Biochimica et Biophysica Acta, 1090(1991)95-101

95

© 1991 ElsevierSciencePublishersB.V. 0167-4781/91/${)3.50 ADONIS 0167478191002043 BBAEXP~ 2 ~

Expression of LDL receptor, apolipoprotein B, apolipoprotein A-I and apolipoprotein A-IV mRNA in various mouse organs as determined by a novel RNA-excess solution hybridization assay R a i A j i t K. S r i v a s t a v a , B a r b a r a A . P f l e g e r a n d G u s t a v S c h o n f e l d Dicision of Atherosclerosis and Lipid Research. Department of Internal Medicine, Washington Unirersi~' School of Medicine. St. Louis, MO (U.S.A.)

(Received 13 November1990)

Keywords: Solutionhybridizationassay:Low densitylipoproteinreceptor; ApolipoproteinB: ApolipoproteinA-i;Apolipoprotein A-IV;(Mouseorgan)

We report expression of LDL receptor, apolipoprotein B (apoB), apolipoprotein A-! (apoA-l) and apolipoprotein A-IV (apoAIV) mRNA in various mouse organs. These mRNA were quantified by an RNA-excess solution hybridization assay. For preparing specific probes, we cloned cDNA fragments of rat LDL receptor, apoB and apoA-i and mouse almA-IV into the polylinher region of pGEM3Zf(+) and used the recombinant vectors for preparing azP-laheled cRNA probes as well as RNA standards using the T7 and SP6 promoters flanking the polylinher regions. Preparation of cRNA probes and RNA standards is faster and more convenient than preparing cDNA probes and ssDNA standards. Absolute levels of mRNA were quantified in the liver, intestine, kidney, heart, lung, spleen and adrenals of females of two mouse strains, C 3 H / H e J and C57BL/6J. ApoB, apoA-! and apoA-/V genes in mice are expressed in the liver and intestine and LDL receptor gene is expressed mainly in liver, intestine and adrenals. APOA-! mRNA levels were found to be 730 and 1039 molecules per cell in fiver and intestine, respectively, in C3H mice and 762 and 952 molecules per cell in C57BL mice. ApoB mRNA levels were 66 and 170 molecules per cell in the liver and intestine of C3H and 83 and 243 molecules per cell in C57BL, respectively. APOA-IV mRNA was found to be 3525 and 2964 molecules per cell in the liver and intestine of female C57BL mice, respectively. LDL receptor mRNA levels were 39, 32 and 14 molecules per cell in the liver, intestine and adrenals of C3H.

Introduction Elevated levels of LDL cholesterol and apoB are associated with an increased risk of atherosclerosis, whereas high levels of HDL cholesterol and apoA-I decrease the risk of atherosclerosis [1-4]. Alterations in the metabolism of LDL receptors [5], apoB and apoA-! affect the level of these constituents in the plasma, and factors like diets, hormones, growth and development, and genetics modulate LDL receptor activities, and apoB, apoA-1 and apoA-IV levels in

Abbreviation:Apo, apolipoprotein. Correspondence: R.A.K.Srivastava.Divisionof Atherosclerosisand Lipid Research. Department of Internal Medicine,4566 Scott Avenue. CampusBox 8046.WashingtonUniversitySchoolof Medicine. St. Louis, MO 63110. U.S.A.

plasma. For studies designed to understand the effect of these factors on the metabolism of LDL receptor, apoB, apoA-I and apoA-IV, it is necessary to quantify these mRNAs in tissues. Most frequently, specific message levels are quantified either by slot blot or Northern blot analysis [6]. However, RNA blot analyses are semiqaanti~ative and unreliable for the quantification of small changes in specific messages [7]. Moreover, RNA blotting methods require several days, and provide only relative and not the absolute values. A more sensitive method, DNA-excess solution hybridization, has been developed that does provide absolute levels of mRNAs values [8], but this method requires a lengthy probe and ssDNA preparation time and the quantification of mRNA values are based on a standard curve generat,~d by the hybridization of ssDNA with a eDNA probe. We have developed RNA-excess .solution hybridizat i o n / RNase A protection assays using specific ribo-

96 probes for the quantification of LDL receptor, apoB, apoA-1 and apoA-IV mRNA. For these assays, probe and RNA standard can be prepared using the same recombinant vectors iinearized at two different sites. Preparation of probe and standard takes about 2 h. We have used this technique to study the expression of LDL receptor, apoB, apoA-I and apoA-IV mRNAs in various organs of two inbred mouse strains and show that liver and intestine are the main organs in mice for apoB, apoA-I and apoA-IV gene expressions. LDL receptor mRNA is expressed mainly in liver, intestine and adrenals. Materials and Methods

quick frozen in liquid nitrogen. Total RNA was isolated by guanidinium-HCi isothiocyanate method coupled to CsCl 2 ultracentrifugation [11]. RNA precipitates obtained were further purified by phenol/ chloroform extraction and ethanol precipitation. For Northern blotting analysis, 10 or 15 lzg of total RNA were electrophoresed in 1.5% agarose gel containing formaldehyde as described [6]. Relative amounts of RNA were compared in each lane by ethidium bromide staining of the gels. RNA from the gel was transferred to Nylon membrane (GeneScreen, NEN) by capillary blotting, baked for 2 h at 80°C and hybridizations were carried out with [ 32p]pi.labeled cRNA probe at 60°C as described [12].

Materials 3-4-month old female mice of the strains C3H/HeJ and C57BL/6J were used in the present investigation. Escherichia coil strain JM109 was transformed with pGEM3Zf(+) plasmids procured from Promega and used as a cloning vector. Restriction enzymes were obtained from Boehringer-Mannheim Biochemicals, New England Biolabs and Promega. A 1.3 kb ~DNA insert of rat LDL receptor cloned into EcoRI site of the bacteriophage M13mpl8 (clone 26.4-1) was obtained from Drs. A.D. Cooper and Y. Nagata, Research Institute, Palo Alto Medical Foundation, Palo Alto, CA, a 240 bp rat apoB cDNA in pBR325 (clone 49a, Ref. 9) and 850 bp mouse apoA-IV eDNA in PUC13 from Dr. A.J. Lusis, University of California, Los Angeles, CA and a 555 bp rat ApoA-! eDNA in pGEMI from Dr. J.l. Gordon, Department of Biologit.~l Ch~.mistry and Molecular Biology, Washington University School of Medicine, St. Louis, U.S.A.

Molecular cloning of cDNA fragments cDNA fragments were cloned into the polylinker region of the pGEM vector employing standard cloning techniques [6]. A 447 bp fragment of rat LDL receptor cDNA was excised from the LDL receptor clone 26.4-1 by Pstl digestion and cloned into the Pstl site of linearized pGEM3Zf( + ). Similarly, a 240 bp fragment of rat apoB cDNA and a 555 bp of rat apoA-I cDNA were excised from apoB and apoA-! clones and cloned into the EcoRI site of the pGEM vector. An 850 bp mouse apoA-IV cDNA was excised from PUC13 by Pst! and cloned into the Pst I site of pGem vector. Ligation of inserts and transformation of E. coli strain JMI09 with the recombinant pGEM plasmid was done following standard methods. The orientation of the inserts in the plasmid were confirmed by sequencing double stranded plasmids as described [10].

Isolation of RNA and Northern blotting analysis Five mice of each strain, C3H/HeJ and C57BL/J were killed and various organs removed, weighed and

Preparation of riboprobes and RNA standard The recombinant pGEM plasmids were linearized by appropriate restriction enzymes and antisense cRNA probes were synthesized (Fig. 1) in a total volume of 20 /~l following the method supplied in the Promega Riboprobe kit. The reaction mixture contained: 4 /~l (5 x ) transcription buffer, 2/zl dithiothreitol (100 raM), 1 /~l each of 10 mM solution of GTP, ATP and CTP, 2.4/zl UTP (100/zM), 1 /tl RNasin (40 U//~l), 2 p.! linearized plasmid (1 /~g), 5 /zl [a-32p]UTP (600 mCi/mM, ICN Biomedicals), 1 /zl T7 or SP6 RNA polymerase and DEPC-water to make 20/zl voi. The transcription reaction was allowed to proceed for 1 h at 37°C followed by addition of 1 /zl DNase I (RNase free, Boehringer Mannbeim Biochemicals) for an additional 10 rain at 37°C. After the DNase i treatment 5 p.I DEPC were added and vortexed vigorously. 80/zl water were added, extracted with phenol/chloroform, ethanol precipitated, washed twice with 70% ethanol, dried and dissolved in 25 /~l TE buffer. Riboprobes were further purified on Quick Spin Column (BMB) to remove unincorporated label. 32p-labeled riboprobes were aliquoted into 5/~l vol. and frozen immediately at -70°C. Frozen riboprobes work well up to 2 weeks. The incorporation of radioactivity was monitored by counting 2 /zl of appropriately diluted probe. Trichloroacetic acid (TCA) precipitable counts were also measured by precipitating 1/tl of riboprobe and filtering through glass fiter. For the preparation of unlabeled RNA standards from each recombinant plasmid, the transcription reaction was performed in a total volume of 50/~l. Plasmids were linearized by Noel (for LDL receptor and apoA-l) or Sphi (for apoA-IV). All the ingredients were same as above except for the additional unlabeled UTP. The concentration of purified RNA standard was determined spectrophotometrically. For all RNA preparations the ratio of A2~a/A~j was found to be close to 2.0. RNA standard were frozen in aliquots at -70°C. Each aliquot was used only once.

97

Solution hybridization assay Hybridizations were carried out in a total volume of 30 /zi containing 40% formamide/400 mM NaCI/40 mM Pipes buffer (pH 6.7)/1 mM EDTA. RNA samples (50 p.g for LDL receptor, 40 g g for apoB and 10 /zg for apoA-I and apoA-IV mRNA quantification) were either dried in a Speed Vac or precipitated with ethanol, dried and dissolved in 28 /zl hybridization buffer. 2 /zl of riboprobe (40000 cpm) was added to each tube and covered with 10/zi parafin oil to prevent the loss in volume. The optimized temperatures for hybridization have been determined to be 63, 55, 65 and 60°C for LDL receptor, apoB, apoA-! and apoA-IV mRNA, respectively. Hybridization was performed overnight (18 h). After hybridization, the solution was mixed with 300 /zl RNase buffer (300 mM NaCl, 10 mM Tris-HCI (pH 7.5), 5 mM EDTA, 40/Lg/ml RNase A, 2 / ~ g / m l RNase TI) and incubated for 1 h at 30°C. The tubes were then transferred to ice bath and 1 ml TCA solution (10% TCA, 1.5% sodium pyrophosphate) was added. The tubes were kept on ice for 15 min and then filtered through G F / C glass fiber filter (S&S). Extensive washing of the filters was performed with cold TCA solution, dried at 80°C in an oven for 1 h and counted in 4 ml scintillation fluid in a Beckman Scintillation Counter.

Results T7 and SP6 R N A polymerases were used for preparing c R N A probes and R N A standards. The use of either o f these R N A polymerases for making riboprobes depends upon the orientation o f the insert in the vector as determined by sequencing the double stranded recombinant plasmid of each clone. Since c D N A fragments of apoA-! and apoB were cloned

into the EcoRl site which is the first restriction enzyme from "1"7 RNA polymerase promoter (Fig. 1), there was no restriction site left to linearize the recombinant vector for the SP6 RNA polymerase directed transcription. We, therefore, linearized the recombinant plasmid at Nael site (a site at nucleotide 2892 in the vector) and used this linearized plasmid to transcribe RNA standard for apoA-I and cRNA probe for apoB by SP6 RNA polymerasc. The RNA standard contain an extra 307 nucleotides from the vector. Therefore, a correction was made for these extra nucleotides, while calculating the absolute values for the specific messages. Conditions for optimum hybridization were established. The standard curves generated by hybridization of cRNA probes with the RNA standard are shown in Fig. 2. To confirm that the riboprobes hybridize to the respective mRNA, Northern

.....

1/I

i: LO~)tm

S P 6 ~-N~dm S6

~ j

,Hindln, cRNA Probe 1 " -~"-SP6 RNA STANDARD

TT~

jA

Nael T7 - - "

Nael

emil,,

~-" SP6 TT'~

Nael

S~emce

..... ~

SP6

EcoRI --

,m- SP6

T7..-~

Sphl

nNASTAm~B

cRNAProbe

J

KI~I, ©RNAProbe 1 flNA STANDARD cRNA Probe RNA STANDARD

JC 1 JO

Fig. I. Strategyfor preparingrilx)probesand RNA standards.On the top the vector and the polylinker region is shown.Panel A, LDL rcccplor probe and RNA standardpreparation;Panel EL,apoB eRNA probe and RNA standardpreparation;Panel C, apoA-i probe and RNA standard preparation; and Panel D, apoA-IV probe and RNA stana~rd preparation. 1"7 and SP6 RNA po|ymera.~-~w¢,~cused for probe and RNA standardpreparationas indicated.

98

A U 14

B I

U

I

K

H

L

S

A

28S-

I

18So

-28S

-18S

.i.

i

| m

' loo

~o

3~o

,~o

Soo '-

pg cRNA Standard

Fig. 2. Standard curve generated by hybridization of cRNA probes with RNA standrad. For details see the text. *, LDL receptor; e, apoA-I; o, apoB. Each point in the figure represents the average of two estimations and the variation in two estimations were lower than 10%.

Fig. 3. Northern blotting analysis for LDL receptor and apoA-I mRNAs. 10 /,tg RNA was electrophoresed in 1.5% formaldebydeagarose gel, blotted to GenScreen membrane and hybridized to the respective cRNA probes. Panel A, LDL receptor and Panel B, apoA-I mRNAs. Li, liver; I, intestine; K, kidney; H, heart; L, lung; S, spleen and A, adrenal.

detected in kidney and heart, but none could be detected in lung, spleen and adrenals. We compared blotting analysis was p e r f o r m e d using c R N A probes

(Figs. 3 and 4). These riboprobes hybridize to the specific mRNA. Levels of LDL receptor, apoA-i, apoB a n d a p o A - I V m R N A were quantified in the total R N A isolated from various mouse organs (Tables I - V ) . Values are expressed as pg m R N A p e r / ~ g total R N A o r as molecules p e r cell [8]. T h e b a c k g r o u n d counts were found to be a b o u t 0.1% of the total p r o b e input. Fig. 3 a n d Tables I a n d !1 clearly show that apoA-I m R N A is synthesized mainly in the liver a n d intestine with levels in C 3 H / H e J mouse strain being higher in intestine t h a n in liver. Very low levels of apoA-I m R N A were

a p o A - I m R N A levels o b t a i n e d from s t a n d a r d curves g e n e r a t e d e i t h e r by using R N A o r s s D N A standards. H i g h e r values were o b t a i n e d w h e n s s D N A s t a n d a r d was used (data not shown). A p o B m R N A was also found to be synthesized only in liver a n d intestine (Table !II). L D L r e c e p t o r m R N A levels were highest in liver, intestine a n d adrenals, b u t low levels o f L D L r e c e p t o r gene expression also were o b s e r v e d in o t h e r o r g a n s (Table IV). L D L r e c e p t o r m R N A levels were higher in liver t h a n in intestine. A p o l i p o p r o t e i n A - I V m R N A was also quantified in various m o u s e o r g a n s a n d the results are shown in T a b l e V. It can b e seen

TABLE I ApoA-! mRNA lerels in rarious mouse organs of strain C3H / HeJ

Organ Liver Intestine Heart Kidney

Organ weight 0.734 -

Lung

0.121 0.082 0.104

Spleen Adrenal

0.106 0.035

ApoA-I mRNA (pg//~g RNA) 25.3 ±2.5 36.9 ±3.1 1.52± 0.39 1.20± 0.40 N.D. c N,D. N.D.

Total RNA ~ (mg/g tissue) 9.44 5.00 2.80 5.92 7.40

Total DNA b (mg/g tissue) 2.95 1.54 ! .26 2.66 2.94

ApoA-I mRNA (molecule per cell) 730 1039 30 24 N.D.

5.56 3.92

i 1.48 2.66

N.D. N.D.

• ,b Total RNA and D N A were measured by colorimetric methods as described by Williams et al. [8]. Calculation for molecules/cell are based on diploid D N A content of 6.4 pg/ocll [22]. c N.D. indicates not detectable.

99

A 1

B 2

1

TABLE IV Let'els o f L D L receptor mRNA in t'arious mouse (C3H / HeJ) organs

2

Organ

ApoS-

-18S

•

8S a

t-ApoAlV

Liver Intestine Heart Kidney Lung Spleen Adrenal

LDL receptor mRNA (pg/#g RNA) 5.04 _+0.36 4.12-+0.31 0.72 .+0.12 0.59 .+0.08 0.45 + 0.15 0.25 -+0. Ifl 3.80± 0.29

LDL receptor mP NA (molecules/cell) 39 32 4 3 3 0.5 !4

TABLE V

18,$-

Apolipoprotein A - I V m R N A contents in carious moux~ organs o f C57BL strain

Mouse organs

Fig. 4. Northern blotting analysis of aooB and apoA-IV mRNA using cRNA probes. 15 ttg (for apoB) and 10 p.g (for aooA-IV).RNA were electrophoresed in 1.5% formaldehyde containing agarose gel, blotted on to GenScreen membrane and hybridized to the respective riboprobes. Pane A represents apoB and Panel B apoA-IV mRNA detection. 1 and 2 represent liver and intestine.

Liver Intestine Heart Kidney Lung Spleen

ApoA-IV mRNA (pg/~tg RNA) 226 + 13 190.+ 10 1.0 _+0.2 0.76 _+0.4 I n.d. 3.2 _+0.7

ApoA-IV mRNA (molecules/cell) 3525 2964 -

t h a t the a p o A - I V gene is expressed mainly in the liver a n d intestine. This was also confirmed ny N o r t h e r n blotting analysis (data not shown).

TABLE I1 ApoA-! m R N A content in t'arious mouse organs o f strain C57BL / 61

Discussion Organ Liver Intestine Heart Kidney Lung Spleen Adrenal

ApoA-I mRNA (pg/~tg RNA) 26.4 + 1.9 32.6 + 1.8 1.91 ± 0.34 0.42 _+0.30 N.D. N.D. N.D.

AooA-I mRNA (molecules/cell) 762 952 38 9 N.D. N.D. N.D.

TABLE !!! Lel'els o f apoB m R N A in L'arious mouse organs

Organ

Liver Intestine Heart Kidney Lung Spleen Adrenal

Strains C3H/HeJ pg/~tg RNA 35.8 _+3.6 44.2 +_3.8 N.D. 2.02 ± 0.24 N.D. 1.01 +_0.15 N.D.

C57BL/6J molecules per cell 66 170 N.D. 4 N.D. 2 N.D.

pg/~tg RNA 45.2 +_ 4.6 62.8 + 10.2 N.D. 3.20-+ 0.8 N.D. N.D. N.D.

molecules per cell 83 243 N.D. 6 N.D. N.D. N.D.

W e describe a solution hybridization assay for determining t h e absolute levels o f L D L receptor, a p o A - l , a p o B a n d a p o A - I V m R N A in different m o u s e organs. This m e t h o d is simpler t h a n o t h e r m e t h o d s employing c D N A probes. T h e same r e c o m b i n a n t vector can be used for p r e p a r i n g b o t h the c R N A p r o b e s and R N A standards. It takes a b o u t 2 h for the p r e p a r a t i o n o f both, w h e r e a s p r e p a r a t i o n of c D N A p r o b e a n d s s D N A s t a n d a r d takes 2 - 3 days. W e have e n c o u n t e r e d difficulties in p r e p a r i n g s s D N A s because of c o n t a m i n a t i o n with the h e l p e r p h a g e t h a t affects t h e m e a s u r e m e n t s o f s s D N A c o n c e n t r a t i o n s using s p e c t r o p h o t o m e t r i c m e t h ods. N o r t h e r n blotting analysis showed t h a t t h e m e t h o d employing riboprobes detects the specific message a n d does not cross-hybridize to o t h e r RNAs. F u r t h e r m o r e , RNA-excess solution hybridization follows R N A . R N A hybridization kinetics, the affinity o f e R N A probes for m R N A s are g r e a t e r c o m p a r e d t o e D N A probes, a n d R N A . R N A hybrids a r e m o r e stable t h a n R N A . D N A hybrids [13]. T h u s t h e m e a s u r e m e n t s using R N A p r o b e s a n d s t a n d a r d s are m o r e likely to b e accurate a n d we have f o u n d t h e m to b e m o r e precise. A f t e r t h e N o r t h e r n blotting analysis using riboprobes, we e x a m i n e d w h e t h e r t h e m e m b r a n e could b e used for reprobing.

100 Since information on the stripping of riboprobe from the membrane is not available, we employed the same procedure of stripping the probe as used for the eDNA probe, but using a higher temperature (800C) and a prolonged duration (5 h). This treatment failed to strip riboprobe from the membrane, indicating a strong and stable RNA.RNA hybrid. The principle of the current quantification method is the same as that reported by Azrolan and Breslow [12], but the purification of in vitro transcribed riboprobes differs from ours. They suggested that the riboprobes should be used within 16 h of synthesis, but we have found that the purified riboprobes could be stored frozen in aliquots at - 70"C for at least 2 weeks. In our qu~,ntifications for various mRNAs, we did not use any internal standard, although internal standards have been used by others [7,12]. Internal standards are necessary in assays involving extract!~n, precipitation and analysis of protected fragments by gel electrophoresis [7] where losses i~ hybrids are likely to occur. We used TCA precipitation and collection of hybrids on a glass fiber filter which minimizes the loss of hybrids. The same technique was used by Azrolan and Breslow [12] who also found that the internal standard gave almost the same counts when added to each R N A samples. We found that the TCA precipitation and counting method is reproducible. ApoB, apoA-I and apoE mRNA were quantified in various cell lines using RNA-excess solution hybridization [12], but not in animal tissues. Pape et al. [7] reported that it is difficult to analyze specific mRNAs by RNA slot-blotting analysis using RNA isolated from fatty tissue, but they could successfully quantif~ apoA-! mRNA by DNA-excess solution hybridization in the same RNA samples. We did not find any difficulty in quantifying mRNA, either by RNA-excess solution hybridization or by Northern blotting analysis (data not shown). Absolute values of apoB, apoA-I and apoA-IV mRNAs have not b,-~n previously reported in mouse. However, chicken liver and intestine are reported to contain 1488 and 1642 molecules of apoA-! per cell in assays using eDNA probes [14]. We found 730 and 1039 molecules of apoA-I in liver and intestine of the C 3 H / H e J mouse strain and 762 and 952 molecules per cell in the C57BL mouse. Miller et al. [15] also showed that apoA-! is expressed mainly in the liver and intestine of mouse. The only other animal in which absolute levels of apoA-I have been determined is the non-human primate [16]. Unlike avian apoA-I mRNA, which is found in many organs, the mouse apoA-! gene is not expressed in organs other than liver and intestine. The ApoB gene is also expressed mainly in the liver and intestine, similar to rat [17] and rabbit [18]. Absolute levels of apoB mRNA in mouse are lower than in primates [19] and HepG2 [12] cells, while absolute levels of LDL receptor mRNA in mouse were compa-

rable to those of primates [19] and rabbits [20]. Apolipoprotein A-IV mRNA have been quantified in mouse by Northern blotting analysis using the cDNA probe [21] and it was shown that different strains of mouse differed in their apeA-IV m R N A levels. Absolute values of apoA-IV mRNA have not been determined in any animal tissue or cell line. We have surveyed various mouse organs for apoA-IV m R N A content and found th~* apoA-IV m R N A is expressed mainly in the liver and intestine of female mice. The cRNA probe used in Northern blotting analysis for detection of a p o A - l ' / m R N A does not cross hybridize to other RNAs like 18S RNA, as reported using eDNA probes [21]. Thus, riboprobes are more specific and can be suitably used for routine Northern blotting analysis. In summary we have described an RNA-excess solution hybridization assay for the quantification of apolipoprotein mRNA levels and have reported for the first time the absolute values of LDL receptor, apoA-I, apoB and apoA-IV m R N A in mouse. We have also shown that apoA-l, apoB and apoA-IV genes are expressed mainly in the liver and intestine, while the LDL receptor gene is expressed at low levels in several organs in addition to liver, intestine and adrenals. Acknowledgements This work was supported by grant PO1 DK 334870AI and RO1 HL 42460-01. R.A.K. Srivatava is a Fellow of the American Heart Association, Missouri Affiliate. References I Mahley, R.W. (1982) in Medical Clinics of North/~merica, Symposium on Lipid Disorders (Hard, RJ., ed.) 60, 375-401. 2 Kwiterovitz, P.O. and Sniderman, ll.D. (1983) Prey. Med. 12,

815-834. 3 Kukita,H., Hiwada. K. and Kokubu,T. (1984) Atheroscierosis51, 261-267. 4 Kukila, H., Hawada, K., Hiwada, K. and Kokubu, T. (1985) Atherosclerosis55, 143-149. 5 Goldstein. J.L. and Brown, M.S. 11985) Annu. Rev. Biocbem. 1-39. 6 Maniatis, T., Fritsch. E.T. and Sambrook, T. (1982) Molecular Cloning. A LaboratoryMannual,Cold SpringHarbor Laboratow, Cold Spring Harbor, New York. 7 Pal)c, H.E., Maroni, K.R. and Melchiar, G.W. (1990) J. Lipid Res. 31,727-733. 8 Williams,D.L, Newman, T.G., Shelvess,G.S. and Gordon, D.A.

(1986) Methods Enzymol. 128, 671-688. 9 Lusis, AJ., West, R., Mehrabian, M., Reuben, M.A, LeBoeuf, R.C., Kaptein, J.S., Johnson, D.F., Shumaker, V.N.. Yuhasz, M.P., Scotz, M.C and Elovson, J. (1985) Proc. Nail. Acud. Sci. USA 82, 4597-4601. 10 Mierendorf, R.C. and Pfeffer, D.C. (1987) Methods Enzymol. 152, 556-562. !1 Davis,L.G., Dibner, M.D. and Battey, J.F. (1986) Basic Methods In MolecularBiology,p. 130, Elsevier,New York. 12 Azrolan,N. and Breslow,J.L (1990)J. Lipid Res. 31, 1141-i 146.

I01 13 Thomson, J., and Gdlespie. D. (19871 Anal. Biochem. 163. 281291. 14 Rajavashistha, T.B., Dawson, P.A.0 Williams, D.L., Shakelford, J.E., Leberz, H. and Lusis, A.J. (19871 J. Biol. Chem. 262, 7058-7065. 15 Miller. J.C.E., Barth, R.K., Shaw, P.H., Elliott, R.W. and Hastie, N.D. (19831 Proc. Natl. Acad. Sci. USA 80, 1511-1515. 16 Sorci-Thomas, M., Prack, M.M., Dashti, hi., Johnson, F.. Rudel. L.L. and Williams~ D.L. (19881 J. Biol. Chem, 263. 5183-5189.

17 Matsumoto, A., Aburatani, H., Shibasaki, Y., Kodama, T.. Takaku, F. and ltakura, H. (1987) Biochem. Biophys. Res. Commun. 142, 92-99.

18 Kroon, P.A., DeMartino. J.A., Thomson, G.M. and Chao. Yushang (19861 Proc. Nail. Acad. ~-i USA 83, 5071-5075. 19 Solci-Thomas, M., Wilson, M.D., Johnson. F.L, Williams, D.L. and Rudel, L.L. (19891 J. Biol. Chem. 264, 9039-9045. 20 Ma, P.T.S., Yamamoto. T. Goldstein, J.L. and Brown, M.S. (19861 Proc. Natl. Acad. Sci. USA 83, 792-796. 21 Williams, S.C., Grant. S.G., Reue, K., Carrasquillo. B., Lusis, A.J. and Kinninhurg, A.J. (19891 J. Bnol. Chem. 264, 191~9-19016. 22 Durnam. D.M. and Palmitor, R.D. (19831 Anal. Biochem. 13. 385 -393.