Microsomal Triglyceride Transfer Protein Activity Is Not Required for the Initiation of Apolipoprotein B-containing Lipoprotein Assembly in McA-RH7777 Cells*

We previously demonstrated that the N-terminal 1000 amino acid residues of human apolipoprotein (apo) B (designated apoB:1000) are competent to fold into a three-sided lipovitellin-like lipid binding cavity to form the apoB “lipid pocket” without a structural requirement for microsomal triglyceride transfer protein (MTP). Our results established that this primordial apoB-containing particle is phospholipid-rich (Manchekar, M., Richardson, P. E., Forte, T. M., Datta, G., Segrest, J. P., and Dashti, N. (2004) J. Biol. Chem. 279, 39757-39766). In this study we have investigated the putative functional role of MTP in the initial lipidation of apoB:1000 in stable transformants of McA-RH7777 cells. Inhibition of MTP lipid transfer activity by 0.1 μm BMS-197636 and 5, 10, and 20 μm of BMS-200150 had no detectable effect on the synthesis, lipidation, and secretion of apoB:1000-containing particles. Under identical experimental conditions, the synthesis, lipidation, and secretion of endogenous apoB100-containing particles in HepG2 and parental untransfected McA-RH7777 cells were inhibited by 86-94%. BMS-200150 at 40 μm nearly abolished the secretion of endogenous apoB100-containing particles in HepG2 and parental McA-RH cells but caused only 15-20% inhibition in the secretion of apoB: 1000-containing particles. This modest decrease was attributable to the nonspecific effect of a high concentration of this compound on hepatic protein synthesis, as reflected in a similar (20-25%) reduction in albumin secretion. Suppression of MTP gene expression in stable transformants of McA-RH7777 cells by micro-interfering RNA led to 60-70% decrease in MTP mRNA and protein levels, but it had no detectable effect on the secretion of apoB:1000. Our results provide a compelling argument that the initial addition of phospholipids to apoB:1000 and initiation of apoB-containing lipoprotein assembly occur independently of MTP lipid transfer activity.

Apolipoprotein B (apoB) 2 is synthesized primarily in hepatocytes and enterocytes and has a fundamental role in the transport and metabolism of plasma triacylglycerols (TAG) and cholesterol (1,2). ApoB is a predominant protein component of very low density lipoproteins (VLDL) and intermediate density lipoproteins and is essentially the only apoprotein component of low density lipoproteins (LDL 2 ) (3,4). ApoB100 (the fulllength protein) is one of the largest monomeric proteins known with 4536 amino acid residues (2). It is expressed primarily in mammalian liver, is an essential structural component for the formation and secretion of VLDL, and serves as a ligand for the LDL receptor (2). ApoB is present as a single molecule per lipoprotein particle (5); and therefore, its concentration in the plasma approximates the number of potential atherogenic lipoprotein particles.
The processes involved in the assembly of apoB-containing lipoproteins in the liver are complex and are regulated at multiple levels throughout the secretory pathway. The assembly of apoB-containing lipoproteins occurs co-translationally (1), i.e. while the C-terminal portion is still being synthesized on the ribosome of the endoplasmic reticulum (ER), the N-terminal portion is translocated across the ER and is assembled as a small lipoprotein particle. The addition of lipids to apoB is widely believed to occur in two steps (2,6,7). The first step involves the addition of small amounts of lipids to apoB, as it is translated and translocated into the lumen of ER preventing its degradation and formation of a partially lipidated small pre-VLDL particle in the high density lipoprotein (HDL) density range (2,7,8). In the second step, this pre-VLDL particle is believed to acquire the bulk of its core lipids and is converted to bona fide VLDL (2,7,9), presumably by fusing with a large, VLDL-sized, apoB-free TAG particle (9). Biochemical studies of VLDL assembly support the concept that the bulk of neutral lipids are added in the second step after apoB translation is completed (10).
* This work was supported by the National Institutes of Health Grants HL084685 and PO1 HL34343. 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 U.S.C. Section 1734 solely to indicate this fact. 1 To whom correspondence should be addressed: Dept. of Medicine, University of Alabama at Birmingham, 1808 7th Ave. South, BDB-D680, Birmingham, AL 35292-0012. Tel.: 205-975-2159; Fax: 205-975-8079; E-mail: ndashti@uab.edu.
One of the most important factors in the assembly and secretion of apoB-containing lipoproteins is microsomal triglyceride transfer protein (MTP), which predominantly resides in the ER of hepatocytes and enterocytes (11)(12)(13)(14). A vital role of MTP in the formation and secretion of apoB-containing lipoprotein particles is further substantiated by the observation that patients with abetalipoproteinemia, an autosomal-recessive disorder caused by mutations in the MTP gene, have low levels of apoB in plasma (11). Although the obligatory role of MTP in the secretion of VLDL is well established, the precise mechanism by which MTP transfers lipids to the nascent apoB polypeptide during its synthesis and assembly into VLDL is not fully understood. Furthermore, the relative importance of MTP in the two steps of VLDL assembly remains unclear and controversial. Some studies indicate that MTP has a crucial role in the first step assembly of VLDL (15)(16)(17)(18)(19)(20), but it is not required for the second-step core expansion during VLDL assembly (17, 19 -21). Others support the concept that MTP is essential for transferring bulk TAG into the lumen of ER for the conversion of small apoB-containing lipoproteins to large VLDL-sized particle (22)(23)(24) and chylomicrons (25). The potential role of MTP in the initial lipidation of apoB and nucleation of the primordial apoB-containing particle has not been elucidated.
Previously, based on experimentally derived results (26,27) and all atom molecular modeling of the ␤␣ 1 domain (amino acid residues 1-1000) of apoB100 (28), we proposed that initiation of apoB particle assembly occurs when the ␤␣ 1 domain, designated apoB:1000, folds into a three-sided lipovitellin (LV)like lipid binding cavity to form the apoB "lipid pocket." Our results supported a model in which the first 1000 amino acid residues of apoB are competent to complete the lipid pocket without a structural requirement for MTP, and that this lipid pocket has a fixed lipid capacity on the order of 50 phospholipids (PL) for a total stoichiometry of 70 lipid molecules, a number in close agreement with that reported for the LV complex (29 -31). We propose that the initiation complex in the assembly of apoB-containing lipoprotein particle is PL-rich suggesting a small bilayer type organization rather than a TAG-rich emulsion proposed by others (32). Although our results (27,28) demonstrated that MTP is not a structural requirement for the formation of the lipid pocket, they did not rule out its potential functional role, i.e. transfer of lipids to the nascent apoB:1000, in this process.
Because MTP both binds to apoB and transfers lipids (33), this study was a logical continuation of our previous work and focused on the following important question: "is MTP involved in the initial addition of PL to nascent apoB:1000 and initiation of apoB-containing lipoprotein particle assembly?" To address this question, we tested the effects of two well characterized inhibitors of MTP lipid transfer activity, BMS-197636 and BMS-200150 (13, 34), on the de novo synthesis, lipidation, and secretion of apoB:1000 in stable transformants of McA-RH7777 cells. We also suppressed the MTP gene expression by miRNA and determined the subsequent effect on the secretion of apoB:1000. Our results provide a compelling argument that the initial addition of phospholipids to apoB:1000 and initiation of apoB-containing lipoprotein assembly occur independently of MTP lipid transfer activity.
Construction of Truncated ApoB Expression Plasmid-Truncated apoB cDNA spanning nucleotides 1-3081 of the fulllength apoB100 cDNA was prepared from pB100L-L (35) as a PCR template and appropriate primers as described previously in detail (26). The amplified PCR product was cloned into the TOPO TA cloning vector and used to transform cells. Clones harboring the vector were selected and identified by restriction enzyme digestion and nucleotide sequencing of the entire open reading frames. Only clones with 100% correct sequence were used in these studies. The 3081-bp fragment (apoB:1000) was excised from the vector, extracted, purified, and ligated into the mammalian expression vector, the Moloney murine leukemia virus-based retrovirus LNCX (36), containing the neomycin phosphotransferase gene, which confers G418 resistance for use as a selectable marker. The apoB expression vector pLNCB: 1000 was used to transform cells, and clones harboring plasmid-containing apoB gene with the correct orientation were identified by restriction enzyme digestion and confirmed by nucleotide sequencing as described previously (26).
Cell Culture and Transfection-McA-RH7777 cells (referred to as McA-RH here) were obtained from American Type Culture Collection (Manassas, VA). Clonal stable transformants of McA-RH cells expressing apoB:1000, denoting amino acid residues 1-1000 of the mature protein lacking the signal peptide, were generated as described previously in detail (26). Cells were grown in DMEM containing 20% HS, 5% FBS, and 0.2 mg/ml G418, and medium was changed every 48 h. All experiments were conducted with 4 -5-day-old cells as described previously (26). The human hepatoblastoma HepG2 cell line (obtained from American Type Culture Collection) was seeded onto tissue culture dishes in MEM containing 10% (v/v) FBS and were incubated at 37°C in a 95% air, 5% CO 2 atmosphere as described previously (37). Medium was changed every 48 h, and all experiments were conducted with 4 -5-day-old cells.
MTP Inhibition-The MTP inhibitors BMS-197636 and BMS-200150 (Bristol-Myers Squibb Co.) (13,34) were dissolved in Me 2 SO at concentrations of 0.025 and 10 mM, respectively. Final Me 2 SO concentration was normalized to 0.4% in all experimental and control dishes.
De Novo Synthesis and Secretion of ApoB and Albumin-Clonal stable transformants of McA-RH cells expressing apoB: 1000 and HepG2 and parental McA-RH cells producing endogenous human and rat full-length apoB100, respectively, were grown for 4 days in 6-well dishes. At the start of experiments, maintenance media were removed; monolayers were washed twice with phosphate-buffered saline (PBS), and cells were preincubated for 45 min in serum-, methionine-, and cysteine-free DMEM. Fresh serum-, methionine-, and cysteine-free DMEM was added, and the incorporation of [ 35 S]Met/Cys (100 Ci/ml of medium) into newly synthesized apoB:1000, apoB100, and albumin in the presence or absence of BMS-197636 and BMS-200150 was determined after 3.5 h of incubation. Control dishes received the same amount of Me 2 SO, i.e. 0.4% final concentration. In a separate experiment, the effects of MTP inhibitor, BMS-197636, on the synthesis and secretion of apoB:1000 was determined as a function of time. After the indicated incubation time, the 35 S-labeled conditioned media were collected; preservative mixtures at final concentrations of 500 units/ml penicillin-G, 50 g/ml streptomycin sulfate, 20 g/ml chloramphenicol, 1.3 mg/ml ⑀-aminocaproic acid, 1 mg/ml EDTA, 1 mM benzamidine, 1 mM phenylmethylsulfonyl fluoride, and 100 kallikrein-inactivating units of aprotinin/ml was added to prevent oxidative and proteolytic damage. The media were centrifuged at 2,000 rpm for 30 min at 4°C to remove broken cells and debris. Cell monolayers were washed with cold PBS, and lysis buffer containing preservative mixture described above plus leupeptin (50 g/ml) and pepstatin A (50 g/ml) was added, and cells were processed as described previously (26). The secreted 35 S-labeled apoB and albumin in the media and 35 S-labeled apoB in cell lysates were isolated by immunoprecipitation as described below.
Metabolic Labeling of the Lipid Content of ApoB-containing Lipoprotein Particles-Cells were grown in 6-well dishes as described above. At the start of experiments, maintenance media were removed; cells were washed twice with PBS and were incubated for 17 h in serum-free DMEM (for McA-RH cells) or MEM (for HepG2 cells) containing either [ 3 H]glycerol (7 Ci/ml of medium) or [ 14 C]oleic acid (0.4 mM) bound to 0.75% BSA. In experiments where the radiolabeled lipids associated with apoB-containing particles were determined by autoradiography, cells were labeled with both [ 3 H]glycerol and [ 14 C]oleic acid to enhance the signal. The labeled conditioned media were processed and supplemented with preservatives as described above. Cell monolayers were washed with PBS, scraped off the plate in 1.0 ml of PBS, and sonicated. The incorporation of [ 3 H]glycerol or [ 14 C]oleic acid into various lipid moieties of apoB-containing lipoproteins secreted into the medium and accumulated in the cells was determined by immunoprecipitation or by nondenaturing gradient gel elec-trophoresis (NDGGE) as described below. Cell protein content was measured by the method of Lowry et al. (38).
Immunoprecipitation-The 35 S-labeled proteins secreted into the conditioned media and accumulated in the cells were immunoprecipitated using monospecific polyclonal antibody to human or rat apoB100 coupled to protein G-Sepharose CL-4B as described previously (27,39). The 35 S-labeled albumin in the conditioned media of HepG2 and McA-RH cells was determined as above using rabbit antibody to human and rat albumin, respectively. The 35 S-labeled proteins were extracted from protein G as described previously (26) and resolved on 4 -12% SDS-PAGE (40). After electrophoresis, the gels were analyzed by autoradiography, in conjunction with computerassisted image processing, or by immunoblotting as described below and in the figure legends.
NDGGE-The [ 3 H]glycerol-labeled apoB-containing lipoprotein particles in the conditioned media were separated on 4 -20% NDGGE as described previously (26). Gels were stained, and the bands corresponding to apoB100 (in parental McA-RH and HepG2 cells) and apoB:1000 (in stably transfected McA-RH cells), identified by immunoblotting of a duplicate gel, were excised and analyzed for lipids as described below. Alternatively, the incorporation of [ 3 H]glycerol and [ 14 C]oleic acid into total lipids of intact apoB-containing lipoproteins was determined by NDGGE of the labeled conditioned media and autoradiography.
Lipid Analysis of Isolated Full-length and Truncated ApoBcontaining Particles-The bands corresponding to labeled apoB100-or apoB:1000-containing particles were excised from NDGGE, and lipids were extracted with chloroform/methanol (2:1) as described previously in detail (27); complete extraction was assessed by liquid scintillation counting the final gel homogenate. Total labeled lipids extracted from gel-isolated apoBcontaining lipoproteins were washed by the Folch method (42) and applied to a TLC plate as described previously (43). The bands corresponding to PL, diacylglycerols, and TAG, identified by comparison to known standards, were visualized with iodine; each band was scraped off the plate, placed in a counting vial, and quantified by liquid scintillation counting.
RT-PCR and Sequencing-Total RNA from parental untransfected McA-RH cells and apoB:1000-expressing stable transformants of McA-RH was isolated with TRIzol reagent (Invitrogen). MTP mRNA was determined by one-step semiquantitative reverse transcriptase (RT)-PCR using sense strand primer AGGCTGGGGAAGGGCCCGTC and antisense strand primer AATGTTCTTCACATCCATGT. Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) mRNA level was used as internal control using sense strand primer GACAA-GATGGTGAAGGTCGGT and antisense primer TTGGC-CCCACCCTTCAGGTG.
Suppression of MTP Expression by miRNA-The pre-miRNA sequences were designed using RNAi designer on-line tool (Invitrogen). Seven different double-stranded oligo duplexes encoding desired miRNA target sequences were selected and cloned into pcDNA TM 6.2-GW, a BLOCK-iT TM pol II miR RNAi expression vector (Invitrogen). The BLOCK-iT TM pol II miR RNAi expression vectors are specifically designed to allow expression of miRNA sequences and contain specific miR flanking sequences that allow proper processing of the miRNA. The vectors also contain the spectinomycin resistance gene for selection in bacteria and the blasticidin resistance gene for selection in mammalian cells. The sequences of one of the most efficient oligo duplexes are as follows: top sequence 5Ј-GCT-GTTTAAGATGACAGCAGCAGCCGTTTTGGCCACTG-ACTGACGGCTGCTGGTCATCTTAAA-3Ј and bottom sequence 5Ј-CCTGTTTAAGATGACCAGCAGCCGTCAG-TCAGTGGCCAAAACGGCTGCTGCTGTCATCTTAAAC-3Ј. The sequences of the negative control oligo duplex are as follows: top sequence 5Ј-GCTGAAATGTACTGCGCGTGG-AGACGTTTTGGCCACTGACTGACGTCTCCACGCAGT-ACATTT-3Ј and bottom sequence 5Ј-CCTGAAATGTACTG-CGTGGAGACGTCAGTCAGTGGCCAAAACGTCTCCAC-GCGCAGTACATTTC-3Ј. Escherichia coli DH5␣ cells were transformed using the vectors harboring the respective doublestranded oligo encoding the engineered pre-miRNAs. Plasmids were purified using standard techniques, analyzed to confirm correct sequence, and used to transfect apoB:1000-expressing stable transformants of McA-RH cells.
Transfecting miRNA into ApoB:1000-expressing McA-RH Cells-Cells were seeded onto 60-mm dishes and grown in DMEM containing 20% horse serum and 5% FBS as described above. After 24 h and at ϳ50 -60% confluency, cells were transfected using TransIT-LT1 transfection reagent (Mirus, Madison, WI) according to the manufacturer's instructions. The cells were trypsinized 48 h post-transfection and were grown in DMEM containing serum and 10 g/ml blasticidin (Invitrogen) to select for clonal stable cells. Cells were then maintained in DMEM containing serum and 5 g/ml blasticidin for 18 -20 days; medium was changed twice weekly, and cells were expanded every 4 days as described previously (26). This approach was necessary to reduce the level of pre-existing MTP, which has a long half-life, i.e. 4.4 days in HepG2 cells (44). MTP mRNA level was determined by RT-PCR, and MTP protein level was assessed by immunoblotting (41) using polyclonal antibody to bovine MTP 97-kDa large subunit.

MTP Inhibitors Have No Effect on the Secretion or Cellular Accumulation of 35 S-Labeled ApoB:1000 Expressed in McA-RH Cells but Markedly Decrease Those of Endogenous Full-length ApoB100 in HepG2 and Parental Untransfected
McA-RH Cells-The liver-derived McA-RH cells retain the ability to synthesize and secrete lipoproteins similar to those in primary hepatocytes (45). McA-RH cells secrete both rat apoB100 and apoB48 in the form of lipoprotein particle (45) and, unlike human-derived hepatoma HepG2 cells, secrete a large fraction of apoB100 in the form of VLDL particles (46). McA-RH cells have a high expression capacity (47) and have successfully been used in numerous studies (23, 26, 27, 48 -50) as a model to investigate the mechanisms of apoB-containing lipoprotein assembly in the liver.
Using stable transformants of McA-RH cells, we demonstrated previously that the first 1000 amino acid residues of apoB100 are competent to form the apoB lipid pocket and initiate apoB particle assembly without the structural requirement for MTP (27). This experimentally derived observation was supported by our all atom molecular model for apoB:1000 (28). In this study, we examined the putative functional role of MTP in the initial lipidation of nascent apoB:1000 and initiation of apoB assembly in stable transformants of McA-RH cells. We tested the effects of two known inhibitors of MTP lipid transfer activity, BMS-197636 and BMS-200150 (13,17,51), on the synthesis, lipidation, and secretion of apoB:1000-containing particles stably expressed in McA-RH cells. Because this class of inhibitors has been shown to markedly decrease the secretion of apoB100-containing particles in HepG2 cells (13) and McA-RH cells (23,48), we included these cells as positive controls.
As the first step, we determined the dose-dependent effects of BMS-200150 on the de novo synthesis and secretion of apoB: 1000 in McA-RH cells and de novo synthesis and secretion of apoB100 in HepG2 cells. Cells were metabolically labeled with [ 35 S]Met/Cys in the presence or absence of MTP inhibitor for 3.5 h. We initially used 5 and 10 M of BMS-200150 that inhibit TAG transfer activity of MTP by 70 and 80%, respectively (13), and have been shown to inhibit apoB100 secretion in HepG2 cells by 65 and 90%, respectively (13). As shown in Fig. 1A, BMS-200150 at 5 or 10 M had no detectable effect on the secretion or cellular accumulation of 35 S-labeled apoB:1000 expressed in McA-RH cells. By contrast, the secretion of 35 Slabeled apoB100 in HepG2 cells was decreased by 86 and 94% with 5 and 10 M of BMS-200150, respectively, and its cellular accumulation was diminished by 65%. This resulted in 83% inhibition in the net de novo synthesis and secretion of apoB100 (Fig. 1A). Similar results were obtained after 17 h of incubation with the inhibitors (data not shown). We reasoned that perhaps higher concentrations of BMS-200150 might be necessary to inhibit the secretion of apoB: 1000-containing particles. As such, we used 20 M BMS-200150, a concentration that inhibits the secretion of apoB100 in HepG2 cells by greater than 90% (13), but we observed no change in either the secretion or cellular accumulation of 35  The labeled conditioned media were concentrated and subjected to NDGGE followed by autoradiography as described previously (27) and under the "Experimental Procedures." Bands corresponding to apoB: 1000-containing particles in stable McA-RH cells, rat endogenous apoB100-containing particles in parental untransfected McA-RH cells, and human endogenous apoB100-containing particles in HepG2 cells, were identified by their Stokes diameter (S d ), immunoblotting with antibody to human or rat apoB100, and their co-mobility with control plasma LDL, when applicable, of a duplicate gel. The intensities of the labeled lipids associated with the apoB-containing particles were measured by computer-assisted image processing. We observed that compared with Me 2 SO control (Fig. 5, A-C, lane 1), 0.1 M BMS-197636 had no effect on the 3 H/ 14 C-labeled lipid content of intact apoB:1000-containing particles in stable transformants of McA-RH cells (Fig. 5B, lane 2) but reduced that associated with intact endogenous apoB100-containing lipoproteins by 70% in both parental untransfected McA-RH cells (Fig.  5A, lane 2) and in HepG2 cells (Fig. 5C, lane 2). BMS-200150 at 40 M caused only a modest 13% inhibition in the 3 H/ 14 C-labeled lipid content of apoB:1000-containing particles secreted by stable transformants of McA-RH cells (Fig. 5B, lane 3) but nearly abolished that of endogenous apoB100-containing particles secreted by both parental untransfected McA-RH cells (Fig. 5A, lane 3) and HepG2 cells (Fig. 5C, lane 3). The inefficacy  The labeled conditioned media were concentrated 10-fold and were subjected to NDGGE for 48 h. Gels were subsequently stained, and bands corresponding to secreted intact apoB:1000-containing particles were excised from the gels. Lipids were extracted from the excised bands and separated on TLC as described previously (27).
Results showed that compared with Me 2 SO control, BMS-197636 at 0.1 M and BMS-200150 at 10 and 20 M had no detectable effect on either the concentration or the composition of 3 H-labeled lipids associated with apoB:1000containing particles (Table 1), confirming the results obtained by autoradiography (Fig. 5B). BMS-200150 at 40 M caused a modest 15% decrease in the 3 H-labeled lipid content of secreted intact apoB:1000-containing particles without altering their lipid composition (Table 1). MTP inhibitors likewise had no effect on the concentration or composition of newly synthesized lipids accumulated in apoB:1000-expressing McA-RH cells (Table 1). Similar results were obtained when apoB: 1000-containing particles were isolated by immunoprecipitation using polyclonal antibody to human apoB100 (data not shown). These results corroborated the effects of these inhibitors of MTP lipid transfer activity on the newly synthesized and secreted 35 S-labeled apoB:1000 (Figs. 1-3) and 3 H/ 14 C-labeled lipids in intact particles determined by autoradiography (Fig. 5B).    35 S-Labeled apoB:1000 in the cell lysate and secreted into the medium was immunoprecipitated with antibody to human apoB100, and proteins were separated by 4 -12% SDS-PAGE and subjected to autoradiography as described in the legend to Fig. 1. media were subjected to NDGGE, and apoB100-containing particles were identified by their co-mobility with human plasma LDL and immunoblotting with antibody to human apoB100. Intact particles were excised from the gels and analyzed for lipids as described above. In sharp contrast to the results obtained for apoB:1000 (Table 1), BMS-197636 (0.1 M) and BMS-200150 at all concentrations tested decreased the 3 H-labeled lipids associated with apoB100 by 75-90% (Table 2). Although MTP inhibitors decreased all lipid species associated with the intact apoB100-containing lipoproteins, this reduction was more pronounced in TAG, resulting in the secretion of particles that contained more PL and less TAG ( Table 2). The MTP inhibitors had no significant effect on the cellular concentration or composition of newly synthesized lipids in HepG2 cells (Table 2). Similar results were obtained when apoB100containing particles were isolated by immunoprecipitation using polyclonal antibody to human apoB100 (data not shown).

Inhibition of MTP Lipid Transfer Activity Drastically Decreases the Concentration and Alters the Composition of Newly Synthesized Lipids Associated With Endogenous
To assess the effects of MTP inhibitors on the lipid content and composition of the secreted rat endogenous apoB100-containing particles, parental untransfected McA-RH cells were metabolically labeled with [ 3 H]glycerol in the presence or absence of MTP inhibitors, as described for apoB:1000-expressing McA-RH and HepG2 cells. The labeled conditioned media were concentrated and applied to NDGGE. Bands corresponding to LDL-sized endogenous apoB100-containing particles, identified by their apparent S d , which was same as that of plasma LDL, and immunoblotting with antibody to rat apoB100, were excised, and lipids were extracted as described under "Experimental Procedures." As shown in Table 3, the total lipid content of rat apoB100-containing particles was decreased by 35 and 65% with BMS-197636 and BMS-200150, respectively. Similarly to that in HepG2 cells (Table 2), the decrease in the presence of 40 M BMS-200150 was more pronounced in TAG, resulting in the secretion of particles that contained more PL and less TAG (Table 3). MTP inhibitors had no effect on the accumulation of lipids in the cells (Table 3).
These results indicate that, as in HepG2 cells, the inhibitors of The labeled conditioned media were concentrated, and intact apoB-containing particles were separated by 4 -20% NDGGE. Gels were stained, dried, subjected to autoradiography, and analyzed by computerassisted image processing. S d , Stokes diameter.

TABLE 1 Effects of MTP inhibitors on ͓ 3 H͔glycerol-labeled lipids accumulated in the cells and associated with secreted human apoB:1000-containing particles in stable transformant of McA-RH cells
Cells were incubated with serum-free DMEM and 3 H-labeled glycerol (  presence of MTP inhibitors could be due to potential alteration in the MTP expression level in these cells. To test this possibility, we measured MTP mRNA and protein levels by RT-PCR and Western blot, respectively, in both the parental untransfected and apoB:1000-expressing McA-RH cells. As shown in Fig. 6A (representative of three individual dishes), we found equivalent levels of MTP mRNA and protein in both cell lines. These results validate the above observations and rule out the possibility that altered MTP expression in transfected McA-RH cells might be the reason for the ineffectiveness of MTP inhibitors to decrease the synthesis and secretion of apoB:1000.

Suppression of MTP Gene Expression in Stable Transformants of McA-RH Cells by miRNA Has No
Effect on the Secretion of ApoB:1000-To further substantiate our hypothesis that addition of phospholipids to apoB:1000 and initiation of apoB particle assembly occur independently of MTP activity, we employed RNAi to suppress MTP gene expression. McA-RH cells stably expressing apoB:1000 were transfected with either MTP miRNA or negative control miRNA as described under "Experimental Procedures." As shown in Fig. 6B (representative of three individual dishes), transfection of cells with MTP miRNA resulted in ϳ60% decrease in MTP mRNA level as compared with cells transfected with negative control miRNA. The GAPDH mRNA level was the same in both cell lines establishing the specificity of MTP miRNA (Fig. 6B). After 18 days in DMEM-containing serum and blasticidin (5 g/ml of medium), the secretion of 35 S-labeled apoB:1000 in both cell lines was determined and was compared with the MTP protein levels. As shown in Fig. 7, transfection of cells with MTP miRNA led to an ϳ70% decrease in MTP protein level when compared with cells transfected with negative control miRNA. In contrast, there was no change in the secretion of 35 S-labeled apoB:1000 in  Fig. 6. Cells were grown for 18 days in DMEM-containing serum and blasticidin (5 g/ml of medium). Both cell lines were incubated for 3.5 h in serum-, methionine-, and cysteine-free DMEM containing [ 35 S]Met/Cys (100 Ci/ml). The 35 S-labeled apoB:1000 secreted into the medium was immunoprecipitated using monospecific polyclonal antibody to human apoB100, and proteins were separated by 4 -12% SDS-PAGE and subjected to autoradiography. Cell monolayers from duplicate dishes were washed with PBS and analyzed for MTP protein level by Western blot. B, intensities of the MTP protein and 35 Slabeled apoB:1000 bands were determined by computer-assisted image processing, normalized for cell protein, and plotted as mean Ϯ S.E. of triplicate dishes.

TABLE 3 Effects of MTP inhibitors on ͓ 3 H͔glycerol-labeled lipids accumulated in the cells and associated with secreted rat endogenous apoB100-containing particles in parental untransfected McA-RH cells
Cells were incubated with serum-free DMEM and ͓ 3 H͔labeled glycerol (7 Ci/ml of medium) in the presence or absence of MTP inhibitors. Secreted particles were isolated by NDGGE and analyzed for lipids. Values are means Ϯ S.E. of nine samples for total lipids and seven samples for lipid composition from three separate experiments normalized to cell protein. DAG indicates diacylglycerol. McA-RH cells transfected with MTP miRNA as compared with cells transfected with negative control miRNA (Fig. 7).

DISCUSSION
We previously suggested (53,54), based on sequence homology between the ␤␣ 1 domain of apoB (N-terminal 1000-residue domain of apoB100) and LV (54 -57), that this region is homologous to the proposed lipid binding cavity in LV, and therefore, this domain might be a lipid-binding pocket for apoB (54). We proposed that formation of an LV-like lipid pocket is necessary for lipid transfer to apoB-containing lipoprotein particles, and we suggested (53, 54) that initiation of particle assembly occurs when the ␤␣ 1 domain folds into a three-sided LV-like lipid binding cavity. Alternatively, the lipid pocket we propose is formed by association of the region of the ␤␣ 1 domain homologous to the ␤A and ␤B sheets of LV with ␤D-like amphipathic ␤-sheet from MTP (53,54). In our initial studies (26), we provided evidence for the formation of a lipid pocket intermediate in the assembly of apoB-containing lipoproteins by the N-terminal 1000 amino acid residues of apoB100. We demonstrated that apoB:1000 is secreted by stable transformants of McA-RH cells as a monodisperse, relatively lipid-enriched particle in the HDL 3 -like density range (26). In subsequent studies, our experimentally derived results (27) and all atom molecular modeling of the ␤␣ 1 domain (28) demonstrated that the N-terminal 1000 amino acid residues of apoB100 are necessary and competent to form the lipid pocket without the structural requirement for MTP and that the lipid pocket is PL-rich (27). These results, however, did not exclude the putative functional role of MTP in the initial lipidation of apoB:1000.
To determine whether addition of phospholipids to apoB: 1000 and the formation of the PL-rich initiation complex in apoB assembly is dependent on MTP lipid transfer activity, we tested the effects of two well characterized (13,34) and widely used (13,17,20,23,33,48,52) inhibitors of MTP lipid transfer activity, BMS-197636 and BMS-200150, on the de novo synthesis, lipidation, and secretion of apoB:1000 in stable transformants of McA-RH cells. To provide more definitive evidence for our hypothesis that the initial step in apoB particle assembly is independent of MTP activity, we suppressed MTP gene expression in apoB:1000-expressing McA-RH cells by miRNA, and we assessed potential correlation between MTP protein level and apoB:1000 secretion.
Results clearly and consistently demonstrated that BMS-200150 at 5, 10, or 20 M and BMS-197636 at 0.1 or 0.2 M had no detectable effect on either the synthesis and secretion of 35 S-labeled apoB:1000 or the content and composition of lipids associated with the secreted intact apoB:1000-containing particles in stable transformants of McA-RH cells. In marked contrast, and consistent with previously reported studies in HepG2 cells (13,51) and McA-RH cells stably expressing human apoB100 (23,48), the secretion of 35 Table 1) and time course (Fig. 4)  MTP has a distinct preference for TAG transfer and its lipid transport rate decreases in the order of TAG Ͼ cholesteryl esters Ͼ diacylglycerol Ͼ cholesterol Ͼ PL (11,52,58,59). In a recent study, Rava et al. (60) have supported our previous stud-ies demonstrating that the primordial apoB-containing particles are PL-rich (26 -28) and have proposed that the PL transfer activity of MTP is sufficient for the assembly and secretion of primordial apoB lipoproteins. This conclusion was based mainly on the observations derived from the effect of Drosophila MTP on the secretion of truncated forms of human apoB in COS cells (60). These investigators showed that Drosophila MTP, which is defective in TAG transfer activity but has PL transfer activity equal to that of human MTP, supports the secretion of human apoB48, apoB53, and apoB72 in COS cells (60). Our results, indicating that MTP activity is not required for the initiation of apoB assembly, are at variance with the conclusion reached by Rava et al. (60). Although we do not know the exact reason for this discrepancy, we speculate that the inconsistency in the results arise, most likely, from the use of cell lines with distinctly different tissue origins. Several differences between the two cell lines, with regard to lipoprotein metabolism, are noteworthy and are outlined below.
First, in our study, we used lipoprotein producing rat hepatoma McA-RH cells, whereas Rava et al. (60) used COS cells, transformed African green monkey kidney fibroblast cells, which normally do not produce lipoproteins. Second, in the study by Rava et al. (60), the secretion of apoB18 in COS cells was considerably higher than that of apoB48 and especially apoB53, whereas in McA-RH cells, apoB18 was secreted much more slowly than apoB48 and apoB53 (47). Third, Rava et al. (60) demonstrated that MTP is not required for the secretion of apoB18. Relevant to this finding are studies demonstrating that apoB17 readily associates with phospholipids (61)(62)(63). Assuming that PL transfer activity of MTP is sufficient for the formation of primordial apoB particles, it is reasonable to expect that the secretion of apoB18 in COS cells would increase, at least partially, by the expression of Drosophila MTP; this was not observed by Rava et al. (60). Fourth, Rava et al. (60) were unable to measure any significant PL transfer activity in cell lysates of COS cells expressing either human or Drosophila MTP. In addition, these investigators (60) observed 72% decrease in TAG transfer activity in mouse Mttp gene-deleted liver homogenates but did not detect any change in PL transfer activity, indicating the presence of other PL transfer activities. In our studies, we found similar levels of MTP mRNA and protein in parental untransfected and apoB:1000-expressing McA-RH cells (Fig. 6), and we demonstrated that miRNA-mediated 70% decrease in MTP protein level had no effect on the secretion of 35 S-labeled apoB:1000 (Fig. 7). Fifth, in the study by Rava et al. (60), the secretion of apoB48 in COS cells was shown to be dependent on MTP expression. In contrast, studies by Wang et al. (23) demonstrated that in McA-RH cells, the assembly/ secretion of apoB48 HDL was relatively unaffected by the MTP inhibitor, suggesting that secretion of apoB48 in liver-derived cells may not depend on MTP activity. Considering the complexity of apoB-containing lipoprotein assembly, COS cells may not have the full complement of numerous factors and chaperons known to be necessary for apoB-containing particle maturation (2). This possibility is supported by studies demonstrating that nonhepatic cell lines synthesize apoB but cannot process it into lipoproteins (64).
Several lines of evidence support our results and conclusion. First, the PL transfer activity of MTP, which is ϳ5% of its TAG transfer activity, is inhibited by 60% in the presence of 0.1 M of BMS-197636 (52). Our results demonstrated that 0.1 M of BMS-197636 had no detectable effect on either the synthesis and secretion of 35 S-labeled apoB:1000 (Figs. 2-4) or the concentration and composition of 3 H-labeled lipids associated with intact apoB:1000-containing particles secreted by stable McA-RH cells (Fig. 5 and Table 1 Table 1). Third, although MTP has been proposed to be crucial for the proper folding and lipidation of apoB during translocation into the lumen of ER (15,18,19), in vitro translation studies have shown that apoB48 is efficiently translocated into the lumen of dog pancreas microsomes in which the activity of MTP is not detectable (65). Fourth, in McA-RH cells, the secretion of human apoB29 and apoB18 was not affected by the MTP inhibitor 4Ј-bromo-3Ј-methylmetaqualone (66). In mouse primary hepatocytes, BMS-197636 inhibited the secretion of apoB100 by 90%, whereas apoB48 secretion was only slightly decreased (21). In mouse hepatocyte-like cell line MhAT3F, which produces both apoB100 and apoB48, MTP inhibitor preferentially blocked the secretion of apoB100 (66). In HepG2 cells, synthesis of apoB polypeptides the size of 100 -200 kDa was insensitive to MTP inhibitors, suggesting that MTP inhibitors did not affect the initiation of apoB100 translation (16). In human-derived differentiated intestinal Caco-2 cells, BMS-200150 impaired the secretion of apoB100, whereas apoB48 secretion was relatively unaffected (67). Fifth, the liverspecific inactivation of the Mttp gene in the mouse lowered apoB100 levels in plasma by Ͼ95% but reduced plasma apoB48 levels by only 20% (24), and MTP-deficient mouse hepatocytes secreted apoB-containing lipoproteins of HDL and LDL sizes but not VLDL-sized lipoproteins (68).
In summary, we investigated the putative functional role of MTP in the initiation of apoB assembly relying on the use of two inhibitors of MTP lipid transfer activity, BMS-197636 and BMS-200150, and suppression of MTP gene expression by miRNA. Both inhibitors consistently failed to affect the synthesis, lipidation, assembly, and secretion of apoB:1000-containing particles in stable transformants of McA-RH cells indicating that MTP activity is not required for the initial addition of PL to the growing polypeptide. The miRNA-mediated suppression of MTP gene expression led to a 60 -70% reduction in MTP mRNA and protein levels but had no detectable effect on the secretion of apoB:1000 in McA-RH cells. From the perspective of apoB assembly, our results do not rule out the role of MTP in the first step assembly of apoB-containing particles. At present, the structural elements within the apoB polypeptide that govern requirement for MTP are not known. Available evidence suggest the following: (i) segments equal to or greater than apoB48 containing the N-terminal 17% of the protein respond to MTP activity (14); (ii) sequences in the C terminus of apoB29 bind PL (69); (iii) sequences between apoB29 and apoB32.5 augment TAG binding (69); (iv) sequences between apoB32.5 and apoB41 account for the marked incorporation of TAG (69); and (v) the domain between apoB51 and apoB53 has a high requirement for MTP (66), and this region is within the ␣ 2 domain of apoB100 (53). Our results provide a compelling argument that the addition of phospholipids to apoB:1000 (apoB22.05), initiation of apoB particle assembly, and the formation of the primordial PL-rich apoB-containing lipoproteins are independent of MTP lipid transfer activity. We propose that factor(s) other than MTP mediate this early stage of apoB lipoprotein particle assembly. Identification of this factor would render it an efficacious pharmacological target to reduce the formation of atherogenic apoB-containing lipoproteins at the very early stage.