Overexpression of Membrane Domain of SCAP Prevents Sterols from Inhibiting SCAP·SREBP Exit from Endoplasmic Reticulum*

SCAP (SREBP cleavage-activating protein) forms a complex with sterol regulatory element-binding proteins (SREBPs) and escorts them from the endoplasmic reticulum (ER) to the Golgi complex where proteases release transcriptionally active segments of SREBPs, which enter the nucleus to activate lipid synthesis. The NH2-terminal segment of SCAP contains eight transmembrane helices, five of which (TM2–6) comprise the sterol-sensing domain. This domain responds to sterols by causing the SCAP·SREBP complex to be retained in the ER, preventing proteolytic release and reducing transcription of lipogenic genes. Here, we use transfection techniques to overexpress a segment of SCAP containing transmembrane helices 1–6 in hamster and human cells. This segment does not interfere with SCAP·SREBP movement to the Golgi in the absence of sterols, but it prevents sterols from suppressing this movement. This block is abolished when SCAP(TM1–6) contains a point mutation (Y298C) that is known to abolish the activity of the sterol-sensing domain. We interpret these findings to indicate that sterols cause the SCAP·SREBP complex to bind to an ER retention protein through an interaction that involves the sterol-sensing domain. The SCAP(TM1–6) segment competes with the SCAP·SREBP complex for binding to this putative retention protein, thereby liberating the SCAP·SREBP complex so that it can move to the Golgi despite the presence of sterols. These studies provide a potential mechanistic explanation for the ability of sterols to block SCAP·SREBP movement from the ER and thereby to control lipid synthesis in animal cells.

The gated movement of sterol regulatory element-binding proteins (SREBPs) 1 from endoplasmic reticulum (ER) to the Golgi complex controls the rate of lipid synthesis in animal cells (1). SREBPs are membrane-bound transcription factors whose NH 2 -terminal domains must be released proteolytically to enter the nucleus. For proteolysis to take place, the SREBPs must be transported from their site of synthesis in the ER to their site of proteolysis in the Golgi complex (2). This transport is accomplished by SCAP (SREBP cleavage-activating protein), a membrane-bound protein that forms a complex with SREBPs in the ER and escorts them to the Golgi complex.
SREBPs are tripartite membrane proteins of ϳ1150 amino acids in length (1). The NH 2 -terminal domain of ϳ480 amino acids contains a basic-helix-loop-helix-leucine-zipper motif that allows it to bind DNA and activate transcription. This domain is followed by a membrane attachment domain of ϳ80 amino acids consisting of two transmembrane helices separated by a short hydrophilic loop that projects into the lumen of the ER. The COOH-terminal domain of ϳ590 amino acids performs a regulatory function. The SREBPs are oriented in the membrane in a hairpin fashion with the NH 2 -and COOH-terminal domains projecting into the cytoplasm (3).
SCAP is a polytopic membrane protein with two distinct domains (4). The NH 2 -terminal domain of ϳ550 amino acids consists of alternating hydrophobic and hydrophilic sequences that are believed to form eight membrane-spanning helices. The COOH-terminal domain of ϳ725 amino acids is hydrophilic and projects into the cytoplasm. This domain contains five copies of a WD-40 repeat sequence that is found in many proteins and mediates protein/protein interactions (5). Indeed, immediately after its synthesis, SCAP forms a complex with SREBPs that is mediated by an interaction of the WD-40 repeat domain of SCAP and the COOH-terminal regulatory domain of the SREBPs (6,7).
When cells are depleted of sterols, the SCAP⅐SREBP complex travels from the ER to the Golgi, where it encounters two proteases that act in sequence to release the NH 2 -terminal domain of SREBP into the cytoplasm (2,8). The first enzyme, designated Site-1 protease, cleaves the SREBP in the luminal loop, thereby separating the two transmembrane helices (9). This allows Site-2 protease to cleave the first transmembrane helix within the plane of the membrane, releasing the NH 2terminal domain with three hydrophobic amino acids attached (10). This domain travels to the nucleus where it activates transcription of multiple genes involved in the synthesis of cholesterol and fatty acids as well as their uptake from plasma lipoproteins through the low density lipoprotein receptor (1).
Movement of SREBP to the Golgi absolutely requires SCAP. In SCAP-deficient mutant cell lines, the SREBPs fail to move to the Golgi complex, and they are not released proteolytically from membranes. As a result, SCAP-deficient cells are cholesterol auxotrophs (11). The SCAP⅐SREBP transport process is controlled by cholesterol. When sterols build up in cells, the SCAP⅐SREBP complex fails to bud from ER membranes and it never reaches the Golgi. 2 As a result, the synthesis of cholesterol and fatty acids is reduced. This sterol-regulated transport mechanism modulates the cholesterol content of cell membranes. * The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18  The sterol-sensing function in the SREBP system has been traced to a portion of the membrane domain of SCAP encompassing transmembrane helices 1-6 (TM1-6). This region contains sequences that are shared with three other proteins that are postulated to interact with sterols (12). This segment has therefore been termed "the sterol-sensing domain" (13). Point mutations at two conserved residues within this domain (Y298C and D443N) abolish the sterol-sensing function of SCAP (12,14). These mutant forms of SCAP form complexes with SREBPs that are transported to the Golgi normally, but they can no longer be blocked by sterols. 2 As a result, cells bearing these SCAP mutations overproduce cholesterol and fail to shut-off cholesterol synthesis when intracellular sterol levels rise.
A crucial question relates to the mechanism by which SCAP is retained in the ER in sterol-overloaded cells. Two mechanisms must be considered. The first model postulates that the SCAP⅐SREBP complex leaves the ER constitutively and that sterols cause the complex to bind to a molecule, presumably a protein, that actively retains it in the ER. The alternative hypothesis states that ER retention of the complex represents the constitutive state. For the SCAP⅐SREBP complex to leave the ER, it must interact with a protein that carries it to the Golgi. This interaction would occur only in sterol-depleted cells.
The current studies were designed to distinguish between these two possibilities. For this purpose, we transfected cells with cDNAs encoding a portion of the transmembrane domain of SCAP that includes the sterol sensor. Overexpression of this domain abolished the ability of sterols to suppress SCAP movement and SREBP cleavage. This effect was not observed when the transmembrane domain contained the Y298C mutation. These findings support the model in which the SCAP⅐SREBP complex is retained in the ER through an interaction between the transmembrane domain and an unidentified retention protein. The overexpressed transmembrane domain competes for this interaction and allows the SCAP⅐SREBP complex to move constitutively even in the presence of sterols.
pCMV-Xp-SCAP(1-448, Y298C) was amplified by PCR of pCMVSCAP(1-764, Y298C) (provided by A. Nohturfft) with the same pair of aforementioned primers and constructed in the same way. pCMV-SCAP(449 -1276) was amplified by PCR of pCMV-SCAP with a pair of primers, 5Ј-GCAACGGATCCATGCTAGCGGACCTGAAC-3Ј, which encodes amino acids 449 -453 preceded by an initiation codon (ATG) and a BamHI site, and 5Ј-CGCGCTCTAGATCAGTCCAGTT-TCTCCAG-3Ј, which encodes amino acids 1272-1276 followed by a stop codon (TGA) and an XbaI site. The PCR fragment was cloned into pcDNA3 vector in a similar fashion as described above.
Culture, Transfection, and Fractionation of HEK-293 Cells-Monolayers of human embryonic kidney (HEK)-293 cells were set up on day 0 at 4 ϫ 10 5 cells/60-mm dish (or 1.5 ϫ 10 5 cells/60-mm dish) and cultured in 8 -9% CO 2 at 37°C in medium A (Dulbecco's modified Eagle's medium containing 100 units/ml penicillin and 100 g/ml streptomycin sulfate) supplemented with 10% (v/v) fetal calf serum. On day 2 (or day 3), the cells were transfected with the indicated plasmids as described previously (3). Three h after transfection, the cells were switched to medium B (medium A containing 10% newborn calf lipoprotein-deficient serum, 50 M sodium compactin, and 50 M sodium mevalonate) in the absence or presence of sterols (varying concentrations of 25-hydroxycholesterol plus 10 g/ml cholesterol added in a final concentration of 0.2% (v/v) ethanol) as indicated in the legends. After incubation for 6 or 12 h, N-acetyl-leucinalleucinal-norleucinal (ALLN) at a final concentration of 25 g/ml was added directly to each dish. After incubation for 1 h, pooled cells from two dishes were harvested and used for preparing nuclear extract and membrane fractions as described previously (17).
Culture, Transfection, and Fractionation of SRD-13A Cells-Monolayers of SRD-13A cells, a line of mutant Chinese hamster ovary cells that lacks SCAP (11), were set up on day 0 (5 ϫ 10 5 cells/60-mm dish) and cultured in 8 -9% CO 2 at 37°C in medium C (1:1 mixture of Ham's F-12 medium and Dulbecco's modified Eagle's medium containing 100 units/ml penicillin and 100 g/ml streptomycin sulfate) supplemented with 5% fetal calf serum, 5 g/ml cholesterol, 1 mM sodium mevalonate, and 20 M sodium oleate. On day 2, the cells were transfected with the indicated plasmids using Fugene 6 reagent as described previously (11). After 10 or 15 h, the cells were washed with medium C and refed with either medium C or medium D (medium C containing 5% newborn calf lipoprotein-deficient serum, 50 M sodium compactin, 50 M sodium mevalonate, and 0.2% ethanol) containing the indicated additions as described in the legends to Figs. 5 and 6.
Trypsin Digestion of Membranes and Glycosidase Treatment-Monolayers of HEK-293 cells were harvested, and membrane fractions were prepared and treated sequentially with trypsin, trypsin inhibitor, and endoglycosidase H (endo H) as described previously (8) except that the endo H treatment was carried out for 4 h at 37°C.
Immunoprecipitation and Immunoblot Analysis-Cell monolayers were harvested, and membrane fractions were prepared as described previously (12). The membrane fractions were then solubilized with 1 ml of Nonidet P-40 lysis buffer as described (6), except that the mixture was passed through a 22.5-gauge needle five times and extracted for 1 h at 4°C. Immunoprecipitation was carried out as described (6). SDS-PAGE and subsequent immunoblot analysis were carried out as described using the Super Signal CL-HRP substrate system (Pierce) (7). Two SCAP antibodies were used: IgG-9D5, a mouse monoclonal antibody against hamster SCAP (amino acids 540 -707) (6); and IgG-R139, a rabbit polyclonal antibody against hamster SCAP (amino acids 54 -277 and 540 -707) (6). Fig. 1 shows a diagram of the membrane topology of SCAP, denoting the two segments of SCAP that were produced by the cDNAs used in these studies. The positions of the various epitopes, the three N-linked glycosylation sites, and the Y298C mutation are also shown. The sterol-sensing domain of SCAP comprises transmembrane helices 2-6 (shaded in Fig. 1). Fig. 2 shows an experiment in which cultured human cells were transfected with a cDNA encoding SREBP-2 with an epitope tag at the NH 2 terminus. The cells were also transfected with cDNAs encoding a portion of the membrane domain of SCAP (transmembrane helices 1-6) tagged with an Xpress (Xp) epitope. The encoded protein contained either the wildtype sequence or the Y298C mutation. The cells were incubated in the absence of sterols or in the presence of increasing concentrations of a mixture of 25-hydroxycholesterol and cholesterol, after which nuclear extracts and membrane fractions were subjected to SDS-polyacrylamide gel electrophoresis and immunoblotted with antibodies against either the HSV-tag on SREBP-2 or the Xpress-tag on the truncated SCAP. When the cells expressing HSV-SREBP-2 alone were incubated in the absence of sterols, the SREBP was cleaved, and the NH 2terminal segment was found in the nuclear extract (top panel, lane 2). The addition of sterols abolished cleavage and the nuclear segment disappeared (lanes [3][4][5]. Expression of the truncated wild-type Xp-SCAP(TM1-6) did not affect SREBP cleavage in the absence of sterols (lane 6), but it prevented suppression of cleavage by sterols (lanes 7-9). When the Xp-SCAP(TM1-6) protein bore the Y298C mutation, it lost the ability to block sterol suppression (lanes 10 -13). The bottom panel of Fig. 2 shows that the wild-type and Y298C-truncated SCAP proteins were expressed at equal levels in the cells.

RESULTS
The experiment of Fig. 3A shows that the abolition of sterol suppression required the first six transmembrane helices of SCAP. Transfection of cDNAs encoding smaller segments that included transmembrane segments 1-5 failed to abolish sterol suppression (lanes 10 and 11). Suppression was abolished only when the construct contained transmembrane domains 1-6 ( lanes 12 and 13).
In the experiment of Fig. 3B, we transfected cells with a cDNA encoding a truncated form of SCAP that extends from amino acids 1 to 1195. This construct encodes all eight transmembrane helices plus the bulk of the cytosolic WD-40 domain (see Fig. 1). However, it is truncated at a position that is 80 amino acids from the COOH terminus of the protein. Recent studies have shown that this truncated protein does not interact with SREBPs as determined by a failure of coimmunoprecipitation, nor does it restore SREBP cleavage to SCAP-defi-cient cells. 3 As seen before, when we transfected the cells with HSV-SREBP-2 alone, sterols suppressed cleavage (Fig. 3B, lanes  2-4). When we cotransfected SCAP(1-1195), sterol suppression was abolished (lanes 5-7). The effect of this truncated SCAP was eliminated when we deleted transmembrane helices 3-6 (lanes 8 -10), supporting the idea that these transmembrane helices were required for this effect. The bottom panel of Fig. 3B shows that comparable amounts of the SCAP(1-1195) construct and the transmembrane 3-6 deletion were expressed in the cells.
The membrane domain of SCAP contains three N-linked carbohydrate chains that are attached to loops that project into the ER lumen (see Fig. 1) (4). In sterol-depleted cells, SCAP moves to the Golgi complex where the carbohydrates are processed by enzymes that render them resistant to digestion by endo H (8). When cells are grown in the presence of sterols, the SCAP⅐SREBP complex is trapped in the ER, and the N-linked sugars remain in their unprocessed, endo H-sensitive state. To detect the changes in electrophoretic mobility upon endo H treatment, we first treat intact membrane vesicles with trypsin to reduce the size of the glycosylated fragment thereby permitting clear discrimination between glycosylated and nonglyco-  10-13). The total amount of DNA was adjusted to 4.5 g/dish by addition of pTK mock vector and/or pcDNA3.1HisC mock vector. After transfection, the cells were incubated in medium B with 0.2% ethanol containing the indicated final concentration of 25-hydroxycholesterol (25-HC). Cells incubated with 25-hydroxycholesterol also received 10 g/ml cholesterol. After incubation for 6 h at 37°C, ALLN was added to each dish, and the cells were harvested after incubation for 1 additional hour. Aliquots of the nuclear extract (70 g representing 1 dish of cells) and membrane (60 g representing 0.5 dish of cells) fractions were subjected to SDS-PAGE and immunoblot analysis with 67 ng/ml anti-HSV IgG or 1 g/ml of anti-Xp IgG, respectively. N and P denote the cleaved nuclear and the uncleaved precursor forms of HSV epitope-tagged SREBP-2, respectively. S denotes the NH 2 -terminal domain of SCAP containing transmembrane segments 1-6. Filters were exposed to film for 20 s for the anti-HSV blots and 1 min for the anti-Xp blot. The asterisk denotes a nonspecific cross-reactive band. sylated forms (8,12). One of the trypsin-resistant fragments contains two of the N-linked carbohydrate chains, and this fragment can be visualized by immunoblotting with an antibody against an epitope on this fragment (see Fig. 1). Fig. 4 shows an experiment designed to explore the glycosylation state of endogenous SCAP in cells that were transfected with cDNAs encoding truncated SCAP. This experiment was possible, because Xp-SCAP(TM1-6) does not contain the trypsin-resistant segment that is glycosylated, and it does not react with the antibody against the glycosylated epitope. Thus, the antibody specifically visualizes the trypsin-resistant fragment from endogenous SCAP. When the cells were transfected only with HSV-SREBP-2 and incubated in the absence of sterols, the trypsin-resistant SCAP fragment retained either one or two N-linked carbohydrate chains after treatment with endo H, indicating that the molecules were largely endo H-resistant (Fig. 4, lane 2). When the cells were incubated in the presence of sterols, the N-linked sugars on SCAP were sensitive to endo H digestion, and the digested protein migrated at a position corresponding to zero sugar chains (lane 3). When we cotransfected the cDNA encoding Xp-SCAP(TM1-6), there was no effect on endogenous SCAP in the absence of sterols (lane 4). However, in the presence of sterols the SCAP molecules remained endo H-resistant (lane 5), indicating that some of the endogenous SCAP continued to traffic to the Golgi complex. When Xp-SCAP(TM1-6) contained the Y298C mutation, it had no effect on the behavior of endogenous SCAP either in the with 0.2% ethanol containing the indicated final concentration of 25hydroxycholesterol (25-HC). Cells incubated with 25-hydroxycholesterol also received 10 g/ml cholesterol. After incubation for 12 h, ALLN was added and the cells were harvested 1 h later. Aliquots of the nuclear extract (35 g) and membrane (30 g) fractions were subjected to SDS-PAGE and immunoblot analysis with 67 ng/ml anti-HSV IgG or 5 g/ml IgG-9D5 (anti-SCAP), respectively. N and P denote the cleaved nuclear and the uncleaved precursor forms of HSV epitope-tagged SREBP-2, respectively. S denotes the truncated forms of SCAP. The asterisk denotes a nonspecific cross-reactive band. Filters were exposed to film for 1 s.  6 and 7); pCMV-SCAP(1-346, TM 1-3) (lanes 8 and 9); pCMV-SCAP(1-415, TM 1-5) (lanes 10 and 11); and pCMV-SCAP(1-448, TM 1-6) (lanes 12 and 13). The total amount of DNA was adjusted to 4.5 g/dish by addition of pTK mock vector and/or pcDNA3.1HisC mock vector. After transfection, the cells were incubated in medium B with 0.2% ethanol in the absence or presence of a final concentration of 1 g/ml 25-hydroxycholesterol plus 10 g/ml cholesterol (sterols). After incubation for 16 h, ALLN was added and the cells were harvested 1 h later. Aliquots of the nuclear extract (40 g) and membrane (60 g) fractions were subjected to SDS-PAGE and immunoblot analysis with 67 ng/ml anti-HSV IgG or 1 g/ml of anti-Xp IgG, respectively. N denotes the cleaved nuclear form of HSV epitope-tagged SREBP-2, and S denotes the various SCAP fragments. Filters were exposed to film for 20 s. B, the plasmids used in the transfection were as follows: 2.5 g of pTK-HSV-SREBP-2 (lanes 2-10); 2 g of pCMV-SCAP (1-1195) (lanes 5-7); and 2 g of pCMV-SCAP(1-1195/⌬308 -480) (lanes 8 -10). The total amount of DNA was adjusted to 4.5 g/dish by addition of pTK mock vector and/or pcDNA3 mock vector. After transfection, the cells were incubated in medium B  6 and 7). The total amount of DNA was adjusted to 4.5 g/dish by addition of pTK mock vector and/or pcDNA3.1HisC mock vector. After transfection, the cells were incubated in medium B with 0.2% ethanol in the absence or presence of 1 g/ml 25-hydroxycholesterol plus 10 g/ml cholesterol (sterols). After incubation for 16 h, the cells were harvested and membrane fractions were prepared as described under "Experimental Procedures." All membranes were incubated with trypsin, proteolysis was stopped by addition of trypsin inhibitor, and the SDS-solubilized membranes were then treated with endo H. The samples were subjected to SDS-PAGE and immunoblot analysis with 2 g/ml anti-SCAP polyclonal antibody (IgG-R139). Filters were exposed to film for 15 s. Numbers 0 -2 on the right denote differentially glycosylated forms of the trypsin-resistant SCAP fragment containing the corresponding numbers of N-linked oligosaccharides.
absence or presence of sterols (compare lanes 6 and 7 with lanes  2 and 3). This experiment demonstrates that the wild-type SCAP(TM 1-6) segment prevents sterols from blocking the movement of endogenous SCAP from ER to Golgi.
The results of Figs. 2-4 support the hypothesis that the SCAP(TM 1-6) segment interacts with a putative protein that retains endogenous SCAP in the ER in the presence of sterols. This interaction displaces the endogenous SCAP⅐SREBP complex from the retaining protein, thereby allowing it to travel to the Golgi even in the presence of sterols.
The experiment of Fig 1-6). The total amount of DNA was adjusted to 5 g/dish by addition of pTK mock vector and/or pcDNA3.1HisC mock vector. 10 h after transfection, the cells were incubated for 12 h in medium D in the absence (lanes 1, 3, and 5) or presence (lanes 2, 4, and 6) of 1 g/ml 25-hydroxycholesterol plus 10 g/ml cholesterol (sterols). ALLN was added for 1 h, and the cells were harvested and fractionated. The membrane fraction was solubilized and subjected to immunoprecipitation as described under "Experimental Procedures" using 2 g of anti-Xp antibody (IgG 1 ). The pellet and supernatant fractions from the immunoprecipitation of two dishes of cells were subjected to SDS-PAGE and immunoblot analysis with 1 g/ml anti-Xp IgG or 5 g/ml of monoclonal IgG-9D5 (anti-SCAP), respectively, followed by a 1:1000 dilution of rat anti-mouse horseradish peroxidase-conjugated antibody ( chain). S NH2 denotes the NH 2 -terminal portion of SCAP containing the transmembrane segments 1-6. S COOH denotes the COOH-terminal portion of SCAP containing the transmembrane segments 7 and 8 and the cytosolic COOH-terminal WD domain. The filters were exposed to film for 4 min.  4 and 5). The total amount of DNA was adjusted to 2.5 g/dish by addition of pcDNA3.1HisC mock vector. On day 2, the cells were washed once with phosphate-buffered saline and refed with medium C supplemented with 5% fetal calf lipoprotein-deficient serum. Cells were refed every day. On day 14, cells were washed, fixed in 95% ethanol, and stained with crystal violet.  6, 7, 9, 12, and 13). The total amount of DNA was adjusted to 5 g/dish by addition of pTK mock vector and/or pcDNA3 mock vector. 15 h after transfection, the cells were incubated for 1 h in medium C supplemented with 5% newborn calf lipoprotein-deficient serum, 0.25% (w/v) 2-hydroxypropyl-␤-cyclodextrin, and 25 g/ml ALLN. The cells were harvested and fractionated as described previously (2) except that the nuclear extract was further spun for 20 min at 100,000 ϫ g. Aliquots of the nuclear extract fraction (35 g) and membrane fraction (30 out of 50 l) were subjected to SDS-PAGE and immunoblot analysis with 67 ng/ml anti-HSV IgG, 5 g/ml monoclonal IgG-9D5 (anti-SCAP), or 1 g/ml anti-Xp IgG as indicated. N and P denote the cleaved nuclear and uncleaved precursor forms of HSV epitope-tagged SREBP-2, respectively. S NH2 denotes the NH 2 -terminal segment of SCAP containing transmembrane segments 1-6. S COOH denotes the COOH-terminal segment of SCAP containing transmembrane segments 7 and 8 and the cytosolic COOHterminal WD domain. Filters were exposed to film for 1 min. The asterisk denotes a nonspecific cross-reactive band.
did not restore cleavage (lane 3), nor did transfection of a cDNA encoding the NH 2 -terminal segment with either the wild-type sequence (lane 4) or the Y298C mutation (lane 5). Cleavage was restored only when we expressed SCAP(449 -1276) together with SCAP(1-448) containing either the wild-type sequence (lane 6) or the Y298C mutation (lane 7). The specificity of this effect was documented by the demonstration that no combination of SCAP segments could achieve cleavage of the R519A mutant of SREBP-2, which is not recognized by Site-1 protease (lanes 8 -13) (9, 16). The two bottom panels of Fig. 5 show that the various segments of SCAP were expressed in the transfected cells as determined by immunoblotting with the anti-Xp, which recognizes the NH 2 -terminal segment, or anti-SCAP, which recognizes the COOH-terminal segment.
The positive result in Fig. 5 suggests that the NH 2 -and COOH-terminal segments of SCAP are able to interact so as to restore SCAP activity even though they are expressed as separate proteins. To study this interaction directly, we performed a coimmunoprecipitation experiment (Fig. 6). SCAP-deficient SRD-13A cells were transfected with cDNAs encoding the COOH-terminal segment of SCAP(449 -1276) together with the NH 2 -terminal segment (1-448) containing either the wild-type or Y298C sequence. Membrane extracts were immunoprecipitated with anti-Xp, which recognizes only the NH 2 -terminal segment. The supernatant and pellet fractions from the immunoprecipitations were subjected to SDS-PAGE and blotted with anti-Xp or with anti-SCAP, which detects the COOH-terminal segment. As shown in the bottom panel of Fig. 6, a fraction of the COOH-terminal segment was found in the pellet after precipitation with the antibody against the NH 2 -terminal segment (lanes 3-6), indicating that these two segments were physically associated. The coimmunoprecipitation was the same in the absence and presence of sterols.
If the NH 2 -terminal and COOH-terminal segments of SCAP can physically interact when transfected into cells, then they should be able to restore the growth of SCAP-deficient SRD-13 cells, which are auxotrophic for cholesterol (11). The experiment of Fig. 7, in which SRD-13 cells were transfected with various SCAP segments and grown for 14 days in the absence of cholesterol, indicates that this was indeed the case. DISCUSSION The current experiments demonstrate that overexpression of a portion of the transmembrane domain of SCAP (transmembrane helices 1-6) abolishes the ability of sterols to cause the retention of the SCAP⅐SREBP complex in the ER. Instead, the SCAP⅐SREBP complex moves to the Golgi complex where the carbohydrates of SCAP are processed to an endo H-resistant form and where SREBP is cleaved by the Site-1 and Site-2 proteases to liberate the transcriptionally active NH 2 -terminal segment. When the SCAP(TM1-6) segment contains the Y298C mutation, its ability to block sterol repression of SCAP⅐SREBP movement is markedly reduced.
The simplest hypothesis to explain these results postulates that the SCAP(TM1-6) segment competes with the endogenous SCAP⅐SREBP complex for binding to a protein that retains the complex in the ER in the presence of sterols. This hypothesis is supported by the observation that the Y298C mutation in the SCAP(TM1-6) segment abolishes this effect. The Y298C mutation in full-length SCAP has already been shown to render the protein unresponsive to sterols (12). Thus, SCAP(TM1-6)/ Y298C does not compete for the retention protein either because tyrosine 298 of SCAP is essential to bind sterols, which in turn triggers binding to the retention protein, or, alternatively, SCAP(TM1-6)/Y298C may continue to bind sterols, but it cannot bind to the retention protein. Either of these mechanisms would explain the failure of SCAP/Y298C to be regulated by sterols as well as the finding that SCAP(TM1-6)/Y298C does not compete with endogenous SCAP⅐SREBP for the retention protein in the presence of sterols. SCAP(TM1-6) includes the sterol-sensing domain (TM2-6) that is shared with three other proteins (reviewed in Ref. 13). It will be important to learn whether the sterol-sensing domains in these proteins interact with the same protein that retains SCAP and whether SCAP(TM1-6) will interfere with the actions of these proteins. One of these proteins is 3-hydroxy-3-methylglutaryl-coenzyme A reductase, an enzyme of cholesterol biosynthesis that is bound to ER membranes (14). Sterols accelerate the degradation of this enzyme in an action that requires the sterolsensing domain (18,19). We are currently conducting experiments to determine whether overexpression of SCAP(TM1-6) will interfere with the ability of sterols to accelerate the degradation of 3-hydroxy-3-methylglutaryl-coenzyme A reductase.
Another important finding in these studies is that the NH 2terminal and COOH-terminal domains of SCAP, expressed separately, can reconstitute the SREBP transport function of SCAP in SCAP-deficient hamster cells. This reconstitution was apparently attributable to the ability of these two segments to form a complex that was susceptible to coimmunoprecipitation (Fig. 6). The COOH-terminal segment that we employed contains three distinct regions: 1) TM segments 7 and 8; 2) the large intraluminal loop that connects these segments; and 3) the cytoplasmic WD-repeat domains. One or more of these regions must be capable of binding to the TM1-6 segment, thereby restoring the function of SCAP.
If the proposed mechanism of SCAP(TM1-6) activity is correct, this protein segment should provide a powerful tool for isolating the postulated ER retention protein with which it interacts.