Complementation of mutation in acyl-CoA:cholesterol acyltransferase (ACAT) fails to restore sterol regulation in ACAT-defective sterol-resistant hamster cells.

A previously described mutant line of Chinese hamster ovary cells, designated SRD-4, fails to synthesize cholesteryl esters, owing to a deficiency in the activity of acyl-CoA:cholesterol acyltransferase (ACAT). These cells also fail to suppress low density lipoprotein receptors or cholesterol synthesizing enzymes in the presence of 25-hydroxycholesterol. In the current studies we show that SRD-4 cells have three defects: 1) a point mutation in one allele at the ACAT locus that changes codon 265 from Ser to Leu, resulting in an inactive enzyme; 2) a silent allele at the other ACAT locus that does not produce detectable mRNA; and 3) a mutation, as yet undefined, that abolishes the ability of 25-hydroxycholesterol to inhibit the cleavage of both sterol regulatory element binding proteins (SREBP-1 and SREBP-2). Correction of the ACAT deficiency by transfection of a wild-type cDNA failed to restore inhibition of SREBP cleavage by 25-hydroxycholesterol, indicating that the ACAT deficiency and the sterol regulatory defect are caused by independent mutations. These data provide further insight into the interplay between ACAT activation and inhibition of SREBP cleavage by 25-hydroxycholesterol, and they indicate that these two processes can be disrupted independently by mutation.

The endoplasmic reticulum (ER) 1 enzyme acyl-CoA:cholesterol acyltransferase (ACAT) esterifies excess cholesterol in animal cells for storage as cholesteryl ester droplets (1). When animal cells are depleted of cholesterol, they do not synthesize appreciable amounts of cholesteryl esters, owing in part to a lack of cholesterol substrate in the ER. When cholesterol enters cells in the core of low density lipoprotein (LDL), the rate of cholesteryl ester synthesis increases markedly (2). This is due to the provision of cholesterol substrate and possibly to an activation of the enzyme's catalytic capacity. New protein synthesis is not required (1,2). ACAT activity also increases when cells are treated with oxysterols such as 25-hydroxycholesterol (3). 25-Hydroxycholesterol itself is esterified, and it also stimulates the esterification of cholesterol, indicating that oxysterols somehow make cholesterol more available to the enzyme (3).
A major step in the molecular understanding of ACAT came from the cDNA cloning of the human enzyme by Chang et al. (4). These workers used a clever selection technique to obtain a line of Chinese hamster ovary (CHO) cells, designated AC29, that is deficient in ACAT activity and therefore fails to accumulate cholesteryl ester droplets when incubated with LDL or 25-hydroxycholesterol (5). They were able to reverse this defect by transfection of human genomic DNA into the cells (6). This eventually led to the isolation of a human cDNA encoding ACAT (4). The protein turned out to be hydrophobic with areas of homology to other acyltransferases. Expression of this protein in insect Sf9 cells, which do not themselves express ACAT, yielded a high level of ACAT activity, confirming that the cDNA encoded the catalytic component (7).
The AC29 cell line that Chang et al. (4) used as a recipient in the cloning studies has two defects, 1) it lacks the ACAT enzyme (8); and 2) it is resistant to feedback repression of cholesterol biosynthesis and LDL receptor activity by 25-hydroxycholesterol (5,6). These two defects arose independently. The cells were first selected for 25-hydroxycholesterol resistance, and they were subsequently selected for ACAT deficiency (5). The molecular nature of the mutation that abolishes ACAT expression and the nature of the defect that leads to a failure of 25-hydroxycholesterol regulation in AC29 cells are unknown.
Our laboratory also isolated a mutant CHO cell line, designated SRD-4 cells, with the same two defects, 1) lack of ACAT activity; and 2) resistance to 25-hydroxycholesterol-mediated feedback regulation (9). Although the SRD-4 cells emerged from a single step selection, we concluded that they also had two independent mutations because treatment of wild-type cells with an ACAT inhibitor blocked cholesterol esterification, but it did not reproduce the 25-hydroxycholesterol resistance phenotype (9). 25-Hydroxycholesterol represses cholesterol biosynthesis and LDL receptor activity at the transcriptional level by regulating a pair of proteins designated sterol regulatory element binding protein-1 and -2 (SREBP-1 and -2) (Refs. 10 -12). These proteins are synthesized as integral components of the membranes of the ER and nuclear envelope (12)(13)(14). In sterol-depleted cells, a protease clips each protein to release an NH 2terminal fragment of ϳ500 amino acids that contains a basichelix-loop-helix-leucine zipper motif and a transcription activating domain (12)(13)(14). This fragment, designated the "mature" form of SREBP, enters the nucleus and binds to sterol regulatory elements in the enhancer regions of the genes encoding 3-hydroxy-3-methylglutaryl-CoA synthase, an early enzyme of cholesterol biosynthesis, the LDL receptor (10 -13), and other enzymes of sterol biosynthesis (15). Binding leads to transcriptional activation, which allows cells to increase their rate of de novo cholesterol synthesis and uptake of cholesterol from LDL through LDL receptors (10 -13). SREBP-1 also activates transcription of the genes encoding acetyl-CoA carboxyl-ase and fatty acid synthetase, two enzymes of fatty acid biosynthesis (16,17). When LDL-derived cholesterol or 25hydroxycholesterol overaccumulates in cells, the proteolysis of SREBPs is reduced, the proteins remain membrane-bound, and transcription of the target genes declines (10 -14).
We previously described three mutant lines of CHO cells, designated SRD-1, -2, and -3 cells, that constitutively transcribe the genes encoding enzymes of cholesterol biosynthesis and the LDL receptor (18 -21). These cells show virtually no suppression when 25-hydroxycholesterol is added. In contrast to the SRD-4 cells, the SRD-1, -2, and -3 cells have normal ACAT activity that is stimulated normally by LDL or 25hydroxycholesterol (20, 21).
The SRD-1, -2, and -3 cells have each undergone genomic recombinations that yield rearranged mRNAs encoding a truncated form of SREBP-2 that terminates before the membrane attachment domain (18,19). Because the truncated SREBP-2 is never attached to the membrane, it does not require proteolysis, and, therefore, it is always active; 25-hydroxycholesterol cannot suppress its activity.
The SRD-4 cells, like the SRD-1, -2 and -3 cells, show nonregulated expression of the enzymes of cholesterol biosynthesis and the LDL receptor (9). We do not know whether the defect in these cells is caused by the production of a truncated form of SREBP. Moreover, we do not know the reason why these cells, in contrast to the SRD-1, -2 and -3 cells, lack ACAT activity.
In the current experiments we have used the human cDNA clone of Chang et al. (4) to isolate a cDNA encoding wild-type hamster ACAT. We have found that the SRD-4 cells produce a form of ACAT with a point mutation that changes a conserved amino acid and abolishes activity of the enzyme. We also show that the SRD-4 cells do not produce a truncated form of SREBP; rather, they continue to process both SREBP-1 and -2 to mature nuclear forms even in the presence of 25-hydroxycholesterol. Finally, we show that correction of the ACAT defect by transfection with wild-type ACAT fails to correct the regulatory response to 25-hydroxycholesterol in SRD-4 cells, confirming that these two abnormalities arose from independent mutations.
Blot Hybridization of RNA and DNA-Poly(A ϩ ) RNA was isolated with oligo(dT) cellulose (Stratagene) and loaded onto a 1.5% agarose gel in 40 mM MOPS. After electrophoresis at 50 mA for 4 h at room temperature, the RNAs were transferred overnight onto Hybond-N membranes (Amersham Corp.) in 20 ϫ SSC, cross-linked with UV light, prehybridized for 30 min, and hybridized for 2 h at 65°C in Rapid-hyb Buffer (Amersham Corp.) with a 32 P-labeled 1.6-kb PCR fragment corresponding to the coding region of the hamster ACAT cDNA (1 ϫ 10 6 cpm/ml). The blot was washed at room temperature with 2 ϫ SSC, 0.1% (w/v) SDS for 20 min, and 0.5 ϫ SSC/0.1% SDS twice at 65°C for 20 min, followed by autoradiography.
Genomic DNA was isolated from CHO-7 and SRD-4 cells with a DNA Extraction Kit (Stratagene). Aliquots of DNA (10 g) were digested with restriction enzymes at 37°C for 2 h and subjected to electrophoresis on a 0.7% agarose gel. The DNAs were transferred to Hybond-Nϩ membranes, cross-linked with UV light, and blotted with 1 ϫ 10 6 cpm/ml of the 32 P-labeled 1.6-kb hamster ACAT probe (see above) in Rapid-hyb Buffer. The blot was then washed twice with 0.2 ϫ SSC at 42°C, followed by autoradiography.
PCR, SSCP Analysis, and Localization of Mutation in ACAT-To detect mutations in the hamster ACAT mRNA in SRD-4 cells, poly(A ϩ ) RNA from SRD-4 and CHO-7 cells was reverse-transcribed with the Stratascript TM reverse transcription-PCR Kit (Stratagene). The resulting RNA/DNA hybrids from each cell line were used as templates with [ 32 P]dCTP in eight separate PCR reactions that were designed to cover the entire 1.6-kb coding region of hamster ACAT. The resulting eight 32 P-labeled PCR products spanning the ACAT cDNA in both CHO-7 and SRD-4 cells were then subjected to SSCP analysis as described (25).
One of the PCR products in the region of codon 265 gave an abnormal pattern in the SSCP analysis. To localize the mutation in this region, we subjected SRD-4 mRNA to PCR with primers from the 5Ј-and 3Јuntranslated regions of the ACAT mRNA. A specific primer corresponding to codons 201-208 was used to sequence this region of the PCR product.
Site-directed Mutagenesis-In vitro mutagenesis was done with a Muta-gene M13 In Vitro Mutagenesis Kit (Bio-Rad). A point mutation (Ser 3 Leu) at amino acid residue 265 in the wild-type ACAT cDNA was produced using a 27-base pair mutagenic oligonucleotide, 5Ј-CTCTCT-GACAAACAAGTGAGCCTTCAT-3Ј. The replication form DNA of the mutant cDNA (cloned in M13mp 19) was digested with SalI and NotI, and the insert was subcloned into pRc/CMV7SB (13). This plasmid was designated pCMV-ACAT(S265L). The mutation was confirmed by DNA sequencing.
Transient Transfection of 293 Cells-Monolayers of human embryonic kidney 293 cells were set up on day 0 at 4 ϫ 10 5 cells/60-mm dish in Dulbecco's modified Eagle's medium with low glucose supplemented with 100 units/ml penicillin, 100 g/ml streptomycin, and 10% (v/v) fetal calf serum. After incubation for 48 h at 37°C in a 5% CO 2 incubator, the cells were transfected with the wild-type or mutant pCMV-ACAT plasmid using a MBS Transfection Kit (Stratagene) according to the manufacturer's instructions. After incubation for 3 h at 35°C in a 3% CO 2 incubator, the cells were washed once with phosphate-buffered saline, refed with fresh medium containing 10% newborn calf lipoprotein-deficient serum (medium A), returned to a 37°C/ 5% CO 2 incubator, and used for experiments 16 h later.
Stable Transfection of SRD-4 Cells-Cells were seeded on day 0 at a density of 5 ϫ 10 5 cells/60-mm dish in Dulbecco's modified Eagle's medium/F-12 medium supplemented with 100 units/ml penicillin, 100 g/ml streptomycin, and 5% newborn calf lipoprotein-deficient serum and transfected on day 1 with 5 g of hamster wild-type pCMV-ACAT using the MBS Transfection Kit as described above. Stable transformants from two dishes were selected by growth in medium supplemented with 700 g/ml G418. G418-resistant colonies were picked and screened for incorporation of [ 14 C]oleate into cholesteryl [ 14 C]oleate by intact monolayers (see below). Positive colonies were cloned by dilution plating and rescreened. Four clones that stably expressed ACAT were isolated, and one of these, designated Tr. 10 -9, was used for experiments.
Immunoblot Analysis-Protein concentrations were measured with a BCA kit (Pierce). After SDS electrophoresis in a 12% gel, the proteins were transferred to Hybond-C extra transfer membrane (Amersham Corp.). Immunoblot analysis was carried out with Enhanced Chemiluminescence Western blotting Detection Kit (Amersham Corp.) according to the manufacturer's instructions with minor modifications (14,23). Filters were exposed to Reflection TM NEF film (DuPont).
ACAT Assays-The rate of incorporation of [ 14 C]oleate into cholesteryl [ 14 C]oleate and [ 14 C]triglycerides by intact cell monolayers was measured as described previously (24).
ACAT activity in membranes from cell extracts was assayed by measuring the rate of conversion of [1-14 C]oleoyl-CoA to cholesteryl [ 14 C]oleate as described previously (3,9) with several modifications. The 2,000 ϫ g cell pellets were resuspended in buffer A (50 mM potassium phosphate and 2 mM dithiothreitol at pH 7.4), homogenized by 10 passages through a 25-gauge needle, and subjected to centrifugation at 10 5 ϫ g for 1 h at 4°C to yield a total cell membrane fraction. This fraction was suspended in buffer A. Membrane fractions (25-100 g of protein) were incubated in a final volume of 0.2 ml for 20 min at 37°C in buffer A containing 5 mg/ml bovine serum albumin, followed by incubation for 20 min at 37°C with 10 g/ml cholesterol (added in 2 l of ethanol). [ 14 C]Oleoyl-CoA (20 dpm/pmol) was then added to a final concentration of 75 M, and the incubations were continued for 1 h at 37°C. Reactions were terminated by addition of chloroform/methanol (1:1, v/v) continued 20,000 dpm [ 3 H]cholesteryl oleate as a recovery control. Neutral lipids were extracted and resolved by thin layer chromatography as described previously (24). RESULTS We isolated a cDNA encoding hamster ACAT by probing a gt22A cDNA library prepared from SRD-2 cells (13), a line of CHO cells that has high ACAT activity, owing to overproduction of cholesterol as a result of a dominant positive mutation in SREBP-2 (19). Fig. 1 shows that the amino acid sequence of hamster ACAT, as deduced from the cDNA sequence, is 88 and 92% identical to the human and mouse homologues, respectively. The hamster and mouse proteins are each slightly shorter than the human sequence (546 and 540 amino acids, respectively, versus 550 amino acids), owing to deletions of blocks of four or six residues. Fig. 2 shows a Northern blot showing that the wild-type CHO-7 cells and the SRD-4 cells each produced a single ACAT mRNA of ϳ3 kb (Fig. 2). The relative abundance was similar in the two cell lines, but a 2-fold difference could not have been detected by this technique.
To search for a point mutation in the ACAT mRNA in SRD-4 cells, we made a cDNA copy of the mRNA with reverse transcriptase, and then we used oligonucleotide primers with [ 32 P]dCTP to amplify by PCR eight overlapping segments that covered the entire 1.6-kb coding region. The PCR products were denatured and subjected to electrophoresis in nondenaturing gels so as to detect single-strand conformational polymorphisms (SSCPs) (25). Seven of the amplified fragments were identical for the cDNAs from CHO-7 and SRD-4 cells (data not shown). The eighth fragment consistently showed a slight mobility difference between the two cell strains (Fig. 3). After denaturation, both strands of the SRD-4 cDNA (lane 4) were displaced relative to the CHO-7 strands (lane 3), and there was no evidence of any wild-type strands.
To identify the putative point mutation in the SRD-4 ACAT, the entire coding region of the mRNA was PCR-amplified, and the DNA sequence in the region of the abnormal SSCP was determined. This sequence showed a single nucleotide abnormality (C 3 T) at the central position of codon 265, changing the amino acid from Ser to Leu. The sequence of the PCR product showed only a T and no C at this position, indicating that the SRD-4 cells produced only the abnormal species of mRNA and none of the normal transcript (data from automated sequencer not shown). It is noted that Ser-265 is conserved in ACAT from all three known species (asterisk in Fig. 1).
To confirm the mutation in the SRD-4 mRNA, we cloned the entire PCR-amplified coding region of the mRNA into a TA vector and isolated 9 independent recombinant plasmids ( Table  I). The DNA sequence of each of these plasmids was determined in the region of codon 265, and all of them showed the C 3 T change.
The data from the SSCP analysis (Fig. 3), the direct sequencing of the PCR product, and the subsequent analysis of cDNA clones (Table I) all indicated that the SRD-4 cells produced only the mutant form of the mRNA. To analyze the situation at the genomic level, we PCR-amplified the region of the gene containing codon 265 and cloned the products into a plasmid vector. Four independent colonies were isolated, and the plasmids were subjected to sequencing. Two of these plasmids showed the wild-type sequence at codon 265 (TCG), and the other two showed the mutant sequence (TTG) ( Table I). We conclude that the SRD-4 cells are compound heterozygotes. One allele contains a C 3 T substitution at codon 265 and produces a mutant ACAT mRNA. The second allele contains a wild-type sequence at codon 265, but it is a silent allele that is not transcribed into mRNA. Southern blots of genomic DNA, probed with the full-length cDNA, failed to reveal an abnormal fragment in the SRD-4 gene after digestion with several restriction enzymes, indicating that the silent allele does not have a gross deletion or rearrangement (data not shown).
Using techniques of in vitro mutagenesis, we introduced the C 3 T mutation at codon 265 into the full-length wild-type cDNA (S265L mutant). To visualize the ACAT protein, we prepared a polyclonal rabbit antibody against a fusion protein containing amino acids 1-140 of hamster ACAT. Fig. 4 shows that this antibody failed to detect a protein in the size range of ACAT (50 kDa) (8) on immunoblotting of extracts of human 293 cells when the cells were incubated either in the absence or presence of sterols (lanes 1 and 2, respectively). When the 293 cells were transfected with a cDNA encoding wild-type hamster ACAT under control of the CMV promoter, a band in the correct size range was detected, and this was not altered by the addition of sterols (lanes 5 and 6). Transfection of the mutant cDNA produced a much lighter band (lanes 3 and 4). This experiment was replicated several times, and we interpreted it to indicate that the S265L mutation renders the protein unstable.
To determine whether the S265L mutant has enzymatic activity, we transfected 293 cells with 5 g of mutant ACAT plasmid and 1 g of wild-type plasmid so as to obtain approximately the same amount of protein (see immunoblot in Fig. 5). The cells were lysed, a membrane pellet was prepared, and ACAT activity was measured with varying amounts of membranes using [ 14 C]oleoyl-CoA and unlabeled cholesterol as substrates. Membranes from 293 cells transfected with an empty vector had endogenous ACAT activity (open triangles, Fig. 5). Transfection with 1 g of wild-type ACAT cDNA led to a severalfold increase in enzyme activity (closed symbols). In contrast, transfection with 5 g of the mutant cDNA produced no detectable increase over the endogenous activity (open triangles). Similar results were obtained in four separate experiments. We conclude that the S265L mutant of hamster ACAT is both unstable and inactive.
We next conducted studies to determine whether the ACAT defect in SRD-4 cells causes the defect in sterol-mediated inhibition of the cleavage of SREBPs and whether this latter defect would be corrected by correction of the ACAT defect. For this purpose, we transfected the wild-type ACAT cDNA into SRD-4 cells under control of the CMV promoter and isolated a permanent line of SRD-4 cells, designated Tr.10 -9, that expresses the wild-type enzyme. Table II shows 1 and 3) and SRD-4 (lanes 2 and 4) cells were incubated with oligonucleotide primers corresponding to amino acid residues 218 -226 and 269 -275 in the wild-type hamster ACAT cDNA. The sequences of these two primers, respectively, are 5Ј-GCCACG-GCTTTTTCTTCTTAGTCTT-3Ј and 5Ј-TAGTACTCTAGGTACATTC-TC-3Ј. PCR was carried out in the presence of [ 32 P]dCTP for 30 cycles at 96°C for 1 min/66°C for 5 min. Aliquots of the PCR products were denatured at 95°C for 5 min in formamide and applied to a 6% glycerol polyacrylamide gel (lanes 3 and 4) adjacent to aliquots of the nondenatured samples (lanes 1 and 2). The gel was subjected to electrophoresis at 300 V for 16 h at room temperature, dried, and exposed to Reflection TM NEF film for 12 h at room temperature.

TABLE I Clonal analysis of PCR products amplified from ACAT mRNA and
genomic DNA from SRD-4 cells Poly(A ϩ ) RNA from SRD-4 cells was reverse-transcribed and used as a template in a PCR reaction containing primers from the 5Ј-and 3Ј-untranslated regions of the ACAT cDNA. The PCR product was cloned into the TA vector. Nine colonies were isolated, and the region surrounding codon 265 was sequenced. Genomic DNA from SRD-4 cells was used as a template in a PCR reaction containing primers corresponding to amino acid residues 218 -226 and 267-275 in the hamster ACAT cDNA. The resulting ϳ1000-base pair PCR product contained an ϳ800-base pair intron situated between the codons for amino acids 256 and 257. The PCR products were cloned into TA vector, and random clones were sequenced using primers derived from the cloning vector.  (Table II).
To assess the sterol-regulated cleavage of SREBPs, we incubated cells in the presence of lipoprotein-deficient serum plus varying concentrations of 25-hydroxycholesterol for 16 h. Nuclear extracts were prepared and subjected to electrophoresis and immunoblotting with antibodies against SREBP-1 or -2 (Fig. 6). CHO-7 cells incubated in the absence of lipoproteins had detectable amounts of SREBP-1 and -2 in the nucleus (lane 1, upper and lower panels). The size of the protein was consistent with the known size of the transcriptionally active NH 2terminal fragment. Addition of 0.1 g/ml 25-hydroxycholesterol markedly reduced the amount of this nuclear fragment, and the protein virtually disappeared at 0.3 g/ml (lanes 2 and   3). The SRD-4 cells also exhibited nuclear forms of SREBP-1 and -2, but neither protein was reduced when the cells were incubated with 25-hydroxycholesterol at concentrations as high as 1 g/ml, which was 10-fold higher than the concentration that produced a detectable decrease in the CHO-7 cells (lanes [5][6][7][8]. This defect in sterol-mediated regulation persisted in the Tr.10 -9 cells (lanes 9 -12).
The experiment of Fig. 6 indicates that the loss of sterolmediated regulation of SREBP processing in SRD-4 cells is not caused by the loss of ACAT activity. However, the experiment does not rule out the possibility that the regulatory defect is caused by some dominant property of the mutant ACAT enzyme. To rule out this possibility, we introduced expressible cDNAs encoding wild-type or S265L mutant ACAT into 293 cells by transient transfection together with expression plasmids encoding epitope-tagged versions of SREBP-1 or -2. The cells were incubated in the absence or presence of sterols (Fig.  7). Nuclear extracts were subjected to SDS-PAGE and immunoblotted with antibodies against the epitope tag on SREBP-1 or -2. The nuclear forms of both SREBPs were down-regulated by sterols whether the cells were transfected with an empty vector (lanes 1 and 2), a vector encoding wild-type ACAT ( lanes  3 and 4), or the S265L mutant of ACAT (lanes 5 and 6). Thus, there was no evidence that the S265L mutant interfered with the sterol-mediated regulation of processing of SREBPs. FIG. 4. Immunoblot analysis of ACAT protein in 293 cells transiently transfected with ACAT cDNAs, comparison of wild-type and S265L mutant. Cells in lanes 3-6 were transfected with 5 g of the indicated ACAT cDNA. Cells in lanes 1 and 2 were transfected with 5 g of the same vector without a cDNA insert. Sixteen h after transfection, the cells received fresh medium A in the absence (Ϫ) or presence (ϩ) of sterols (2.5 g/ml 25-hydroxycholesterol plus 50 g/ml LDL) as indicated. After incubation for 5 h at 37°C, the 10 5 ϫ g pellet was isolated as described (14) and subjected to SDS-PAGE and immunoblot analysis. The filter was exposed to film for 1 min at room temperature. Closed arrow denotes position of ACAT monomer.
FIG. 5. ACAT activity in membranes from 293 cells transiently transfected with ACAT cDNAs, comparison of wild-type (q) and S265L mutant (E). Cells were transfected with 1 g of wild-type cDNA (q), 5 g of mutant cDNA (E), or 5 g of vector containing no cDNA insert (Ç). Sixteen h after transfection, the cells received fresh medium A containing 2.5 g/ml 25-hydroxycholesterol plus 50 g/ml LDL. After incubation for 5 h at 37°C, the cells were harvested; a 10 5 ϫ g membrane pellet was isolated, and the indicated amount of membrane protein was assayed for in vitro ACAT activity as described under "Experimental Procedures." Inset, an aliquot of the same 10 5 ϫ g membrane pellet (50 g of protein) was subjected to immunoblot analysis as described in the legend to Fig. 4. The filter was exposed to film for 1 min at room temperature.  After incubation for 16 h at 37°C, the cells were harvested, and the high salt nuclear extract fraction was prepared as described (23). An aliquot of each fraction (50 g of protein) was subjected to SDS-PAGE and immunoblot analysis for SREBP-1 and SREBP-2. The filter was exposed to film at room temperature for 1 min.

DISCUSSION
The current data indicate that the SRD-4 cells contain at least three independent defects as follows: 1) a mutant allele at the ACAT locus that contains a C 3 T substitution, changing codon 265 from serine to leucine and producing an inactive enzyme; 2) a silent allele at the other ACAT locus that does not produce mRNA; and 3) a mutation at an unknown locus that prevents 25-hydroxycholesterol and cholesterol from inhibiting the cleavage of SREBP-1 and -2. All of these abnormalities appear to have arisen independently, and all of them act together to allow the SRD-4 cells to survive selection in the presence of 25-hydroxycholesterol.
The amino acid substitution in the ACAT protein reduces activity by two mechanisms. 1) It reduces the amount of ACAT protein, most likely because the mutant protein is unstable. 2) It severely reduces the catalytic activity of the enzyme. The first abnormality became manifest when we transfected 293 cells with a cDNA encoding the S265L mutant ACAT and found that it produced much lower levels of protein than the cDNA encoding the wild-type enzyme (Fig. 4). We were unable to measure the amount of ACAT protein in the SRD-4 cells themselves because our current antibodies do not have the requisite sensitivity, but we suspect that these cells must also have a reduced level of ACAT protein. The second abnormality was manifest when we transfected 293 cells with a 5-fold excess of plasmid encoding the mutant form of ACAT as compared with the wild-type form so as to obtain approximately equal levels of ACAT protein (Fig. 5). Although the amount of protein was similar, membranes from the cells transfected with the mutant cDNA failed to show an increase in ACAT activity, whereas membranes from cells transfected with the wild-type cDNA showed a clear increase.
The most likely explanation for the silent ACAT allele in the SRD-4 cells is that this allele is in a region of a chromosome that is heavily methylated. Many genes in CHO cells are heavily methylated and thereby silenced from expression leaving the cells functionally hemizygous at many loci (reviewed in Refs. 26 and 27).
The third abnormality in the SRD-4 cells, namely an inability of 25-hydroxycholesterol to suppress cleavage of either SREBP-1 or -2, is so far unique. In previous studies of three independent CHO cell lines that were selected for 25-hydroxycholesterol resistance, we found that each one produced a truncated form of SREBP-2 that terminated at amino acid 460 and was therefore constitutively active (18,19). Whereas the mature form of SREBP-2 was found in the nucleus under all conditions, only trace amounts of SREBP-1 were found there, owing to an apparent down-regulation of the cleavage of SREBP-1 by the truncated SREBP-2 (18,19). The situation with the SRD-4 cells is markedly different. In these cells we find approximately equal amounts of the cleaved forms of SREBP-1 and SREBP-2 in the nucleus under inducing conditions, and neither is suppressed when 25-hydroxycholesterol is added (Fig. 6).
It is likely that the SRD-4 cells have a mutation in a gene whose product is necessary for sterol-mediated inhibition of proteolysis of both SREBPs. This putative protein is not ACAT because correction of the ACAT deficiency by transfection of the wild-type ACAT cDNA failed to correct the regulatory abnormality (Fig. 6). We also considered the possibility that the point mutation in ACAT might produce a dominant effect that prevented 25-hydroxycholesterol from regulating proteolysis of SREBPs. Against this hypothesis is the experiment of Fig. 7, which shows that cleavage of epitope-tagged SREBPs is regulated normally in 293 cells that have been transfected with a cDNA encoding the S265L ACAT mutant.
An interesting sidelight to these studies is the observation of a high basal rate of cholesteryl ester synthesis in SRD-4 cells transfected with the wild-type ACAT cDNA (Table II). The rate was more than 2-fold above normal when the cells were deprived of exogenous sterols, and it was equal to normal when the reaction was stimulated with 25-hydroxycholesterol and LDL. The high rate of esterification in the absence of sterols is most likely due to a high rate of cholesterol synthesis, owing to the defect in feedback suppression. Previous studies with SRD-1, -2, and -3 cells showed that these cells also have a high rate of cholesterol synthesis and esterification in the absence of exogenous cholesterol (20,21). Presumably the sterol feedback regulatory system operates to some degree even in the absence of exogenous sterol, and this is abolished in the sterol-resistant cells.
It is of interest that the only two cell lines so far described with ACAT deficiency both have concurrent defects in sterol regulation of transcription. These are the SRD-4 cells of the current paper and the AC29 cells of Cadigan et al. (5,6). This coincidence raises the possibility that the defects in ACAT and the defect in sterol regulation somehow interact to improve survival of cells in the presence of 25-hydroxycholesterol.
It is straightforward to rationalize how an ACAT deficiency might help cells resist killing by 25-hydroxycholesterol (9). The sterol kills cells by suppressing the cleavage of SREBPs, thereby preventing transcription of genes encoding LDL receptors and enzymes in the cholesterol biosynthetic pathway. The resultant cholesterol deficiency is made worse because 25-hydroxycholesterol also activates ACAT, which esterifies some of the residual free cholesterol, preventing it from being used for membrane synthesis. A deficiency in ACAT would ameliorate this problem partially by abolishing esterification of cholesterol, thereby making more cholesterol available for membrane function.