Biochimica et Bioph+~+icaAc:., 1086(1991) 245-254 © 1991 ElsevierScience PublishersB.V. All rightsreserved (I(XI5-2760/91/$ID.50

245

BBALIP 53768

The effect of apo E secretion on lipoprotein uptake in transfected cells Hitoshi Shimano, Chikafusa Fukazawa, Yoshikazu Shibasaki, N a t u s k o Mori, Takanari Gotoda, Kenji Harada, M a s a k o Shimada, N o b u h i r o Y a m a d a , Yoshio Yazaki and Fumimaro Takaku Tile Third Departmentof b~ternalMedici.he.Facul~ of Mt'dicim; Unu'ersityof Tokyo. Tokyo (Japan)

(Received 15 March 19911 (Revised manuscriptreceived4 July I~,~l)

Key words: ApolipoptoieinE; Lipoptoteinmetabolism:Transfection:(Chinesehamsterovarycell) To investigate the role of apolipoprotein E (apo E) secreted by peripheral tissues in local lipoprotein metabolism, we developed a cell strain that constitutively produced and secreted apo E. A fusion plasmid containing rat ap~ E genomic DNA under control of mouse metallothionein promotor was constructed and transfected into Chinese hamster ovary cells. A stable transformant designated CHO-MAEll constitutively secreted rat apo E mainly in the form of sialylated free protein. The secretion was further enhanced by metal induction up to l / z g apo E / m l per 12 h. When incubated with t2Sl-iaheled very low density lipoprotein (tZsl-VLDL) at 37°C, CHO-MAEli took up and degraded tZsl-VLDL with higher affinity than control cells. Furthermore, considerable amount of methylated tT"SI-VLDL was degraded by CHO-MAEII, while no metbylated I~I-VLDL was degraded by control cells. No significant differences were foqmd in the uptake of tZSl-LDL The data indicated that apo E molecules secreted by CHO-MAEll were !,-'=nsferred to tT"s-VLDL particles, which caused a higher affinity of these particles for LDL receptors on the cells. It is suggested that apo E secreted from peripheral tissues enhances the uptake of Ilpoproteins by themselves or by surrounding cells in the local environment which demand cholesterol and express LDL receptors. CHO-MAEll was a good model for these 'autn- or paracrine-like functions' of agn E.

Introduction Apolipoprotein E (apo E ) is mainly p r o d u c e d in the liver and is an important constituent of plasma iipoproteins such as chylomicron, very low density (VLDL), intermediate density lipoprotein (IDL) and high density lipopr~tein (HDL) (for review, see Ref. l). Apo E has a high affinity for LDL receptors on the surface of liver and extrahepatic cells as well as apolipoprotein BI00. It is also thought to be a sole and specific ligand

Abbreviations: CHO. Chinese hamsterovary:VLDL, very low density lipoprotein; IDL, intermediate low densitylipoprotein; LDL. low density lipoprotein; HDL, high density lipoprotein; apo E, apolil~protein E, apo BI00, apolipoproteinBI00; NCS. newborncalfserum: BSA, bovine serum albumin: LPDS, lipoprotein deficient serum: HAM, Ham'smedium. Correspondence: N. Yamada, The Third Department of Internal Medicine, Faculty of Medicine, Universityof Tokyo, 7-3-1 Hongo, Bunkyo-ku, Tokyo 113,Japan.

for apo E receptor (chylomicron remnant receptor). A ]ipoprmein panicle whh ~vera] apo E m0]ecu~es has a

24-foM greater affinity tor LDL receptors than LDL [2-4]. As a result, V L D L particles with apo E are removed more rapidly by the liver than those without apo E [5]. Therefore, apo E regulates the catabolism of these lipoproteins and plays an important role in cholesterol metabolism. On the other hand, extrahepatic tissues such as resident macrophages, brain, adrenals and kidneys secrete substantial amounts of apo E [6-8]. To explain the significance of the local secretion of apo E, it was proposed that peripheral tissues laden with excess cholesterol, such as foamed macrophagcs, excrete cholesterol into HDL and finally back to the liver in the system called reverse cholesterol transport [6]. In this process, apo E secreted from macrophage can be transferred to these HDL particles and function as a ligand for hepatic LDL receptors to be taken up by the liver, in addition to plasma-mediated reverse transport of cholesterol, Mahley et al. and we have proposed that

246 apo E plays a crucial role in the redistribution of lipids among cells within an organ or tissue [1,9]. Secreted apo E captures lipids in the interstitial fluid and delivers them to cells in the local environment which demand lipids for membrane repair or cell proliferation such as in peripheral nerve injury and regeneration. Wc reported that both apo E and lipoprotein lipase secreted by monocyte-derived macrophages enhance the uptake of VLDL by themselves [9]. In the process of atherogenesis, because of endothelial damage and increased permeability, VLDL might be recruited into the artery wall and the autocrine-like function of apo E might contribute to atherogenesis. In the current study, we made cell strains which constitutively produced and secreted apo E as an in vitro model, and tested the effect of apo E secretion on the cellular uptake of lipoproteins. Materials and Methods

Materials CHO-K cells were purchased from ATCC. Na~2Sl, [a-~2P]dCTP, [T-32p]dATP, TranSlabel (L-[35S]methionine), and )251-Protein A were purchased from ICN. Ham's medium (HAM) FI2, phosphate-buffered saline (PBS) and new-born calf serum (NCS) were obtained from GIBCO. Geneticin (G418) and bovine serum albumin (BSA) were from Sigma. The restriction or modifying enzymes used were from Nippon Gene or Toyobo (Japan). Other reagents used were all of the first grade. Cell cultures Wild-type Chinese hamster ovary cells (CHO-K) were maintained in HAM FI2 medium containing 10% Newborn Calf Serum (NCS) in humidified 5% CO 2 incubators at 37°C. Construction of the fit,zion plasmid pMAEll Fusion plasmid pMGH which contains the promotor of mouse metallothionein I (MTHi) and rat growth hormone gene was kindly provided by Dr. Nakanishi [10]. The unique Xhol site of the plasmid was converted to the BamHI site by linker ligation as follows: digestion with Xhol, filling in of the sticky ends with Klenow enzyme in the presence of all tbur dNTPs and ligating BamHI linker to the blunt ends. Bacterial strain RRI was transformed with this DNA. The resultant plasmid containing the ne.,v BamH1 site was prepared on a small scale from the transformant, digested with Barn HI and dephosphorylated with bacterial alkaline phosphatase. The plasmid pALE 31, which contains whole genomic .rat apo E gene, was also digested with BamHl [11,12]. The first BamH1 fragment of pALE31 covering from 11 base pa~rs upstream from

the transcription initiation site to the second BamHI site in the third intron was isolated by preparative 0.9% agarose gel electrophoresis, using a commercial DNA purification kit (Gone Clean). These two fragments were ligated with T4 ligase to yield a plasmid which contained the MTH promotor, the first BamH! fragment of pALE31 and pBR322. Then RRI was transformed with these DNAs. The plasmid DNAs were prepared on a small scale from transformants and digested with several restriction enzymes to select the plasmid in which BamH1 fragment of pALE31 was inserted orthodoromically. The new plasmid purified by large scale preparation and pALE31 were then digested with Hincll and Accl. The fragments were ligated and RRI was transformed with this mixture. To prepare the probes for colony hybridization, pALE31 was digested with Aval, and the sticky ends were filled in with [a-32p]dATP and other dNTPS with Klenow enzyme. The labeled Aval fragments were run in the 7% polyacrylamide gel, and the gels corresponding to the two fragments (130 bp and 328 bp) were excised on the basis of the bands on the autoradiogram. These fragments, 130 bp and 328 bp, are located in the second intron and fourth exon of rat genomic genes, respectively, and serve as adequate probes to pick up the clone containing the whole structure of the rat apo E gone. Clones positive for the two probes were iso. lated and cultured. Plasmid DNAs were prepared on a small scale, and by analyzing their digestion patterns with some restriction enzymes, we finally selected the clone which consistently contained the MTH promotor and the whole structure of rat genomic apo E without 5' flanking DNA and vector PUCI2 derived from pALE31. The new fusion plasmid was designated pMAEII. The nucleotide sequence at the junctional site of the MTH premotor and rat apo E gone was sequenced by the dideoxy method using Sequenase sequence kit (United States Corporation, U.S.A.) and confirmed to be identical with the predicted sequence.

Transfection and selection of CHO cells expressing rat apo E CHO-K cells were seeded at 0.8.107 cells/10 cm dish and cultured in HAM FI2 with 10% NCS for 24 h. 4 h prior to transfection, the medium was changed to 5 ml of DMEM with 10% FCS. DNA-calcium phosphate coprecipitates were prepared as follows [13]: 200 p,I of the solution containing 10/zg DNA of pMAEIi, 2 p,g DNA of pSV2Neo [14] and 250 mM CaCl: was added dropwise to the same volume of 2 x HBS buffer (280 mM NaCI, 50 mM Hopes, !.5 mM Na2HPO4, pH 7.0-7.1) and the mixture was kept still at room temperature for 30 min. Then the mixture was added to each medium and incubated for 6 h. The cells were washed once with phosphate-buffered saline (PBS) without CaCI 2 and shocked for 90 s with 15% glycerol in 1 x

247 HBS. The cells were washed with PBS and allowed to grow for an additional 48 h in Ham FI2 with 10% NCS. At this time the cells were subcultured and diluted 1:5 in the complete medium supplemented with 600 /~g/ml of (]418. After 3 weeks of G418 selection, each resistant colony was isolated and cultured individually in a 24-well plate. In the confluency, the medium was changed to Ham FI2 with 0.2% BSA and 50 p.M ZnSO 4 and incubated for 24 h. The conditioned medium were subjected to the first screenir, g by immunoblotting for rat apo E (described below), and the positive clones which fully expressed rat apo E were selected and maintained in the complete medium with 3 0 0 / z g / m l of G418. One of the highly expressive transfectants was subjected to limiting dilution in 96well plate to confirm a single clone. A strong positive clone was obtained in the second screening and designated CHO-MAEII. For transient expression, the procedure for transfection was the same as that for stable expression except that pSV2Neo was not used. A transfectant with only pSV2Neo was also acquired for control cells. lmmunoblot or slot blot for rat apo E or LDL receptor Anti-rat apo E rabbit serum was used for immunoblotting to detect secreted rat apo E by transfected cells. The conditioned media from CHO-MAEll or control cells cultured in HAM FI2 with 0.2-4% BSA for 24 h were collected. Then 50 p.I of sample mixed with 50 p.I of 2 x buffer O (the final solution containing 5.5% (v/v)/$-mercaptoethanol, 62.5 mM Tris pH 8.4, 2.3% sodium dodecyl sulfate (SDS), 10% glycerol and 0.01% blue phenol blue) was subjected to electrophoresis in sodium dodecyi sulfate polyacrylamide gel (11%) and electrophoretically transferred to nitrocellulose paper in methanol buffer containing 20% methanol, 230 mM glycine, 25 mM Tris and 0.02% SDS at 9fl V of constant voltage for 5 h according to the method of Laemmli [15]. For slot blot analysis, the samples (1013 ALl of the cell-conditioned medium or purified rat a0o E in HAM containing 0.2% BSA and 1% SDS) were directly slotted to the paper. The papers were rapidly washed with the rinse buffer containing 1% Triton X-100, 1 mM EDTA, 150 mM NaCI, 10 mM Tris (pH 7.5) and immersed in the rinse buffer containing 5% skim milk at room temperature for 2 h. After two washes with the rinse buffer, the filters were incubated in the rinse buffer containing anti-rat apo E rabbit serum (0.2 # l / m l ) [16] and 3% BSA at 4°C overnight. The papers were washed three times with 50 ml of the rinse buffer by shaking for 30 rain and incubated in the rinse buffer containing 12-Sl-labeled protein A (2000 c p m / n g , 0.5 /zg/ml) and 3% BSA. Then the membranes were washed again three times with the rinse buffer containing 1 M NaCI, followed by air-drying and autoradiography on XAR-5 (Kodak).

The immunoblot procedure for LDL receptors was as described [17]. The cells were dis~lved in the lysis buffer containing 50 mM Tris-maleate, 2 mM CaCI 2, 1% Triton X-100 and I mM PMSF, pH 6.0. The protein contents of the lysates were measured with a BCA (bicinchoninie acid) protein assay kit (Pierce) [18]. The cell lysates containing 1 mg of protein were adjusted to provide 1% $DS, 10% glycerol and 0.1% bromphenol blue without heating or adding a reducing agent. Then, the samples were subjected to sodium dodecyl sulfate polyacry!amide gel electrophoresis (37 ~ gradient gel) at 10 mA at 4°C for 12 h. After electrophoresis, the samples were electrophoreticaUy transferred from the gel to nitrocellulose membranes in the buffer containing 20 mM Trisbase, 150 mM glycine, 0.02% SDS and 20% methanol at 200 mA for 16 h. The sheets were immersed in blocking buffer containing 150 mM NaCI, 10 mM Tris-HCI (pH 7.4), 50 m g / m l BSA for i h at room temperature and incubated with the same buffer containing anti-LDL receptor serum for 24 h at 4°C. Polyclonal antibody against rabbit LDL receptors was produced as described previously [19]. Then the sheets were washed twice for 15 min with the washing buffer containing 150 mM NaCI, 50 mM Tris-HCI (pH 7.4), 0.1% SDS, 0.2% Nonidet P40 and 0.25% deox-yeholate, and incubated in the buffer containing n~Sl-protein A (2000 c p m / n g , 0.5 p.g/ml) for 24 h at 4°C. The sheets were washed again, air-dried and autoradiographed. Preparation of RNA and Northern blot Total ccUular RNA was prepared by extraction with acid guanidinium thiocyanate, phenol and chloroform as described [20]. 10/s.g of total RNA was subjected to electrophoresis in formaldehyde-agarose (1%) gel and transferred to a nylon membrane. The filter was baked, prehybridized and hybridized with the probe of 1.2 kb BamH! ~ragment of pALF_31, which was prepared as describe I above and labeled with [32P]dCTP by the random ~rimer method according to the manufac'~er'~ protocol (specific activity = ! • 10~ cpm/p.g) [21]. The filter was washed, air-dried and autoradiographed for one day. The conditions for (pre-)hybridization and washing were as described [22]. Metabolic labeling and immunoprecipitation of secreted rat apo E The cultured cells were grown in a 3.5 cm dish in HAM FI2 with 10% NCS until confluent. The cells were washed extensively with PBS and labeled in methionine free DMEM with I00 p.Ci of Tran35S Label containing L-[3~S]methionine and cysteine for 24 h. The media were dialysed against 25 mM ammonium bicarbonate. 1 /zl of antiserum and immu,loprecipitation buffer were added to 50/zl of sample to give 500 p.I containing 0.02% NaPO 4, 150 mM NaCI, 1% Triton-X

248 100, 1% deoxycholate and 100 #M phcnylmethylsulfonyl fluoride. The mixture was rotated at 4°C overnight and immunoprecipitated with a 50 pA suspension of Pansorbin, followed by 20 rain rotation at 4°C. Immunoprecipitates were centrifuged at 13 000 rpm for 2 rain and washed with 1 ml of immunoprecipitation buffer without deoxycholate and spun down five times. The final pellets were dissolved in 1 × buffer O and boiled for 3 rain. An aliquot of supernatant was subjected to electrophoresis in 11% SDS-PAGE and stained with Coomassie blue, fixed with acetate and enhanced in autoradiography enhancer (EN3HANCE, New England Nuclear). The gel was washed, heat-dried and autoradiographed. In another experiment, the sample was treated with neuraminidase (2 U / m l ) in 0.l M sodium acetate (pH 6.0) for 18 h at room temperature before electrophoresis.

Lipoproteins and binding assay for LDL receptor Lipoproteins were prepared from human volunteers' plasma by sequential ultracent,ifugation [23]. Each fraction was floated again at the same density. Reductive methylation of VLDL and LDL was performed by adding sodium borohydride and aqueous formaldehyde to lipoproteins in borate buffer exactly as described

[24]. Lipoproteins were labeled with Na1251 by a modification of the iodine monochloride method of McFarlane [25], CHO-MAEII and control cells were seeded at 1.0.105 cells/2.2 cm well and cultured in HAM FI2 with 10% NCS. In the state of sub-confluency (usually 3-4 days after seeding) the medium was changed to HAM F12 with 0.2% BSA and 50 p.M ZnSO 4 to induce the promotor of MTHI integrated into chromosomal DNA of CHO cells and incubated for 12 h. The labeled lipoproteins were added to the medium to yield the indicated concentrations. Binding assay at 37°C f(~r LDL receptors was performed as described [26]. The incubation period was 12 h, and surface-bound and internalized 1251-1ipoproteins were exhibited together as cell associations. The contents of cell proteins dissolved in 0.1 M NaOH were determined by the method of Lowry et al. [27]. Results We constructed the plasmid containing the rat genomic apo E gene whose 5' flanking region is replaced with the promotor of MTHI as shown in Fig. 1. In the construction carried out in this study, the predicted transcription of rat apo E gene under control of the MTH promotor must be initiated at the site l l bp upstream from the authentic transcription initiation site of rat apo E gene under control of the intrinsic promotor. The transcript will be I I bp longer than authentic mRNA. To determine whether pMAEII was

II Mouse MTI-I-Iproulolor -~,-,-i----...-r-'~ Ratt,~ enumle ' Apo 1'~ (l~Tkb ~,TG77//l':1mli~._it-i~

)

~

~

( 2.9kb)

PUCI2\ \

~/-U

( 2 7k b ) ~ , . , ~

J

B

Fig. I. Constructionof the fusion plasmid pMAEII.The promotor (metal regulatoryregion) and transcription initiation signalsderived from the mousemelallothionein I gene were comained in a 1.7 kb fragment (striped box) derived from pMGH. Rat apo E genomic gene and plasmid vector pUCI2 were derived from pALE31. Rat apo E gene contains four exons(black boxes). Replication origin of the plasmid(Ori) and translation initiation site of apo E gene (CTG) were indicated. Restriction enzymes used for construction were BamH! (B), Hictl (H) and Accl (A). functional in spite of this extra sequence on transcription and translation, CHO cells were preliminarily transfected with MAEII for transient expression by the method of CaPO 4 co-precipitation [13]. Immunoblotting of the cell lysates and the conditioned medium using anti-rat apo E serum was performed. The specific band of apo E was detected in both the cell lysates and the media at the same size as rat apo E in plasma, while no band was found in the conditioned medium in untransfected cells (data not shown). Thus we confirmed normal transcription and translation of pMAEII and secretion of rat apo E by the transfeeted mammalian cells. Then CHO cells were co-transfected with pMAEII and pSV2Neo, which contains neomycln-resistant gene for stable expression [14]. Stable transfectants were selected on the basis of resistance to G418 in the culture medium. After screening the conditioned media for rat apo E by immunoblotting, culturing the strongly positive clones in limiting dilution and con, ducting a second screening, we acquired a cell line that stably expressed rat apo E and designated it CHOMAEII.

Characteristics of secreted apo E Fig. 2 shows the representative results of immunoblotting of the conditioned media from transfeetants for rat apo E. Northern blot analysis showed that rat apo E messenger RNA tlas transcribed constitutively and increased further by metal induction (Fig. 3). 35S-labeled rat apo E secreted in the medium was immunoprecipitated and analyzed on SDS-PAGE. As

249

1

Mr (kDa)

2

3

4

5

42.7

31.0

Fig. 2. Immunoblot of apo E in the conditioned medium by CHO-MAEII. For ~reening transfectcd ClIO cells highly expressive of rat aoo E, conditioned media (50 /~l/each lane) from nansfected cell clones were subjected to immunoblot as de~ribed in Materials and Methods. Molc,,:ularweight standards are indicated. I, rat ~rum (I /.all as positive control of rat apo E, 2-4, ~mples from GM418 resistant transfected CHO cell clones; 5. sample from untransfectod CHO-K cells as negative control. ".:'heauthentic rat apo E protein was detected on lane I at ,~ kDa with heaw chain of immunoglobulin G at 50 kDa. The strong positive clone on lane 2 was designated CHO-MAEII.

1

2

3

4

28S

18S

Fig. 3. Nonhero blot analysis of rat apo E mRNA from CHO-MAEII. Total cellular RNA (10 tz.g) from the transfectants was subjected to Northern blot as described in Materials and Methods. I. control cells; 2, control cells treated with zinc ion; 3, CHO-MAEII uninduced; 4. CHO-MAEII zinc-induced. 18S and 28S ribosomes are indicated.

s h o w n in Fig. 4. b a n d s w e r e d e t e c t e d in doublet at 34 k D a a n d 38 k D a , a n d a single 34 k D a b a n d w a s f o u n d a f t e r t r e a t m e n t with n e u r a m i n i d a s e . T h i s suggests t h a t rat a p o E particles s e c r e t e d f r o m C H O - M A E l l a r e almost sialylated. T h e ratio o f i m m u n o p r e c i p i t a b l e c o u n t by anti-rat a p o E a n t i b o d y versus T C A precipitable c o u n t in the c o n d i t i o n e d m e d i u m w a s 8 - 1 0 % . C a l c u l a t e d from the p r o t e i n m a s s in the m e d i u m by the m e t h o d of Lowry, the a m o u n t o f a p o E secreted in the m e d i u m after 12 h o f i n c u b a t i o n with C H O - M A E I 1 w a s a b o u t 1 p . g / m l o n the a s s u m p t i o n t h a t secreted proteins w e r e equally labeled with [35S]methionine. T o m e a s u r e the a m o u n t o f s e c r e t e d a p o E m o r e precisely, slot blot analysis with specific a n t i b o d y was p e r f o r m e d (Fig. 5). Linearity w a s observed b e t w e e n 2 - 5 0 n g rat a p o E protein in a s t a n d a r d curve d r a w n f r o m the d a t a f r o m a d e n s i t o m e t r i c scan o f slot blot o f purified rat a p o E (Fig, 5a). Using this assay, the time-course o f the a m o u n t o f rat a p o E s e c r e t e d into the m e d i u m by C H O . M A E l l was d e t e r m i n e d (Fig. 5b). T i m e - d e p e n d e n t a c c u m u l a t i o n o f a p o E in the m e d i u m w a s observed, but the increase w a s sluggish a f t e r 12 h of incubation, Induction with zinc ion c a u s e d a n increase in the a m o u n t s of s e c r e t e d a p o E. W h e n i n d u c e d with zinc ion, the c o n c e n t r a t i o n s o f a p o E in the m e d i u m a f t e r 12 a n d 24 h o f incubation, w~:ich c o r r e s p o n d to

250

1

Mr

2

3

Binding assays of 1251-1ipoproteins to CHO-MAEH

(kDa) --

:

97.4

66.2

42.7

t

__

__

: :.'i~i;~i'-.:~-:.::.~i~'~, - . . ~ ~ ~i:::~:!::-~:~iiii~i~i~ili!~. -'~,;~i : ( ~ : ~ ! : ~ ¥ : ' : .::~ ~'.~,~:~,i~'~.~-~, ':::;ii!ii~:~:ii~'~i~ ;;'~;~!i~?~'~ii!~i!i ~:~!?~;~.~%!~i~:~:~.i::;~i:~/~i~2 / i~!~,,i ~ ~iii: ~ ~''ffr ;

~

.

~

;

~

,

~

~.

~

; ~:~'-,~,~~:~,~:ff,'- ":"~;=....

Binding assays of 125I-VLDL were performed for C H O - M A E I I and control cells. Prior to the assay, the cells were pre-incubated in the m e d i u m containing 0.2% BSA for 12 h and the indicated . . . . . . trati . . . . f 125I-VLDL were added, followed by another 12 h of incubation. W e did not use LPDS because LPDS contains a significant amount of apo E derived from serum, which would influence the results of the assay. Saturation kinetics of cell-association and degradation at 37°C

2.0

31.0

21.$

/

0..¢

0

Fig. 4. lmmunoprecipitation of apo E secreted by CHO-MAEII. Cells were incubated for 24 h in the presence of L-[35S]methionine with methionine free medium in 2.2 cm dishes. Cell-conditioned medium was mixed with anti-rat apo E serum, followed by the addition of protein A. lmmunoprecipitates were subjected to 10% SDS-polyacrylamide gel electrophoresis and autoradiographed, l, CHO-MAEll conditioned medium; 2, CHO-MAEII conditioned medium treated with neuraminidase; 3, control cell conditioned medium.

the condition of the binding assay, were 1.07 and 1.28 p . g / m l (0.99 and 1.18/zg/104 cells), respectively. These values were higher than that of apo E secreted by human monocyte-derived macrophage (0.2-0.4 p . g / m l ) [9]. The amount in the m e d i u m with the control cells was undetectable at any incubation time. All the subsequent experiments were performed with C H O - M A E I 1 and control cells induced with zinc ion. Analysis of secreted aSS-labeled apo E by gel filtration exhibited a major peak after the authentic peak of HDL. The immunoprecipitable counts distributed in H D L or larger particles represented 11% of total immunoprecipitable counts in the medium. The ratio of floated counts at the density of 1.025 versus total immunoprecipitable counts in the medium was 7.9%. These data indicate that approx. 90% of secreted apo E exists as free protein and the rest as lipid particles.

20

40

60

80

I00

rat 8po g (ng)

z,

/

Oo~° t'o-" so-

3'o " ,'o -s'o

hours Fig. 5. Standard curve for slot blot assayof rat apo E and time.course changes in concentrations of apo E in the medium. The indicated amounts of purified rat apo E and CHO-MAE]] conditioned medium at the indicated incubation times was subjected to slot blot. The amounts of purified rat apo E and optical densities of those slots on autoradiogram were plotted (a). From this standard curve, the contents of apo E in conditioned medium were determined. Changes in contents of secreted rat apo E at the indicated incubation times by CHO-MAEI] zinc-induced (o), uninduced (e) and control cells ( • ) are shown (b).

2~1

~

12o -

too.

80.

•

o

E et ~

• ~

.

/ ° ~.~o

o

300-

,

.

1

/

I

,

.

2

~ r3 =

"

,

.

,

.

4

3

20o.

tu

m ..u

, s -

t2s I-VLDL (pg / ml)

, 10

-

, 24

-

. 34

-

. ao

-

, so

12SI-VLDL (pg / ml)

400" eo i

3Og

I

20S

100.

0 0

1

~t

3

4

$

12s I-VLDL (pg / ml)

0

I0

20

30

40

5'0

t2s i-VLDL (pg / ml)

Fig. 6. Saturation kinetics in cell-association and degradation of t~I-VLDL in CHO-MAEII. CHO-MAEII and control cells subconfluenl were preincubatcd in HAM with 0.2% BSA and 50/.tM ZnSO4 for 12 h, and the indicated concentrations of t-~I-VLDL(specific activity= 150-400 cpm/ng) in the presence or absence of 50-fold unlabeled VLDL were added. After another 12 h of incubation at 3"PC, the medium and the cells were harvested to measure cell-associated and degraded amounts of *~I-VLDL. Saturation curves of cell-association (upper) and degradation (lower) in CHO-MAEIi (o) and control cells (o) at low concentrations (1-5 ;tg/ml, a) and high concentrations (I-50 p.g/ml, b) are shown. The values are expressed as means of the values in tbe absence of exce~ cold minus those in the presence in triplicate assays from the representative results of three independent experiments. Scotcbard plots from cell-association are shown in the inset. Kd values from three independent experimz,~ts (where three different lots of VLDL were used) were CHO-MAEII: 7.67. 8.97 and 6.67, and control cells: 50.45, 32.2 and 14.8, respectively(.ttg/ml).

w e r e shown in Fig. 6. C H O - M A E I I took up and deg r a d e d m o r e t251-VLDL than control cells in which only p S V 2 N e o was transfected, especially at low concentrations of V L D L ( 1 - 1 0 / t g / m i ) (Fig. 6a). T h e difference was less p r o m i n e n t at high concentrations, and no difference was found at the concentration of saturation ( 5 0 / ~ g / m l ) (Fig. 6b). Scatchard plot showed

that the affinity of V L D L for C H O - M A E I I was about 4.2-fold g r e a t e r than that for control cells in the uptake (Fig. 6b inset, the m e a n Ko values from three independent experiments are 32.47 p , g / m l vs. 7 . 7 7 / ~ g / m l ) . Meanwhile, there were no differences in maximal uptake and degradation. T h e data indicated increased affinity but not increased capacity of V L D L for L D L

252 1

2

3

:.~'~. . . W ~..-~ ~. ,.~.,/;.=~ . ~"

4

~"

apoE

'Na;~.

1

f,.~ t

2.5

$

.

12SI-LDL (lag / nil) Fig. 7. Uptake of '2~I-LDL in CHO-MAEll (open columns) and control cells (closed columns). The experimental procedures were as described in the legend to Fig. 6 and Material,; and Methods, 12~I-LDL used were 2.5 and 5 /.tg/ml. Cell-associations were expressed as means (column)+_S.D. (bar) ng/mg of quadruplicate assays. receptors in interactions between V L D L and LDL receptors on the transfected cells. This suggested that rat apo E molecules secreted by C H O - M A E I ! were transferred onto ~2SI-VLDL in the medium and that the increase in the number of apo E molecules on V L D L caused the increase in affinity for LDL receptors on the cells, On the other hand, the results of binding assays for 12SI-LDL showed no difference between transfected and control cells at ~.5 and 5 n g / m l (Fig. 7). lmmunoblots of L D L receptors of both cell lines are shown in Fig. 8. There was no significant difference in the amounts of LDL receptor proteins 1

2

3

4

200 k D a - -

116 k D a - 97 k D a

Fig. 8. Immunoblot of LDL receptor in CHO-MAEII. CHO-MAEII (3, 4) and control cells (l, 2) were preincubated in HAM with 10% NCS (l. 3) or 0.2r/~ BSA (2, 4} for 24 h. Cell-lysates (I mg protein/lane) were subjected to immunoblot as described in Materials and Methods using polyclonal antibody against LDL receptor. Molecular weight standards are indicated.

~

Fig. 9. Immunoblot of apo E on reimlated VLDL in the conditioned medium by CHO-MAEll. CHtO-MAEI1(I, 3) and control cells (2, 4) were preincubated in HAM containing 0.2% BSA and 50/tM ZnSO4 for 12 h. Human VLDL (10 ,¢g/ml) was added and the cells were further incubated for 12 h. The conditioned medium was ultracentrifuged and refloated VLDL was subjected to immunobloning using anti-rat aim E antibody (1, 2) or anti-human aim E antibody (3, 4) as described in Materials and Methods. (130 kDa), supporting the results of the L D L binding assay. To confirm the transfer of secreted rat apo E o n t o h u m a n VLDL, the following experiment was performed. C H O - M A E l l was incubated with h u m a n V L D L and the V L D L in the conditioned m e d i u m was reisolated by ultracentrifugation and subjected to immunoblot analysis (Fig. 9). Using anti-rat apo E antibody, rat apo E was detected in V L D L reisolated from C H O - M A E I I , while that from control cells contained only minimal a p e E detectable probably due to cross reactivity with h u m a n a p e E. A n t i - h u m a n a p e E antibody detected a greater a m o u n t of a p e E in V L D L from C H O - M A E I I than that from control cells, suggesting additional rat a p e E onto h u m a n V L D L particles from C H O - M A E I I . We performed another binding experiment using methylated VLDL. N,~ methylated tesI-VLDL was degraded by control cells because reductive methylation of lysyl residues in the binding domain of a p e BI00 a n d / o r a p e E abolishes the binding activities of the ligands to L D L receptors on the cells• However, when methylated te~I-VLDL was a d d e d at 2.5 ~ g / m l , C H O - M A E I ! d e g r a d e d t h e m significantly• T h e amounts of methylated 'a51-VLDL degraded by CHOM A E l l induced and uninduced with zinc ion were 42% and 27% of tasI-VLDL degraded by control cells, respectively (Fig. 10). This indicates that secreted a p e E molecules were exchanged for methylated a p e E or a p e C molecules on tesI-VLDL and caused recovery of

253 pared with control cells (Fig. 11). However, no significant differences in the degradation were found. Discussion

i

2

3

4

s

6

Fig. 10. Degradationof methylated II~I-VLDLin CHO-MAEll and control cells, i~-~I-VLDLwas methylatedas described in Materials and Methods. CHOoMAEll(3, 4, 5. 6) induced(5, 6) or uninduced (3, 4) with zinc ion and controlcells (1.2) treated with zinc ion were subjected to the experimentalproced-res as de~ribed in the legend to Fig. 6 and Materialsand Methods. t~I-VLDL (I, 3. 5) or methylated 8251-VLDL(2. 4, 6) was added to the medium at the concentration of 2.5 p.g/ml. Degradationwas expressedas means(columns) :i:SD (bars) (ng/ms) in triplicateasSaysof two independentexperiments. binding activity of methylated t251-VLDL for LDL receptors. No methylated 1251-LDL was degraded by CHO-K or CHO-MAEII cells (data not shown). In comparison, binding assays for 12-*I-HDL.~ were also carried out. A slight increase in the uptake of I~-~IHDL 3 by the transfected cells was observed as c o r n -

1/o 2~

t2s I-IIDL p g / m !

Fig. 11. Saturation kinetics in cell-associationand degradation of tz~I-HDL in CHO-MAEII. The experimentalprocedures were de~ribed in Materialsand Methods. The indicated amounts of la~IHDL3 were added to the media. Cell-associationand degradationof CHO-MAEil(o) and controlcells(