THE JOURNAL OF BIOLOGICAL CHEMISTRY Vol. 259, No. 10, Issue of May 25, pp. 6459465,1984 Printed in U.S.A.

Rat Liver and Small Intestine Produce Proapolipoprotein A-I Which Is Slowly Processed to Apolipoprotein A-I in the Circulation*

(Received for publication, September 21,1983)

Mary B. SliwkowskiS and Herbert G. Windmueller From the Laboratory of Cellular and Developmental Biology, National Institute of Arthritis, Diabetes, and Digestive and Kidney Diseases, National Institutes of Health, Bethesda, Maryland 20205

Two-dimensional electrophoretic analysis of plasma lipoproteins from male Osborne-Mendel rats consist- ently reveals three isoforms of apolipoprotein A-I (apo- A-I) with the following apparent PI values and quan- titative distribution: isoform 3, PI = 5.68,69%; isoform 4, PI = 5.55, 29%; isoform 5, PI = 5.44, 2%. The two major isoforms were obtained by preparative isoelec- tric focusing and subjected to NH,-terminal amino acid sequence analysis with the following results: isoform 3, (Asp)-Glu-Pro-Gln-Ser-Gln-Trp-Asp-Arg-Val; iso- form 4, X-Glu-Phe-X-Gln-Gln-Asp-Glu-Pro-Gln-Ser. By comparison with the amino acid sequence previ- ously reported for the primary translation product of rat intestinal apo-A-I mRNA (Gordon et al. (1982) J. Biol. Chem. 257,971-978), isoform 3, the more basic isoform, is identified as mature apo-A-I and isoform 4 as its proform (proapo-A-I). The proform differs from mature apo-A-I by a 6-amino acid extension at the NH, terminus. Isoform 5 was not identified further. The plasma steady state distribution of the apo-A-I forms indicates that proapo-A-I is relatively stable in the circulation. Virtually all plasma proapo-A-I is lipopro- tein-associated. No significant differences in the steady state proportions of plasma apo-A-I forms were observed between male and female rats, or among var- ious subfractions of plasma high density lipoproteins obtained by heparin-Sepharose affinity chromatogra- phy or by density gradient ultracentrifugation. Rats fed a high fat, high cholesterol diet, however, showed an increase in the proportion of circulating proapo-A- I. The relative increase in proform was even more pronounced in rats fed a fat-free diet containing orotic acid.

The biosynthesis, secretion, and metabolism of the various apo-A-I forms were also studied. In liver and intestine, the only known sites of apo-A-I synthesis in the rat, -85% of the newly synthesized intracellular apo-A-I, was the proform. Proapo-A-I was also the predominant form (-80%) released into the circulation by isolated, perfused livers and by autoperfused intes- tinal segments in vivo. Gradual processing of circulat- ing proapo-A-I to mature apo-A-I was observed in vivo following pulse-labeling of apo-A-I with r3H]leucine. Processing in vivo was -80% complete in 10 h. In heparinized blood or plasma incubated in vitro at

* A preliminary report of this work was presented at the annual meeting of the American Society of Biological Chemists, San Fran- cisco, California, June 5-9, 1983, and published in abstract form (1). 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.

$. To whom all correspondence should be addressed at: Building 10, Room 5N-102, National 1nst.itutes of Health, Bethesda, MD 20205.

~ "~ " __

37 "C, processing reached only 22% in 10 h, nearly all during the first 2 h; processing was totally inhibited when EDTA or EGTA plus Mg2' replaced the heparin, suggesting a divalent metal requirement. Also, no sig- nificant processing was observed during continuous recycling for up to 5 h through isolated rat livers either in the presence or absence of serum. Thus, a labile plasma factor may participate in the proteolytic cleav- age of proapo-A-I.

The inverse correlation between plasma HDL' levels and development of coronary artery heart disease has been well documented ( 2 ) . However, the role of HDL in lipid transport and metabolism is not well understood. Apo-A-I is the major apoprotein of HDL and as such may be important to an understanding of the biochemistry of atherosclerosis. Apo-A- I is thought to occupy a surface position on HDL particles, surrounding a neutral lipid core (3). HDL-like particles can be reconstituted in aqueous solution from lipid and apoprotein components (4). Apo-A-I may have a structural role in direct- ing the assembly of such particles ( 5 ) . Additionally, apo-A-I activates 1ecithin:cholesterol acyltransferase (6), which is the cholesterol-esterifying enzyme of plasma and is involved in the production of mature circulating HDL (7).

The amino acid sequences of human (8) and canine (9) apo- A-I have been determined. Rat apo-A-I has a similar compo- sition (10). Although various peripheral tissues of the chicken appear to synthesize apo-A-I (ll), liver and intestine are the only known significant sites of synthesis in the rat (12), mouse (13j, and baboon (14). Using human liver and intestine cul- tured in uitro, Zannis and co-workers (15, 16) reported that apo-A-I occurs as several isoproteins with the same molecular weight but different isoelectric points. Furthermore, the two major tissue isoforms were more basic than the two major isoforms of plasma. Rat serum apo-A-I has also been shown to occur as isoforms (17). Ghiselli et al. (18) found that rat liver produces a pattern of isoforms that is quantitatively different from that of plasma. Cell-free translation of rat liver (19, 20) and intestinal (21) apo-A-I mRNA has recently been reported. Gordon et a/. (21) also determined the NH2-terminal amino acid sequence of rat plasma HDL apo-A-I. By compar- ing this sequence with that of the primary translation product, they determined that apo-A-I is initially synthesized as a larger precursor which is co-translationally processed by mi- crosomal membranes. In addition to an 18-amino acid preseg-

The abbreviations used are: HDL, high density lipoproteins; apo- A-I, apolipoprotein A-I; proapo-A-I, the proform of apo-A-I; SDS, sodium dodecyl sulfate; Hepes, 4-(2-hydroxyethyl)-l-piperazineeth- anesulfonic acid; EGTA, ethylene glycol bis(0-aminoethyl ether)- N,N,N',N'-tetraacetic acid.

_____

6459

by guest on Novem

ber 8, 2018http://w

ww

.jbc.org/D

ownloaded from

6460 Processing of Rat Proapolipoprotein A-I to Mature Form

ment, an unusual hexapeptide prosegment was identified. The prosegment was presumed to be removed in a post-transla- tional event(s). The NH,-terminal sequence of the plasma apo-A-I reported by Gordon et al. (21) differed, after the first three amino acids, from the sequence reported by Lin-Su et al. (19). However, in each case, only one sequence was detected in the plasma and this was cited as evidence for occurrence of a single plasma apo-A-I species. These data suggested that removal of the prosegment must occur before or immediately after secretion of apo-A-I.

The present studies were initially undertaken in an attempt to explain the steady state apo-A-I isoform pattern of rat plasma. Additionally, we investigated the relationship of the isoforms to the post-translational processing of the apopro- k in . Employing in uiuo labeling wherever possible, newly synthesized intra- and extracellular forms of apo-A-I have been examined..Both rat liver and intestine have been found to synthesize and secrete an isoform which we subsequently identified as the proform of apo-A-I. We find rat proapo-A-I to be a stable plasma species that is slowly converted in the circulation to the more basic "mature" apo-A-I. The conver- sion proceeds to a limited extent in blood incubated in vitro and exhibits an apparent requirement for divalent metal. Further, the hexapeptide prosegment does not appear to sub- stantially alter the ability of apo-A-I to associate with lipid.

EXPERIMENTAL PROCEDURES

Rats and Materials-All rats were 300-400-g Osborne-Mendel males fed NIH-07 open formula stock diet unless otherwise noted. L- [4,f1-~H]Leucine (40-65 Ci/mmol) was from Amersham Corp., Arling- ton, IL. Ampholine, pH 4-6, was obtained from LKB. Urea was the highest grade available and all solutions were freshly prepared and deionized with Rexyn 1-300 mixed bed resin from Fisher Scientific. All other electrophoresis chemicals were from Bio-Rad. PD-10 gel filtration columns and heparin-Sepharose were from Pharmacia. Aprotinin, 6000 kallikrein inhibitor units/mg, was obtained from Boehringer Mannheim.

Lipoprotein Isolation-Total lipoproteins, d < 1.21 g/ml, were isolated by ultracentrifugation (12). All salt solutions contained 3 mM EDTA, 0.1% NaN3, pH 7.4. HDL, d = 1.040-1.21 g/ml, were prepared similarly except that lipoproteins with d < 1.040 g/ml were removed following an initial centrifugation at d = 1.040 g/ml.

Two-dimensional Electrophoresis-Lipoprotein samples were de- salted by gel filtration or dialysis with 0.1 M NH4HC03, pH 7.0. Desalted samples were lyophilized and initially they were then de- lipidated with chloroform-methanol (18). Subsequently, delipidation was found to be unnecessary. Dried samples were dissolved in 8 M urea, unless otherwise noted, for electrophoretic analysis.

Two-dimensional analysis with isoelectric focusing as the first dimension and electrophoresis in SDS as the second dimension was conducted essentially according to Ghiselfi et al. (18). Focusing gels contained 8 M urea and 2% pH 4-6 ampholytes. Good agreement was obtained between pH measurements made directly on the gel with a surface electrode or following extraction from 5-mm gel slices. Ver- tical slab gels (1.5 mm thick) containing 0.1% SDS were prepared with the pH 8.8 Tris buffer of Laemmli (22). The electrode buffer contained 0.125% SDS, 31 mM Tris, and 0.24 M glycine at pH 8.2. Slab gels were stained with 0.25% Coomassie blue R-250 in 50% methanol, 10% acetic acid. Stained areas were cut from the gels and digested in scintillation vials with 30 pl of 60% perchloric acid and 300 pl of 30% H202 at 70 "C for 3-4 h. Ready-Solv H P (Beckman), 10 ml, was added and the vials were held overnight before counting. Protein in two-dimensional gel spots was quantitated by the method of Fenner et al. (23) involving pyridine extraction of dye from macer- ated gel pieces.

Preparative Isoelectric Focusing and Amino Acid Sequence Analy- sis-Apo-A-I isoforms were obtained by preparative flat bed electro- focusing, conducted in an LKB 2117 Multiphor at 5 "C as per Marcel et al. (24) with the following changes. To increase protein recovery, Bio-Lyte electrofocusing gel (Bio-Rad) was used as stabilizing me- dium (25). The gel was washed according to the manufacturer's suggestions with distilled H20 and then equilibrated with 6-8 M urea.

The gel slurry was adjusted to yield final concentrations of 6.5 M urea and 2.5% pH 4-6 ampholytes. Desalted HDL (-20 mg) was delipi- dated and dried under N2, dissolved in 8 M urea containing 2% Triton X-100, and mixed with the gel slurry. The slurry was dried under a fan until a stable gel was obtained (-120 g final wt.). Only a single focusing step was required. Using a paper print (24) as guide, gel sections were removed with a spatula and eluted by shaking for 1 h with 2 volumes of 5 M urea. Gel particles were removed by centrif- ugation and washed with an additional volume of urea. Pooled super- natant solutions were filtered through 0.45-pm membranes and then electrodialyzed at 1200 V according to Bloomster and Watson (26) with 10 mM Hepes, pH 7.5, in dialysis bags (M, = 12,000 cutoff). Samples were then dialyzed extensively against 0.1 M NH4HC03, pH 7.0, and lyophilized.

Automated solid phase Edman sequencing was conducted by Se- quemat, Inc., Watertown, MA. The proteins were immobilized on p- phenylene diisothiocyanate-activated aminopropyl-glass (27). The phenylthiohydantoin derivatives of the released amino acids were identified by reversed phase high performance liquid chromatography.

Labeling Liver and Intestinal Tissue in Vivo-A 176-g rat was anesthetized by an intraperitoneal injection of sodium pentobarbital (50 mg/100 g of body weight). Then 4.5 mCi of [3H]leucine in 0.9 ml of 0.15 M NaCl was injected into a saphenous vein. After 18 min, blood was withdrawn from the left ventricle of the heart, adjusted to 3 mM EDTA, 0.1 % NaN3, and held at 0 "C. Through the same site, the entire body was flushed with -500 ml of ice-cold 0.15 M NaCI, which was allowed to run out by severing the thoracic vena cava. A 0.5-g portion of blood-free liver was homogenized in 13 rnl of phos- phate-buffered saline, pH 7.4, containing 1 mM EGTA, 0.1% NaN3, 50 pg of aprotinin and 30 pg of phenylmethylsulfonyl fluoride. A second 0.5-g portion was similarly homogenized, and to this portion Triton X-100 and deoxycholate were added to a final concentration of 1% each and homogenization was repeated. Similarly, two 0.5-g portions of jejunum, taken just distal to the ligament of Treitz, were homogenized after the tissue was dissected free of mesentery and the lumen rinsed. The tissues were kept at 0 "C throughout. All four homogenates were then sonicated for three 12-s bursts. The sonicates were centrifuged at 21,000 X g for 30 min at 2 "C. The supernatant tissue extracts, including all floating lipid, were decanted and mixed to resuspend the lipid.

The detergent-containing extracts were used for immunoprecipi- tation of apo-A-I (see below). To the extracts without detergent was added %o volume of freshly prepared rat plasma containing 3 mM EDTA, 0.1% NaN3, to provide carrier lipoproteins, and then the total lipoproteins, d < 1.21 g/ml, were isolated by ultracentrifugation.

Labeling Liver and Intestinal Tissue in Vitro-Tissues and solu- tions were maintained at 0-4 "C throughout the procedure unless otherwise noted. The liver and a jejunal segment were obtained from a 275-g rat anesthetized with ether. The intestinal lumen was flushed with 0.15 M NaCI. Tissue pieces (-30 mg in six pieces) were incubated in an atmosphere of 95% 02, 5% COZ at 37 "C with 0.1 ml of Krebs- Ringer bicarbonate buffer containing 100 pCi of [3H]leucine. After 75 min, the incubation mixtures were chilled and the media removed. The tissues were sonicated in 0.3 ml of phosphate-buffered saline, pH 7.4, containing 1% (v/v) Triton X-100, 1% (w/v) sodium deoxy- cholate, with EGTA, NaN3, and inhibitors as before. Supernatant solutions were obtained following centrifugation at 12,000 x g for 2 min.

Immunopreclpitation-Rabbit antiserum specific for rat apo-A-I was a gift from Peter Herbert, Miriam Hospital, Providence, RI. For immunoprecipitation of tissue apo-A-I, 0.27 ml of antiserum was added to 0.2-0.5 ml of tissue extract containing 40 pg of carrier lipoprotein, d < 1.21 g/ml, and the mixture was diluted to a final volume of 1.0 ml with 0.1 M phosphate buffer, pH 7.1. Immunopre- cipitates were collected and washed according to Zannis et al. (15). For electrophoretic analysis, immunoprecipitates were dissolved in 9.6 M urea containing 2% ampholytes, pH 4-6,2% Triton X-100, and 5% 2-mercaptoethanol. In control experiments, the isoform pattern obtained by immunoprecipitation of apo-A-I from a 3H-labeled liver perfusate sample (see below) was compared with that obtained using total lipoproteins isolated from the same sample by ultracentrifuga- tion. Identical patterns of 3H distribution were obtained.

Liver Perfusion-Livers were perfused in situ using minor modifi- cations (28) of the method of Mortimore (29). Liver donor rats weighed 270-290 g and were prepared under a variety of dietary conditions, including 24-h fasted, meal-fed, or ad libitum feeding of fat-free, semisynthetic diet or stock diet. Because the dietary condi-

by guest on Novem

ber 8, 2018http://w

ww

.jbc.org/D

ownloaded from

Processing of Rat Proapolipoprotein A-I to Mature Form 6461

tions did not affect the experimental data, results were combined. All perfusates (see Table 11) contained antibiotics and were maintained between pH 7.35 and 7.40 (12). After 5 min of recycling, 1.0 ml of 0.15 M NaCl containing 1-2 mCi of 13H]leucine was injected into the portal cannula. Samples of perfusate (5-10 ml) were removed a t various intervals, chilled in ice, and centrifuged to remove red cells. The supernatant fluid was adjusted to 3 mM EDTA, 0.1% NaNa. Isolation of total lipoproteins ( d < 1.21 g/ml) was begun within 18 h. After 3 h of perfusion, a mean of 30.6% of the radioactive dose had been incorporated into perfusate proteins, 2.8% into perfusate apoli- poproteins, and 0.4% into perfusate apo-A-I (determined by analysis of the perfusate apolipoproteins on 10% polyacrylamide gels (30)).

RESULTS

Apo-A-I Isoforms in Rat Plasma-A representative pattern obtained by two-dimensional electrophoretic analysis of rat plasma apolipoproteins is shown in Fig. 1. In particular, repeated analyses consistently revealed the presence of three isoforms of apo-A-I. The plasma apo-A-I isoforms have the following apparent PI values and quantitative distribution, as determined by pyridine elution of Coomassie blue-stained spots: isoform 3, PI = 5.68, 69 & 3%; isoform 4, PI = 5.55, 29 f 3%; isoform 5, PI = 5.44, 2 f 1% (mean f S.D., n = 27 determinations on 11 rats). Thus, isoform 3, the more basic form, is the major circulating apo-A-I species.

Newly Synthesized Intracellular Apo-A-I Isoforms-The pattern of apo-A-I isoform biosynthesis in liver and small intestine was determined by exposing them to ['Hlleucine both in uiuo and in uitro. When the ['Hlleucine was admin- istered intravenously in uiuo, samples of blood-free tissue were harvested to ensure isolation of only intracellular apo-A-I. Less than 10% of the 'H incorporated into liver or intestinal apo-A-I was in isoform 3, while 70-90s appeared in isoform 4 (Table I). Apo-A-I isolation from the tissue extracts by ultracentrifugal flotation and by immunoprecipitation gave similar results.

Examination of 'H incorporation into tissue apo-A-I by liver and small intestine incubated in uitro (Table I) gave results similar to those obtained in uiuo. In every case, isoform 4 was the predominant form of newly synthesized apo-A-I in the tissues (65-90% of total).

Apo-A-I Isoforms Released into the Circulation-To identify the form of apo-A-I released by liver and intestine into the circulation, perfusion studies were performed with the isolated

pH 6 pH 4 I (.Vt

IEF

S D S T - A-IV

JE

- A-I

- C-lIl3 -C-llIo - A-ll - c-ll

FIG. 1. Two-dimensional gel electrophoresis of rat plasma apolipoproteins. Desalted, delipidated lipoproteins, d < 1.21 g/ml ( -30 p g ) , were analyzed and stained for protein with Coomassie blue as indicated under "Experimental Procedures." For comparison, the corresponding isoelectric focusing (IEF) pattern is shown above the two-dimensional gel. The isoform numbering system and apolipopro- tein identification were adopted from Ghiselli et al. (18). A-IV, apolipoprotein A-IV; E, apolipoprotein E etc.

4''

4 0 3 4 5

TABLE I Distribution of labeled apo-A-I isoforms in tissues after labeling with

['Hlleucine in vivo or in vitro ~~~~~ ~.

Labeling Tissue APO-A-I ApoA-I isoform

method isolation method 3 4 5 Othef

~~~

%total apo-A-I radioactiu- ity

In vivob Liver Immunoprecip- 6.2 91.4 2.4 0.0

Ultracentrifu- 3.8 71.6 11.8 12.8

Intestine Immunoprecip- 9.0 75.4 3.6 12.0

Ultracentrifu- 4.3 79.8 5.3 10.6

In vitro' Liver Immunoprecip- 11.1 87.0 1.9 0.0

Intestine Immunoprecip- 12.6 65.1 2.9 19.4

(18 min) itation'

gationd

itation

gation

(75 min) itation

itation ~ - ~~~ ~~~~

"Includes one or more gel areas in the apo-A-I region which are more basic than isoform 3. With the samples prepared by ultracen- trifugation, this material appeared at the isoelectric focusing origin (pH > 6), while in the immunoprecipitates, this radioactivity was spread over a wide area.

"Rats were injected intravenously with ['Hjleucine and exsanguin- ated after 18 min, and their entire vascular system was flushed with 500 ml of chilled 0.15 M NaCI. Portions of blood-free liver and intestine were removed and treated further as indicated.

'Tissues were homogenized and sonicated in phosphate-buffered saline containing preservatives, 1% deoxycholate, and 1% Triton X- 100. Apo-A-I isoforms were immunoprecipitated from these detergent extracts with rabbit anti-rat apo-A-I. Precipitates were analyzed on two-dimensional gels and the stained isoform areas were excised, digested and the 3H quantitated. (See "Experimental Procedures" for additional details.)

dAdditional tissue portions were sonicated in the absence of deter- gents. Total lipoproteins, d < 1.21 g/ml, were isolated from these sonicates by ultracentrifugation. Lipoproteins were analyzed on two- dimensional gels as above.

'Pieces of rat intestine or liver were incubated with ['Hlleucine at 37 "C for 75 min in an atmosphere of 95% oxygen and 5% C02. Apo- A-I isoforms were immunoprecipitated and analyzed as above. (See "Experimental Procedures" for additional details.)

organs. Isolated rat livers were perfused with a freshly pre- pared mixture of defibrinated blood and serum plus ['HI leucine. After 3 h of recirculation through the liver, approxi- mately 80% of the radioactivity in perfusate apo-A-I was found in isoform 4 (Table II) , similar to results previously reported (18). When the serum in the perfusate was replaced with a defined physiological buffer, isoform 4 again accounted for over 80% of the labeled apo-A-I released by the liver. This was true not only when the perfusate was sampled a t 3 h after addition of the ['H]leucine dose (Table 11) but also in perfus- ate sampled earlier (0.8, 1.3, and 2 h) or later (4 and 5 h). Since the release of labeled apo-A-I into the perfusate is quite rapid (at least 75% complete by 1.3 h),2 these results indicate that recycling of newly synthesized apo-A-I through the livers in the presence or absence of serum did not significantly alter the isoform pattern.

Most apo-A-I synthesized by the small intestine reaches the circulation via mesenteric lymph while some is released directly into intestinal venous blood (32). To identify the apo- A-I isoform products released by rat small intestine, ["HI leucine was administered into the lumen of a segment of jejunum from which all lymph and venous blood were collected for 3 h in uiuo as previously described (12, 30). Thus, all labeled proteins recovered in these fluids were exclusively of

~ ~~~ . . ~

H. G. Windmueller, and A. E. Spaeth, unpublished observation. ~ -~

by guest on Novem

ber 8, 2018http://w

ww

.jbc.org/D

ownloaded from

6462 Processing of Rat Proapolipoprotein A-I to Mature Form

TABLE I1 Release of labeled apo-A-I isoforms by isolated, perfused organs

jolbwing a ( 3 H / ~ ~ i n e dose [3H]Leucine was injected into the portal vein cannula of isolated,

perfused livers and into the lumen of autoperfused jejunal segments in uiuo. Lipoproteins were isolated by ultracentrifugation from liver perfusate and from intestinal lymph and venous blood collected at 3 h. Apo-A-I isoforms were analyzed on two-dimensional gels and radioactivity was quantitated as before. (See “Experimental Proce- dures” for additional details.)

Apo-A-I isoform’

3 4 5 % total upo-A-I radioactivity

Isolated, perfused liverb Recirculated perfusate

Defibrinated blood’ ( n = 3) 13.3 (3.7) 80.4 (6.0) 6.3 (2.5) Red cells in buffef‘ (n = 4) 7.1 (2.3) 86.8 (1.8) 6.4 (2.1)

Autoperfused small intestine‘ ( n = 2)

Fluid analyzed Mesenteric lymph 6.5 (0.2) 87.7 (0.6) 5.8 (0.5) Intestinal venous blood 6.6 (4.0) 88.6 (4.0) 4.7 (1.1)

“Values are means (S.D.) of n perfusion experiments. ‘The indicated perfusate containing 13H]leucine was recirculated

for 3 h as described under “Experimental Procedures.” ‘Perfusate consisted of 50% defibrinated rat blood and 50% rat

serum, both freshly prepared from rats anesthetized with pentobar- bital.

dPerfusate consisted of 22.5% washed, bovine red blood cells and 77.5% Krebs-Ringer bicarbonate buffer, pH 7.4, containing 10 mM glucose, 1% bovine serum albumin, and 20 L-amino acids a t 5 X plasma concentration except L-leucine at 1x plasma concentration

‘The intestine of an intact rat, prepared as previously described, was infused with lipids, starting 2 h before perfusion (12, 30). [3H] Leucine was then introduced into the lumen of a jejunal segment from which all lymph and venous blood were collected for 3 h in vivo exactly as described (12, 30). The 3-h lymph was visibly lipemic. All labeled proteins obtained by this method are exclusively of intestinal origin. Lymph and plasma were adjusted to 3 mM EDTA and 0.1% NaN3 before isolation of lipoproteins.

(31).

intestinal origin. As shown in Table 11, nearly 90% of the labeled apo-A-I recovered in both lymph and blood was iso- form 4. To verify these findings in intact animals, rats were exsanguinated 18 min after receiving an intravenous injection of [3H]leucine. The distribution of label among the apo-A-I isoforms in the plasma at 18 min was as follows (mean -+ S.D., n = 3 ) : isoform 3, 6.5 f 3.9%; isoform 4, 87.5 +- 6.3%; isoform 5,6.0 2 2.6%. Therefore, isoform 4 is the predominant form of apo-A-I synthesized and released into the circulation by both rat liver and small intestine, while the major steady state form observed in rat plasma is isoform 3.

NH2-terminal Amino Acid Sequence of Isoforms--In an attempt to establish the relationship (if any) of the apo-A-I isoforms to the processing intermediates identified by Gordon et al. (21), we initiated NH,-terminal amino acid sequence analysis of the major rat plasma isoforms. The following results were obtained: isoform 3, (Asp)-Glu-Pro-Gln-Ser-Gln- Trp-Asp-Arg-Val; isoform 4, X-Glu-Phe-X-Gln-Gln-Asp- Glu-Pro-GIn-Ser. Isoform 4 has six additional amino acids a t the NH2 terminus compared with isoform 3. Furthermore, the amino acid sequence of isoform 4 is identical to that obtained by Gordon et al. (21) for proapo-A-I, while the NH2 terminus of isoform 3 corresponds to that of the “mature” form? The

”

For the remainder of the report, apo-A-I refers to all forms of the apoprotein, proapo-A-I refers only to isoform 4, and mature apo-A-I refers to isoform 3.

solid phase method (27) used here does not permit identifi- cation of the initial NH2-terminal residue of either protein, although the NHz-terminal residue of isoform 3 can be iden- tified as aspartic acid by inference from the internal aspartic acid at the corresponding position of isoform 4. Isoform 5 could not be obtained in enough quantity for sequencing; its relationship to the other apo-A-I forms is uncertain (see “Discussion”).

Conversion of Proapo-A-I to the Mature Form in Viuo-If apo-A-1 is secreted as a proform, then it must be converted to mature apo-A-I, the predominant species in the circulation. To further document this conversion, pulse-chase experi- ments were conducted in intact rats. Rats were given an initial [3H]leucine pulse via a jugular vein cannula, followed by a continuous infusion of unlabeled leucine. Analysis of serial blood samples (Fig. 2 A ) clearly shows the in uiuo conversion of the apo-A-I proform (isoform 4) to mature apo-A-I (isoform 3 ) in the circulation. Assuming similar plasma clearance rates for proapo-A-I and the mature form, conversion reaches about 80% in 10 h, which is approximately the circulating half-life of apo-A-I (34). These data, then, explain the steady state proportions of apo-A-I forms observed in rat plasma. To rule out effects of the large leucine chase, additional similar ex- periments were conducted in which the chase was omitted.

5o A (chase) B (no chase1 T

l h 1 10

0 0

d I J I I I I I I I 2 4 6 8 1 0 0 2 4 6 8 1 0

TIME AFTER 13HILEUCINE DOSE Ihl

FIG. 2. Radioactivity of apo-A-I forms in plasma after a [‘Hlleucine dose. A, rats were prepared with a jugular vein cannula, as previously described (12), and allowed 20 h to recover in restraining cages during which time they were infused continuously with a 10% glucose solution at 2.2 ml/h and allowed free access to stock diet and water. Then, at t = 0, the glucose infusion was momentarily inter- rupted and 2.4 mCi of [3H]Ieucine in 0.8 mi of 0.15 M NaCI was infused into the jugular cannula. After 15 min, the glucose infusion was stopped and a 2-ml priming dose of unlabeled leucine (183 mM L-leucine, 62.5 mM NaCI) was given via the cannula. Thereafter, a continuous infusion with a solution of 183 mM L-leucine, 125 mM glucose at 4.0 ml/h was administered throughout the remainder of the experiment. At various times, 1 ml of blood was withdrawn via the jugular cannula and immediately adjusted to 3 mM EDTA and 0.1% NaN3 at 0 “C. Total lipoproteins, d < 1.21 g/ml, were isolated by ultracentrifugation from the corresponding plasma. To determine the total plasma radioactivity of each apo-A-I form, total apolipopro- tein radioactivity was first measured in the trichloroacetic acid- insoluble pellet prepared from an aliquot of the d < 1.21 g/ml lipoprotein fraction (33); the proportion of apolipoprotein radioactiv- ity due to apo-A-I was then determined by analyzing a second aliquot on 10% acrylamide gels containing 0.1% SDS (30); and, finally, the distribution of apo-A-I radioactivity among the apo-A-I forms was determined from analyzing a third aliquot on two-dimensional gels. Plasma volume (in milliliters) was assumed to be body weight (in grams) times 0.042. B, same as A except the L-leucine chase was omitted. The glucose infusion was continued throughout the experi- ment.

by guest on Novem

ber 8, 2018http://w

ww

.jbc.org/D

ownloaded from

Processing of Rat Proapolipoprotein A-I to Mature Form 6463

Again, conversion of the proform to mature apo-A-I can be readily seen (Fig. 2B). The increase in labeled proapo-A-I in plasma between 0.8 and 2.5 h in the nonchase experiments can be explained by ( a ) the earlier time of the first sample, before all newly synthesized intestinal proapo-A-I had time to traverse the lymphatic system into the blood, and (b ) the continuing production of labeled proapo-A-I from recycled [3H]leucine.

Conversion of Proapo-A-I to the Mature Form in Vitro-To establish whether the conversion of proapo-A-I to mature apo-A-I occurs in freshly drawn blood, incubations in vitro were conducted using apo-A-I labeled with [3H]leucine in vivo. As shown in Fig. 3, no conversion of pmform to mature apo- A-I was seen in the presence of EDTA. The conversion was also prevented by EGTA even in the Fresence of added M$+, suggesting a requirement for some other divalent metal. The conversion observed in vitro when h.eparin was used as anti- coagulant was much slower than the conversion in vivo and appeared to stop after about 4 h. An additional experiment (data not shown), in which cells and platelets were removed from the heparinized blood by centrifugation prior to incu- bation, produced a curve similar to that obtained with hepa- rinized whole blood. Thus, the formed elements of blood were not required for the limited conversion observed. The slow conversion observed in whole blood incubated in vitro may explain the slightly increased conversion seen in liver perfu- sions conducted with defibrinated blood in comparison to those using serum-free perfusate (Table 11).

Factors Affecting Distribution of Apo-A-I Forms-Apo-A-I

100 proapoA-I EDTA

mature apoA-I

0 2 4 6 8 1 0 0 2 4 6 8 1 0 J

INCUBATION TIME (h)

FIG. 3. Conversion of proapo-A-I to mature apo-A-I in blood incubated in vitro versus in vivo. Rats, under ether anesthesia, were injected via a saphenous vein with 2.2 mCi of [3H]leucine in 0.9 ml of 0.15 M NaCI. After 48 min ( t = O ) , again under anesthesia, the rats were exsanguinated. The drawn blood was immediately treated with the indicated anticoagulant. The blood (-1.5 X lo6 dpm in total apo-A-I) was incubated aseptically a t 37 "C in vitro in an atmosphere of 95% oxygen and 5% COO. The sample pH remained within the range of pH 7.2-7.5 throughout the incubations. A t various times, aliquots were removed and adjusted to 3 mM EDTA, 0.1% NaN3 at 0 "C. The blood samples were centrifuged and total lipoproteins, d < 1.21 g/ml, were isolated by ultracentrifugation from the corresponding plasma. Radioactivity in the apo-A-I forms was determined on two- dimensional gels as usual. Anticoagulants: M, 25 mM EDTA M, 0.7 mg heparin/ml; W, 25 mM EDTA plus 0.7 mg heparin/ml; c"., 25 mM EGTA plus 25 mM MgC12. The data in viuo (A- - -A) were calculated from the experiment in Fig. 2A. All data were plotted after normalization to a mean value at t = 0; actual values for proapo-A-I a t t = 0 ranged from 84.9 to 89.8% with a mean of 87.5%. Data for isoform 5 showed no discernible trends; the proportion (-5%) remained relatively constant under all conditions examined and therefore is not shown.

(A) ( B )

DENSITY GRADIENT HEPARIN-SEPHAROSE 3H6 7 - - - - - w

" I , - - "

-t-

A - = = -

II

pH4 ._ -

I- - - mature A - I " w - proA- I **

1 ;> 3 'I 5 ApoE ApoE Whole

HDL Fraction poor rlch HDL HDL HDL

FIG. 4. Isoelectric focusing pattern of plasma HDL subfrac- tions. A, sequential HDL subfractions were prepared from normal rat plasma by the density gradient method of Redgrave et al. (35). The density of the fractions was determined by pycnometry. The density range of each fraction is shown in Table 111. Isoelectric focusing was conducted on 5 X 95-mm tube gels prepared as described under "Experimental Procedures" for first dimension gels. Gels were stained with an aqueous solution containing 0.04% Coomassie blue R-250,0.05% crocein scarlet, 0.5% CuSO,, 27% isopropanol, and 10% acetic acid. B, heparin-Sepharose affinity chromatography of rat plasma HDL (d = 1.040-1.21 g/ml) was conducted as described (36) except that MnCll was omitted and 0.1% NaN3 was added to all buffers. Apo-E-poor HDL passed through the column unretarded while apo-E-rich HDL required 100 mM NaCl for elution. Isoelectric focusing was conducted as in A. *, unidentified band; **, apo-A-I isoform 5.

TABLE I11 Distribution of apo-A-I forms among plasma HDL subfractions

Subfractions of HDL were prepared and focused as described in Fig. 4. Stained bands were quantitated by scanning of gels at 500- 550 nm in a Gilford Model 2520 gel scanner.

Distribution of Fractionation apo-A-I forms"

method Fraction Density Proaw- Mature

glml Whole HDL 1.040-1.210 Density gradient 1 1.040-1.059

2 1.059-1.077 3 1.077-1.100 4 1.100-1.130 5 1.130-1.180

Heparin-Sepharose Apo-E-poor HDL Apo-E-rich HDL

~

% to&

35.9 64.1 29.2 70.8 32.7 67.3 32.0 68.0 36.6 63.4 33.9 66.1 35.4 64.6 29.7 70.3

"Isoform 5 was present in concentrations too low to quantitate by this method. Therefore it has been omitted from the analysis.

is the major apolipoprotein in HDL. The distribution of apo- A-I forms in various subfractions of HDL was examined in a search for functional differences between the forms. Five subfractions of normal rat plasma HDL were prepared by density gradient ultracentrifugation and apo-A-I forms were quantitated by scanning of isoelectric focusing gels (Fig. 4A). Isoform 5 was present in such low concentrations that it could not be quantitated by this method and was therefore omitted from the analysis. The proportion of proapo-A-I was similar throughout the density fractions (Table 111). The proportions of the proform and mature apo-A-I were similar in apo-E-rich and apo-E-poor HDL, separated by heparin-Sepharose affin- ity chromatography (Fig. 4B and Table 111). The plasma apo-

by guest on Novem

ber 8, 2018http://w

ww

.jbc.org/D

ownloaded from

6464 Processing of Rat Proapolipoprotein A-I to Mature Form TABLE IV

Effect of sex and diet on distribution of rat apo-A-I f o r m Rats, 200-250 g, were fed as indicated for 15 days and then, without

fasting, they were exsanguinated under ether anesthesia. Total plasma lipoproteins, d < 1.21/g/ml, were isolated and apo-A-I forms analyzed as in Table III. Values are means (S.D.) for n rats. "" "

Distribution of apo-A-I forms

Proapo- Mature Sex n Diet

A-I A p - A - I

% total Female 10 Stock" 34.3 (2.9) 65.7 Male 7 Stock 30.7 (2.4) 69.3 Male 4 High fat, high cholesterolb 44.5 (3.5) 55.5 Male 4 Fat-free + orotic acid' 50.6 (2.9) 49.4 "NIH-07 open formula stock diet. bFinely ground stock diet with the following additions (final con-

centration, w/w): 5% lard, 1% cholesterol, and 0.3% taurocholic acid. 'AIN-76 purified diet (37) modified by replacement of the 5% corn

oil with an equivalent amount of glucose monohydrate and addition of 1% (w/w) orotic acid. These rats had grossly fatty livers when killed.

A-I pattern was also examined in female and male rats main- tained on different diets (Table IV). Males and females had similar steady state proportions of plasma proapo-A-I al- though the concentration of total apo-A-I was higher in the females (data not shown). Rats fed a high cholesterol, high fat diet, however, developed an increased proportion of cir- culating proapo-A-I. This increase in proform was even more pronounced for rats fed a fat-free diet containing orotic acid.

DISCUSSION

In this report, we have shown by NHz-terminal amino acid analysis of the two major apo-A-I isoforms from rat plasma that the more acidic isoform is identical to the proform obtained by Gordon et al. (21) by cell-free translation of rat intestinal apo-A-I mRNA. Similarly, the more basic isoform was identified as mature apo-A-I. We have also shown that both rat liver and small intestine synthesize and secrete i n vivo predominantly proapo-A-I, which is slowly processed in the circulation to mature apo-A-I. Our results further indicate that slow and incomplete processing occurs in heparinized blood or plasma incubated in vitro at 37 "C, may involve a labile factor, and exhibits an apparent requirement for diva- lent metal. Additionally, circulating proapo-A-I is shown to be virtually all associated with lipoproteins.

Since the completion of our work, the release of proapo-A- I by cultured human (38, 39) and rat (40) hepatocytes has been reported. Our observation that liver and intestine release the same form of apo-A-I is consistent with recent reports that the human (41) and the mouse (13) each express only a single apo-A-I gene. Apo-A-I isoform 5 rarely accounts for more than -6% of the newly synthesized apo-A-I secreted by liver and intestine and represents only 2% of the plasma steady state forms. The relationship of this minor acidic isoform to the pro- and mature forms of apo-A-I is uncertain.

In human plasma, the processing of proapo-A-I to mature forms appears to occur quite rapidly (after 10 min, only -50% of the injected, radiolabeled proapo-A-I remained in the pro- form) (42). Our results, however, indicate that processing in the rat occurs much more slowly, several hours being required to reach 50% completion in the circulation in uiuo. This is in contrast to the report by Stoffel et al. (40) that proapo-A-I released by cultured rat hepatocytes is immediately cleaved to mature apo-A-I in the presence of serum. However, they provided no data to support this assertion.

We found slow and incomplete processing of proapo-A-I in whole, heparinized blood or in plasma incubated for 10 h in vitro. This indicates that while the protease presumably needed for the conversion is present in plasma, it may be labile and, in uiuo, may be replenished continuously from a tissue source. Additional processing in vivo may also occur when circulating proapo-A-I comes in contact with certain tissues. Our failure to see progressive processing when proapo- A-I was recirculated through isolated livers for 3-5 h makes it unlikely that the liver is a significant site for further processing of circulating proapo-A-I. The role of the liver as a source of the circulating protease is difficult to assess from these experiments because the relatively large perfusate vol- ume (at least 3 times the rat's blood volume) and the apparent lability of the putative protease could have prevented the attainment of effective enzyme concentrations in the perfus- ate. Liver and intestinal tissue sampled as early as 18 min after exposure to [3H]leucine always contain small amounts (-5% of total) of a newly synthesized apo-A-I species with charge properties similar to mature apo-A-I. This suggests that some processing may occur even before release of apo-A- I from the tissues. Such heterogeneous processing is not without precedent (43).

Similar proportions of rat plasma proapo-A-I (-29%) were observed whether isolation involved ultracentrifugal floating, which depends on lipid-protein association, or specific im- munoprecipitation, which does not. These results indicate that the proform is not lost disproportionately from HDL during ultracentrifugal isolation. Since there is almost no loss of total apo-A-I when HDL is so isolated (44), we conclude that rat plasma proapo-A-I, like the mature form, is virtually all lipoprotein-associated. Additionally, the distribution of apo-A-I forms remains relatively constant among subfractions of rat plasma HDL prepared by affinity chromatography or density gradient ultracentrifugation. These fractions have been previously shown to vary in lipid as well as apoprotein composition (36, 45).

In the present studies, the relative concentration of proapo- A-I in rat plasma increased as a result of a high fat, high cholesterol diet. Similar observations have been made for human lymph (46). In that situation, however, the increased proportion of proapo-A-I in lymph chylomicrons and very low density lipoproteins could, in part, be a reflection of the greater retention of newly synthesized intestinal apo-A-I (i.e. proform) in the chylomicron-rich lymph, with diminished release directly into the blood (32). In our studies, the relative concentration of plasma proapo-A-I was even more markedly increased in rats maintained on a fat-free diet supplemented with orotic acid. By inhibiting hepatic very low density lipo- protein production totally and HDL production partially, this diet lowers plasma triglyceride and cholesterol concentrations to 10 and 25% of normal, respectively (47), and was chosen to provide the lower extreme in plasma lipid levels. HOW orotic acid or a high fat diet effect an increase in the steady state proportion of plasma proapo-A-I is unclear, but the plasma concentration of neutral lipids per se does not appear to be a major regulatory factor.

Thus, the physiological relevance of release of a stable apo- A-1 proform from tissues and further processing in the circu- lation to mature apo-A-I remains uncertain. Apo-A-I is known to activate 1ecithin:cholesterol acyltransferase, the major cho- lesterol-esterifying enzyme of plasma (6). However, other proteins and peptides with a large amphiphilic surface will activate this enzyme (48) and apo-A-I possesses several such regions (49). It seems unlikely, therefore, that loss of a hexa- peptide from the NH, terminus of proapo-A-I would influence

by guest on Novem

ber 8, 2018http://w

ww

.jbc.org/D

ownloaded from

Processing of Rat Proapolipc

the activation significantly unless the tertiary structure and surface features are markedly altered by the NHderminal modification.

Proapo-A-I processing has now been observed in both the human and the rat. Two important differences exist between the conversion processes in these two species. 1) In the human, processing to mature form involves conversion to a more acidic form due to the presence of a histidine in the propeptide (38, 50), whereas the opposite change occurs in the rat system due to substitution of glutamic acid for histi- dine in the rat propeptide. Artifactual changes, such as deam- idation and cyanylation, would mimic the human charge conversion but not the conversion to a more basic form seen with rat apo-A-I. 2) The steady state proportion of proapo-A- I in human plasma has been shown to be almost negligible (<2%) (50,51). In the rat, however, the proform accounts for -29% of the total plasma apo-A-I. This may be a reflection of a slower conversion of proapo-A-I to mature apo-A-I in rats as well as the widely different circulating half-life values reported for apo-A-I in these two species, -4.8 d in humans ( 5 2 ) but only -10 h in rat (34). Thus, the rat may have several advantages as a model system for further study of the site, mechanism, control, and physiological importance of proapo- A-I processing.

Acknowledgments-We thank Albert E. Spaeth for expert technical assistance throughout this work, Ai-Lien Wu for surgical assistance in several experiments, Giancarlo Ghiselli for guidance in setting up the two-dimensional gel system, Peter Herbert for his generous gift of apo-A-I antiserum, and Julie Worden for analysis of the plasma incubation in vitro.

REFERENCES 1. Sliwkowski, M. B., Wu, A.-L., and Windmueller, H. G. (1983)

Fed. Proc. 42, 1817 (abstr.) 2. Heiss, G., Johnson, N. J., Reiland, S., Davis, C. E., and Tyroler,

H. A. (1980) Circulation 6 2 , Suppl. IV, 116-136 3. Smith, L. C., Pownall, H. J., and Gotto, A. M., Jr. (1978) Annu.

Reu. Biochem. 47, 751-777 4. Pownall, H. J., Massey, J. B., Kusserow, S. K., and Gotto, A. M.,

Jr. (1979) Biochemistry 18,574-579 5. Tanford, C., and Reynolds, J. A. (1979) in The Chemistry and

Physiology of Human Plusma Proteins (Bing, D. H., ed) pp. 111-126, Pergamon Press, New York

6. Fielding, C. J., Shore, V. G., and Fielding, P. E. (1972) Biochem. Biophys. Res. Commun. 46, 1493-1498

7. Scanu, A. M., Byrne, R. E., and Mihovilovic, M. (1982) CRC Crit. Rev. Biochem. 13 , 109-140

8. Brewer, H. B., Jr., Fairwell, T., LaRue, A., Ronan, R., Houser, A., and Bronzert, T. J. (1978) Biochem. Biophys. Res. Commun.

9. Chung, H., Randolph, A., Reardon, I., and Heinrikson, R. L.

10. Swaney, J . B., Braithwaite, F., and Eder, H. A. (1977) Biochem-

11. Blue, M.-L., Ostapchuk, P., Gordon, J. S., and Williams, D. L.

12. Wu, A.-L., and Windmueller, H. G. (1979) J. Biol. Chem. 254,

13. Miller, J. C. E., Barth, R. K., Shaw, P. H., Elliott, R. W., and Hastie, N. D. (1983) Proc. Natl. Acad. Sci. U. S. A. 80, 1511- 1515

14. Cheung, P., and Chan, L. (1983) Nucleic Acids Res. 11, 3703- 3715

15. Zannis, V. I., Breslow, J. L., and Katz, A. J. (1980) J. Biol. Chem. 255,8612-8617

80,623-630

(1982) J. Biol. Chem. 257,2961-2967

istry 16,271-278

(1982) J. Biol. Chem. 2 5 7 , 11151-11159

7316-7322

?protein A-I to Mature Form 6465

16. Zannis, V. I., Kurnit, D. M., and Breslow, J. L. (1982) J. Biol.

17. Gidez. L. I.. Swaney, J. B., and Murnane, S. (1977) J. Lipid Res. Chem. 257,536-544

18,.59-68 18. Ghiselli. G.. Brewer. H. B.. Jr., and Windmueller, H. G. (1982)

Bwchkm. Biophys.’Res. Comkun. 107, 144-149

(1981) Biochemistry 2 0 , 2470-2475

120,519-522

G., and Strauss, A. W. (1982) J. Biol. Chem. 257,971-978

19. Lin-Su, M.-H., Lin-Lee, Y.-C., Bradley, W. A., and Chan, L.

20. Stoffel, W., Blobel, G., and Walter, P. (1981) Eur. J. Biochem.

21. Gordon, J. I., Smith, D. P., Andy, R., Alpers, D. H., Schonfeld,

22. Laemmli, U. K. (1970) Nature ( L o n d . ) 227 , 680-685 23. Fenner, C., Traut, R. R., Mason, D. T., and Wikman-Coffelt, J.

24. Marcel, Y. L., Bergseth, M., and Nestruck, A. C. (1979) Biochim.

25. Forgez, P., and Chapman, M. J. (1982) J. Biochem. Biophys.

26. Bloomster, T. G., and Watson, D. W. (1981) Anal. Biochem. 113 ,

27. Laursen, R. A., and Horn, M. J. (1977) in Advanced Methods in Protein Sequence Determination (Needleman, S. B., ed) pp. 21- 37, Springer-Verlag, Berlin

28. Windmueller, H. G., and Spaeth, A. E. (1966) J. Biol. Chem.

29. Mortimore, G . E. (1963) Am. J . Physiol. 204,699-704 30. Wu, A.-L., and Windmueller, H. G. (1978) J. Biol. Chem. 253,

31. Tolman, E. L., Schworer, C. M., and Jefferson, L. S. (1973) J.

32. Windmueller, H. G., and Wu, A.-L. (1981) J. Bid. Chem. 256 ,

33. Windmueller, H. G., and Spaeth, A. E. (1972) J. Lipid Res. 13,

34. Eisenberg, S., Windmueller, H. G., and Levy, R. I. (1973) J. Lipid

35. Redgrave, T. G., Roberts, D. C. K., and West, C. E. (1975) Anal.

36. Weisgraber, K. H., and Mahley, R. W. (1980) J. Lipid Res. 21,

37. Windmueller, H. G., and Spaeth, A. E. (1980) J . Biol. Chem.

38. Zannis, V. I., Karathanasis, S. K., Keutmann, H. T., Goldberger, G., and Breslow, J. L. (1983) Proc. Natl. Acad. Sci. U. S . A. 80,

39. Gordon, J. I., Sims, H. F., Lentz, S. R., Edelstein, C., Scanu, A. M., and Strauss, A. W. (1983) J . Biol. Chem. 258,4037-4044

40. Stoffel, W., Bode, C., and Knyrim, K. (1983) Hoppe-Seyler’s 2. Physiol. Chem. 364,439-445

41. Shoulders, C. C., and Baralle, F. E. (1982) Nucleic Acids Res. 10,

42. Ghiselli, G., Schaefer, E. J., Law, S., Light, J. A., and Brewer, H.

43. Hanley, J. M., Haugen, T. H., and Heath, E. C. (1983) J . Bid.

44. Van Lenten, B. J., and Roheim, P. S. (1982) J. Lipid Res. 23,

45. Oschry, Y., and Eisenberg, S. (1982) J. Lipid Res. 23,1099-1106 46. Ghiselli, G., Schaefer, E. J., Light, J. A., and Brewer, H. B., Jr.

47. Windmueller, H. G., and Levy, R. I. (1967) J. Biol. Chem. 242 ,

48. Yokoyama, S., Fukushima, D., Kupferberg, J. P., Kizdy, F. J., and Kaiser, E. T. (1980) J . Biol. Chem. 255,7333-7339

49. Fitch, W. M. (1977) Genetics 86,623-644 50. Brewer, H. B., Jr., Fairwell, T., Kay, L., Meng, M., Ronan, R.,

Law, S., and Light, J. A. (1983) Biochem. Biophys. Res. Com- mun. 113,626-632

51. Zannis, V. I., Lees, A. M., Lees, R. S., and Breslow, J. L. (1982) J. Bwl. Chem. 257,4978-4986

52. Malmendier, C. L., Delcroix, C., and Ameryckx, J. P. (1983) Clin. Chim. Acta 131.201-210

(1975) Anal. Biochem. 63,595-602

Bwphys. Acta 573, 175-183

Methods 6,283-296

79-84

241,2891-2899

2525-2528

Biol. Chem. 248,4552-4560

3012-3016

92-105

Res. 14,446-458

Biochem. 65, 42-49

316-325

2 5 5 , 107-112

2574-2578

4873-4882

B., Jr. (1983) Clin. Res. 31,500A

Chem. 258,7858-7869

1187-1195

(1983) J . Lipid Res. 24, 731-736

2246-2254

by guest on Novem

ber 8, 2018http://w

ww

.jbc.org/D

ownloaded from

M B Sliwkowski and H G Windmuellerprocessed to apolipoprotein A-I in the circulation.

Rat liver and small intestine produce proapolipoprotein A-I which is slowly

1984, 259:6459-6465.J. Biol. Chem. 

  http://www.jbc.org/content/259/10/6459Access the most updated version of this article at

 Alerts:

  When a correction for this article is posted• 

When this article is cited• 

to choose from all of JBC's e-mail alertsClick here

  http://www.jbc.org/content/259/10/6459.full.html#ref-list-1

This article cites 0 references, 0 of which can be accessed free at

by guest on Novem

ber 8, 2018http://w

ww

.jbc.org/D

ownloaded from