatty acids and we present evidence consistent with the hypothesis that the primary defect conferring the 25-hydroxycholesterol resistance to this mutant appears to lie in the fatty acid elongation pathway. complexed to various fatty acids. Revertants of crB to fatty acid prototrophy were selected by standard procedures (15) involving overnight mutagenesis with ethyl methanesulfonate (150 rg/ml) followed by incubation in Nutridoma for 10 days. The frequency of spontaneous reversion in crB is less than 1 in 106.
Zsohtion of RNA and Northern Blot Analysis-Cells in F12DIPE2 were treated with 0.5 PM 25-hydroxycholesterol for 16 h as described before (13). Total cellular RNA was extracted (16) for Northern blot analysis (13). The blots were probed with either the 3.8-kilobase EcoRI HMG-CoA reductase insert from XDSll (18), the 1.5-kilobase Hind111 fragment from the hamster HMG-CoA synthase clone p53K-312 (9), or the 1.3-kilobase EcoRI fragment from the rat farnesyl pyrophosphate synthetase clone CR39 (11). The human or-tubulin insert (courtesy of Dr. W. Salser, UCLA) was used as an internal standard. The cDNA probes were labeled using random priming to a specific activity of at least 10s cpm/rg (19).

HMG-CoA
Reductase-Cells were seeded at 1 X 106/100-mm culture dish in 8 ml of F12FC5 and grown for 24 h (Protocol A). The monolayers were then rinsed twice with Puck's Saline G and refed 5 ml of F12DIPE5. Varying amounts of 25hydroxycholesterol were added in ethanol such that the final concentration of ethanol in the culture medium did not exceed 1% (v/v). After 16-18 h of incubation, the relative rate of synthesis and the half-life of degradation of HMG-CoA reductase were determined following [35S]methionine labeling as described previously (13). For the measurement of reductase activity, the cells were seeded at 3 x 106/60-mm culture dish in 3 ml of F12FC5 and grown for 24 h (Protocol B). After rinsing twice with Saline G, they were refed 2 ml of F12DIPE5. Enzyme activity was assayed in detergent extracts of cells as detailed elsewhere (20). Upon incubation of both CHO-Kl and crB cells in F12DIPE5, there was a rapid induction of HMG-CoA reductase activity during the next 16 h (Fig.  1). The enzyme activity was relatively stable between 16 and 24 h and then declined rapidly. This decrease in reductase activity was not prevented by refeeding of the cells with fresh F12DIPE5 (data not shown). In view of these findings, all measurements of HMG-CoA reductase activity were carried out in cells after 18 h of incubation in F12DIPE5 f 25-hydroxycholesterol.
We noticed that while the time course of derepression of reductase activity was very similar in the two cell lines, the in vitro specific activity of reductase in crB was 4-fold lower than that in the wildtype cells (Fig. 1). Treatment of cell extracts with E. coli alkaline phosphatase (21) or complete solubilization of reductase in a lysis buffer containing 1% Triton X-100 and a mixture of protease inhibitors (13), did not increase the assayed activity of reductase in crB. However, immunotitration of similar amounts of protein from CHO-Kl and crB cells with a monospecific, polyclonal, anti-reductase IgG revealed nearly identical equivalence points (Fig. 2). Anti-reductase antiserum was a kind gift from Dr. Peter Edwards, UCLA. Preliminary measurements of the mass of HMG-CoA reductase protein using immunoblots also revealed similar levels of enzyme protein in derepressed CHO-Kl and crB cells (data not shown). These results indicate that while the specific mass of reductase in CHO-Kl and crB cells is comparable, the enzyme from crB cells may be readily inactivated in broken cell preparations.
BiosynthesB of Fatty Acids-CHO-Kl and crB cells were grown for 24 h in F12FC5 and then incubated for 18 h in F12DIPE5 with ' OOOQ-B and crB cells. Cells were grown as per Protocol B. After 18 h of incubation in F12DIPE5, monolayers were rinsed twice in icecold saline and drained well. Cells from each culture dish were solubilized in 300 ~1 of ice-cold lysis buffer containing 50 mM potassium phosphate buffer, pH 7.0,lOO mM sucrose, 500 mM KCl, 20 mM EDTA, 1% (v/v) Triton X-100, 10 mhi dithiothreitol, 3 mM phenylmethylsulfonyl fluoride, and 0.3 mM leupeptin (13). Pooled lysates from triplicate dishes were incubated on ice for 30 min and centrifuged at 12,000 X g for 15 min. Aliquots of supernatant fraction (22 pg of protein) were incubated with indicated amounts of anti-reductase IgG at 37 "C for 30 min prior to reductase assay.
13H]acetate (lo-15 &i/ml). Cellular lipids were extracted into hexane:isopropyl alcohol (60:40) as described before (22). Lipids were saponified in 1 N KOH in methanol:benzene (80:20), and non-saponifiable lipids were extracted into hexane. The aqueous phase was aciditied with HCl and fatty acids were extracted into petroleum ether followed by diethyl ether. Pooled extracts were dried under N2 and the fatty acids were redissolved in methanol for analysis by reverse phase HPLC essentially as described before (23).
Fatty acid synthesis was also measured in uitro in Dounce homogenates of CHO-Kl and crB cells in 1 mM Tris-HCl, pH 7.0, 1 mM EGTA. Aliquots (200-300 pg of protein) were assayed for 30 min at 37 "C in a final volume of 1 ml in 0.1 M potassium phosphate, pH 7, 1 mM dithiothreitol, 50 pM [3H]acetyl-CoA (2000 dpm/nmol), 50 pM malonyl-CoA, and 2 mM NADPH. The reaction was stopped with 0.1 ml of 6 N HCl and fatty acids were extracted into petroleum ether for HPLC analysis.

Metabolism of Radiolabeled
Fatty Acids by Cells-Cells were grown for 24 h in F12FC5 and were then incubated for 24 h in FlZDIPE5 with either [3H]palmitate or with [14C]stearate (1 &i/dish) complexed to fatty acid-free BSA. The final concentration of BSA was 0.17%. Fatty acids were isolated and analyzed as described above.

Acyl-CoA:Cholesterol
Acyltransferase-The activity of cellular acyl-CoA:cholesterol acyltransferase was assayed indirectly by following the esterification of endogenous cholesterol by exogenous [3H]oleate (22).
Composition of Phospholipid Fatty Acids of Cultured Cells-Cells were grown overnight in F12FC5 and were switched to Nutridoma for 24 h. Total lipids were extracted according to Bligh and Dyer (24). Phospholipids were separated from neutral lipids by silicic acid column chromatography (25) and were saponified in 10% methanolic KOH at 65 "C for 60 min. Extracted fatty acids were methylated with diazomethane and a portion was analyzed by gas chromatography on a 10% DEGS-PS column (2 mm x 6 feet) using N, as the carrier gas at 20 ml/min. Initial column temperature was held at 165 'C for the first 4 min and was then increased at a rate of 8 "C/min to 180 "C. The response of the flame ionization detector was integrated using the System Gold software (Beckman).

Fatty
Acid Auxotrophy of crB-In the process of analyzing the nature of the regulatory defect in crB, we attempted to adapt these cells to growth in a chemically defined, lipid-poor culture medium (Nutridoma).
We found that CHO-Kl cells and the dominant 25-hydroxycholesterol-resistant mutant, CR1 cells, grew in Nutridoma medium at rates comparable to those in FlZFC5 (Fig. 3). However, crB cells grew at a much slower rate (1-2 doublings of cell population in 72 h in Nutridoma compared to 5-6 doublings in FlZFC5). Essentially similar results were obtained in the lipid-poor F12DIPE5 medium as well, although F12DIPE5 is not a good growth medium for even the wild-type cells. The rate of growth of crB in Nutridoma could be restored to the level observed in F12FC5 by the addition of 1% tissue culture grade bovine albumin (Fraction V). Supplementing with 1% albumin did not produce any stimulation of cell growth in either the wild-type CHO-Kl or the mutant CR1 cells, indicating that this observation was characteristic of the crB mutant.
Since tissue culture grade albumin contains bound fatty acids, we considered the possibility that the growth promoting ability of albumin may entirely be due to the presence of such fatty acids. We therefore tested the ability of various fatty acids to stimulate the growth of crB. Sodium salts of fatty acids (10 pM each) ranging in chain length from 14 to 20 carbons and in unsaturation from 0 to 4 double bonds were complexed to a small amount (0.03%) of fatty acid-free BSA for presentation to the cells either singly or in several combinations. Maximal growth of crB cells could be restored by supplementation of lipid-poor culture medium with a mixture of a saturated and an unsaturated fatty acid, particularly stearate (l&O) and oleate (l&l) (Fig. 3). These fatty acid supplements stimulated growth somewhat when added singly (data not shown), but complete restoration of growth was reproducibly achieved only with the combination shown. Fatty acid-free BSA did not enhance the growth of crB cells even at 1% level (data not shown).

Nature of the Biosynthetic Defect Leading to Fatty Acid
Auxotrophy in crB-To determine the site of the biosynthetic lesion of fatty acid synthesis, we incubated CHO-Kl and crB cells with radiolabeled acetate and examined the lipid products. Both cell lines incorporated similar amounts of radiola-be1 into total cellular lipids. However, the distribution of the label between non-saponifiable lipids (mostly sterols) and fatty acids was quite different in the two cell types (Table I).
Cells were seeded at 5 X 104/60-mm culture dish in F12FC5 and incubated for 4 h. The culture medium was then changed to Nutridoma containing either no addition (A), 1% tissue culture grade albumin (B), 0.03% fatty acid-free BSA (C), or 10 PM each of stearate and oleate complexed to 0.03% BSA (D). Cells were harvested after 3 days by trypsinization and counted. The results are presented as a percentage + SE. of the cell number for cells maintained throughout in F12FC5 (1.75 x 106, 1.32 x 106, and 1.94 x lo6 cells/dish for CHO-Kl, crB, and CRl, respectively).
Incorporation of [3H]acetate into total fatty acids in crB was 4-fold lower than that in CHO-Kl while that into nonsaponifiable lipids was 3-fold greater. This difference in the distribution of radiolabel persisted (data not shown) over a wide range of exogenous acetate concentrations (1 /zM to 4 mM), suggesting that the acetate pool sizes in the two cell lines were not very different (26,27). Analysis of the radiolabeled non-saponifiable lipids of CHO-Kl and crB by thin layer and high performance liquid chromatography (20) revealed that >95% of the radiolabel migrated as C&sterols, with cholesterol accounting for 50 and 85% of the total in CHO-Kl and crB, respectively. As expected, treatment of CHO-Kl cells with 25-hydroxycholesterol caused a 5-fold inhibition of the rate of sterol synthesis (Table I). A similar treatment of crB cells resulted in a much smaller degree of inhibition such that the rate of synthesis of sterols even in oxysterol-treated cells was greater than that in untreated CHO-Kl cells (Table I).
A comparison of the fatty acids formed from exogenous acetate in CHO-Kl cells and crB cells (Fig. 4A) indicated that, whereas in wild type cells substantial amounts of stearate (l&O) and oleate (l&l) are formed, in crB cells there did not appear to be any elongation of palmitic acid (160). Instead, a significant accumulation of radioactivity in myristate (14:0) could be observed. Examination of the specific activities of acetyl-CoA carboxylase and fatty acid synthetase by direct enzyme assay in cell-free homogenates indicated that these enzyme activities were present at near normal levels in crB (data not shown). When the products of the in vitro fatty acid synthetase reaction were analyzed (Fig. 4B), the myristate accumulation seen in intact crB cells was not observed. The absence of fatty acids of chain length longer than 16 in assays with CHO-Kl homogenates is explained by the short incubation period employed (30 min).
The lack of formation of stearate and oleate observed in intact crB (Fig. 4A) is most consistent with a defect in fatty acid elongation reactions. To test this possibility, we examined the incorporation of labeled palmitic acid into products in CHO-Kl cells and crB. The results (Fig. 5A) demonstrate that exogenously added palmitate could not be converted to stearate in crB although this process could clearly be observed in the wild-type CHO-Kl cell. The data also indicate that the source of myristate in crB cells is not p-oxidation of palmitate. These observations are consistent with a defect in the chain elongation reaction that adds a two-carbon unit onto palmitic acid. Also, in both crB and CHO-Kl cells exogenous stearic acid was converted to oleic acid (Fig. 5B). The extent of desaturation of cell-associated [14C]stearate ranged from 47 to 52% in CHO-Kl cells and 33 to 41% in crB cells. While this experiment was not designed to yield quantitative results, these data do demonstrate that crB cells are capable of fatty acid desaturation at a level not very different from that in and crB (lower panels).

Intact cells in F12DIPE5 (A) or Dounce homogenates of cells (B) were incubated with [3H]acetate or
[3H]acetyl-CoA, respectively, and the labeled fatty acids were isolated as described under "Materials and Methods." Individual fatty acids were resolved by reverse phase HPLC on an Alltech Versapak Cls column (4.1 X 300 mm) equilibrated in 58% acetonitrile in 14.7 mM phosphoric acid at 2 ml/min. The concentration of acetonitrile was increased to 61% over a I-min period starting at 30 min and then to 83% over a 4-min period starting at 45 min. The column effluent was monitored by an on-line Flo-one p radioactivity detector (Radiomatic). Peaks were identified by their coelution with authentic standards. 14:0, myristate; 16:0, palmitate; 18:0, stearate; 18:1, oleate. Cells in F12DIPE5 (Protocol B) were incubated for 24 h with 1 PCi of either [3H]palmitate (A), or ['"Clstearate (B) complexed to 0.17% BSA, and labeled fatty acids were isolated as described under "Materials and Methods." Individual fatty acids were resolved by reverse phase HPLC as described in Fig. 4 (Table II) indicate that like CRl, crB is also pleiotropically defective in oxysterol-mediated regulation of mRNA levels for enzymes of sterol biosynthesis.
We also examined the effect of 25-hydroxycholesterol on the rates of synthesis and degradation of HMG-CoA reductase in crB and CHO-Kl cells. The results (Table III) demonstrate that in contrast to the findings in wild-type cells, 25-hydroxycholesterol is a poor regulator of the synthesis and degradation of this enzyme in crB cells. These observations, including the constitutively rapid rate of HMG-CoA reductase degradation in crB are similar to the previously reported findings with CR1 (13).
Relationship between Fatty Acid Auxotrophy and Defective Regulation of Cholesterogenic Enzymes by 25Hydroxycholesterol-These observations immediately raised the question of whether the defective regulation of cholesterol synthesis by exogenous 25-hydroxycholesterol in crB arises from the same genetic defect responsible for the fatty acid auxotrophy. In order to examine this question, we first attempted to isolate revertants of crB to fatty acid prototropy by selection for growth in lipid-poor medium after mutagenesis. Isolation of such mutants was achieved by standard methods (15) and the ability of the putative revertants to synthesize long chain fatty acids was examined. The results with one such revertant, crB R7, clearly demonstrate synthesis of stearate and oleate (Fig. 6) similar to that observed with the wild-type cells (Fig.  4A). The putative revertants were then analyzed for the regulatory response of HMG-CoA reductase activity to 25hydroxycholesterol.
The results (Fig. 7A) indicate that in the two revertants examined, regulation of HMG-CoA reductase activity by 25-hydroxycholesterol was restored to the same levels as that found in parental CHO-Kl cells. This finding clearly establishes a genetic link between fatty acid elongation and sterol-mediated regulation of isoprenoid biosynthesis. The capacity of exogenous fatty acid supplements to restore regulation of HMG-CoA reductase by 25-hydroxycholesterol in the crB mutant was also tested (Fig. 7B). With a mixture of fatty acids (oleate and stearate) that enhanced the growth of crB in lipid-poor medium, regulation of HMG-CoA reductase activity by 25-hydroxycholesterol was essentially restored. The restoration of sterol-mediated regulation by fatty acid supplementation of crB culture medium could be shown to occur at the level of reductase synthesis (Table IV). When added singly, oleate restored the regulation partially and stearate was ineffective (Fig. 7B). Single supplements of myristate, palmitate, linoleate, or arachidonate were also incapable of restoring oxysterol regulation of HMG-CoA reductase (data not shown). We also tested the possibility that the peculiar accumulation of myristate in crB cells (Fig. 4A)    Cells were grown and treated as described in Table II. The relative rates of synthesis and the rates of degradation of HMG-CoA reductase polypeptide were determined following [YS]methionine labeling as described under "Materials and Methods." The rate of synthesis is expressed as a ratio (X103) + S.E. of 35S counts/min in reductase to those in total proteins determined as trichloroacetic acid-precipitable material and is the average of triplicate cultures. The numbers in the parentheses show normalized results taking the control values to be loo. The revertant crB R7, capable of growth in Nutridoma, was labeled with [aH]acetate as described under "Materials and Methods." Isolated fatty acids were resolved by HPLC as described in Fig. 4. cells with hexadecynoate caused a substantial abolition of the synthesis of fatty acids longer than palmitate (Fig. 8), resulting in a phenotype quite similar to that of crB. As shown in Table V   Cells were grown and treated as described in Table II except that the culture medium for one set of crB dishes was supplemented with 10 PM each of stearate and oleate complexed to BSA (0.03% final concentration). The cells were labeled with [YZl]methionine for 1 h before immunoprecipitation of HMG-CoA reductase as described under "Materials and Methods." The rates of synthesis are expressed as a ratio (~10~) of "S counts/min in reductase to those in total proteins determined as trichloroacetic acid-precipitable material. The results are the mean + S.E. of triplicate cultures. of a specific inhibitor, appears to cause a loss of regulation of reductase by sterols. Specificity of the Regulatory Defect in crB-Since the experiments on the regulation of HMG-CoA reductase in crB were conducted on cells incubated for 18-24 h in lipid-poor media, we considered the possibility that growth arrest resulting from fatty acid starvation may have played a part in the observed loss of regulation. However, as shown in Fig. 9, fatty acid depletion had no adverse effect on the rate of total protein synthesis in crB cells for up to 24 h. It may be pointed out that in this experiment as well as in other experiments of short duration (~24 h) investigating the regulation of HMG-CoA reductase activity, the initial cell density was 6-fold higher than that used in cell growth experiments (Fig. 3). The rate of total protein synthesis in crB under these culture conditions did decline upon longer incubation in F12DIPE5 (data not shown). It should be noted that the derepressed rates of synthesis of HMG-CoA reductase protein are also very similar in crB and CHO-Kl (Tables III and IV). In addition, treatment of crB with mevalonate produced a dosedependent suppression of reductase activity which was only slightly smaller than that seen in CHO-Kl cells (Fig. 10). This finding suggests that the reductase activity in crB can indeed be down regulated in the presence of suitable regula- CHO-Kl cells, continuously grown in Nutridoma, were seeded in the same medium at 3 x 106/60-mm culture dish. The next day, they were refed fresh Nutridoma medium without (A) or with (B) 20 FM hexadecynoate complexed to 0.06% BSA. After 1 h, [3H]acetate (15 &i/ml) was added and the cells were incubated for 18 h. Fatty acids were isolated and analyzed by reverse phase HPLC as described in Fig. 4. tors. The small difference in the magnitude of inhibition of reductase activity by mevalonate between the two cell lines can be explained as due to the loss of the sterol component of inhibition in crB. We have previously reported (20) that specific blockage of the pathway of sterol synthesis in CHO-Kl cells results in a similar minor loss of inhibition of reductase activity by mevalonate. Thus, these results indicate that the component of regulation mediated by non-sterol products of mevalonate remains intact in crB cells starved for fatty acids. We have observed that incubation of crB cells in F12DIPE5 beyond 18-20 h resulted in progressive loss of DNA synthesis as measured by the incorporation of [3H]thymidine (Fig. 11). The extent of inhibition at 20 h was 25-30% compared to controls supplemented with fatty acids. However, in other CHO cell lines in which we caused a similar degree of inhibition of DNA synthesis such as in CHO-Kl cells treated with 0.2 pM aphidicolin (31) or in the inositol auxotroph Ino-A3 cells' starved for myo-inositol, regulation of reductase activity by 25-hydroxycholesterol was not compromised (Table VI) and crB phospholipids by gas chromatography after a 24-h incubation of both cell types in lipid-poor medium also did not reveal any obvious differences in the pattern of fatty acids between the two cells (Fig. 13). The ratio of palmitate (l&O) to staarate (l&O) plus oleate (l&l) in total phospholipids was thus unchanged by this incubation. DISCUSSION We have previously reported that crB is one of two comple-  occurs through control of its synthesis by sterols via transcriptional (6)(7)(8) and translational (32, 33) mechanisms. In addition, there is clear evidence that HMG-CoA reductase is also regulated through control of its rate of degradation (34). Therefore, it is not surprising that as we have previously reported for a dominant 25hydroxycholesterol-resistant mutant CR1 (13), and currently report for a recessive mutant crB, resistance is manifested as defective regulation of mRNA levels encoding enzymes known to be transcriptionally regu- lated and of degradation of the rate-limiting enzyme. The major conclusions of the current report are that there is a total absence of fatty acid chain elongation of palmitate to stearate in crB cells and that this defect is probably responsible for the loss of regulation by 25-hydroxycholesterol of mRNA levels of several enzymes of sterol biosynthesis and of the synthesis and degradation of HMG-CoA reductase. The defect in fatty acid elongation is manifested as loss of capacity to convert radioactive acetate to stearate and oleate relative to wild-type cells, loss of capacity to convert exogenous palmitate to stearate relative to wild-type cells, and the resultant fatty acid auxotrophy.
Single-step revertants of crB to fatty acid prototrophy regain both the capacity to synthesize stearate and oleate and the capacity to be regulated by 25-hydroxycholesterol.
Interestingly, incubation of crB cells in medium supplemented with stearic and oleic acids results in restoration of regulation of the activity as well as the synthesis of HMG-CoA reductase by 25-hydroxycholesterol to a wild-type response. The link between long chain fatty acids and sterol regulation of cholesterogenic enzymes is further demonstrated by the fact that resistance to regulation by 25-hydroxycholesterol can be generated in the wild-type cells by treatment with hexadecynoate, a specific inhibitor of fatty acid elongation. Some tentative conclusions regarding the mechanisms by which fatty acids may be regulating HMG-CoA reductase can also be drawn from the data reported in this manuscript. We have determined that under the conditions in which regulation of HMG-CoA reductase activity by 25-hydroxycholesterol is compromised there is no gross change in the fatty acid composition of cellular phospholipids. This result implies that the regulatory response is produced by fatty acids in a labile pool. We have also noted that the regulatory response to 25hydroxycholesterol in crB is restored to near normal by fatty acids corresponding to those which are normal components of bulk lipids of cells. Some known signaling mechanisms that might involve fatty acids such as post-translational modification by acylation with myristate or palmitate, or through oxidation of arachidonic acid do not, therefore, seem to be responsible for the regulation observed in this case. Neither myristate, palmitate, linoleate nor arachidonate when given alone were able to restore regulation of HMG-CoA reductase activity in crB. There is some prior evidence of the effects of fatty acids on hepatic cholesterol biosynthesis (35-41). The fatty acid composition of dietary lipids is reported to affect the rate of cholesterol biosynthesis measured either as ['"Clacetate incorporation or as the activity of HMG-CoA reductase (35-37). Dietary stearate is found to stimulate cholesterogenesis in rat liver more than other fatty acids tested (38). Stimulation of hepatic HMG-CoA reductase activity by dietary fatty acid is reported to be proportional to the chain length and inversely related to the degree of unsaturation. In perfused rat liver, however, oleate caused the most pronounced rise of reductase (39,40). Treatment with oleic acid is also reported to enhance HMG-CoA reductase activity of rat hepatocytes in culture by 2-3-fold (41, 42). Other fatty acids such as palmitate, stearate, linoleate and linolenate were much less effective than oleate. Arachidonate treatment of hepatocytes actually led to a small (33%) decrease in reductase activity (42). Oral administration of olive oil is reported to enhance the hepatic activities of several other enzymes of cholesterol biosynthesis (43).
We have noted that the specific activity of HMG-CoA reductase in crB incubated in lipid-poor medium was 4-5-fold lower than that observed in CHO-Kl cells, and was stimulated by fatty acids. The lower in vitro activity of reductase is inconsistent. with the observations of higher mRNA levels for HMG-CoA reductase in crB than in CHO-Kl, an equivalent rate of synthesis of the enzyme, and comparable amounts of immunotitratable mass (see Tables II-IV and Fig. 2). In addition, the conversion of [3H]acetate into non-saponifiable lipids was 2-4-fold higher in crB compared to that in CHO-Kl over a wide range of exogenous acetate concentrations (Table I). These data are consistent with the conclusion that while the amount of HMG-CoA reductase protein in crB cells is comparable to that in the wild-type cells, the enzyme from the mutant cells may be rapidly inactivated upon cell lysis. It should, however, be pointed out that despite the lower specific activity of HMG-CoA reductase in crB lysates, the magnitude of the changes in the measured activity upon treatment of cells with 25-hydroxycholesterol is consistent with the changes observed in the levels of reductase mRNA and the rate of enzyme synthesis. Thus, measurement of reductase activities has proven to be a reliable preliminary measure of the resistance of crB cells to 25-hydroxycholesterol.
The crB mutant provides a system in which to examine several questions related to the enzymology of fatty acid chain elongation. The chain length specificity of fatty acid elongating enzymes and their subcellular distribution have been difficult to analyze by biochemical procedures alone. The crB mutation should allow us to answer the question of whether or not there are fatty acid elongating enzymes with different chain length specificities and whether the functional site of fatty acid elongation, at least in CHO cells, is mitochondrial or microsomal.