THE JOURNAL OF I~IOLOGICAL CHEMISTRY Vol. 241, No. 8, Issue of April 25, 1966

I’ririnki in U.S.A.

Biosynthesis of Lipoprotein by Rat Intestinal Mucosa*

(Received for publication, July 13, 1965)

I:IIEUEI~ICK T. HATCH,$ YOSHIRO Aso, LILLIAN M. HAGOPIAN, AND JOEL J. RUBENSTEIN

From the Arteriosclerosis Unit, Medical Xervices, Massachusetts General Hospital, and the Department of Medi- cine, Harvard Medical School, Boston, Massachusetts

SUMMARY

Lymph chylomicrons are at the lipid-rich end of the spec- trum of the lipoproteins. The nature of their protein moiety and its biosynthesis have not yet been fully elucidated. Observations are reported on the incorporation of radioactive leucine into a “lowest density lipoprotein” by mucosal cells of the small intestine isolated from the rat during absorption of fat. The amino acid incorporation exhibited the char- acteristic energy requirement and sensitivity to inhibitors of other systems for biosynthesis of proteins. The protein moiety was not homogeneous, but appeared to consist of a small number of species that could be separated by thin layer electrophoresis in starch granules. Similar, but non- identical, electrophoretic patterns were obtained with the residual soluble intracellular protein of the intestinal mucosa and with chylomicrons of intestinal lymph.

The lowest density lipoprotein was isolated from a homog- enate of the mucosal cells. After removal of lipids by ex- traction with organic solvents, the protein moiety exhibited several unusual properties in addition to its affinity for lipids: (a) a predominance of apolar amino acids, (b) resistance to fractionation procedures involving ionic properties, (c) con- siderable solubility in organic solvents, and (d) peculiarities of the peptide map that were related to the foregoing features. Peptide maps were obtained after proteolytic digestion of the protein moieties of the lowest density lipoprotein fraction, the residual soluble protein, and lymph chylomicrons by means of two-dimensional electrophoresis and chromatog- raphy. Unusual maps were obtained in which most of the peptide fragments failed to move upon electrophoresis over a broad pH range, but were readily resolved by chromatog- raphy in organic solvents. Strong, although imperfect, resemblance was noted among the peptide maps of all three protein fractions. Preliminary studies of low and high density lipoproteins of rat serum revealed different, more complex peptide maps.

No definitive conclusion could be drawn about the speci- ficity of the protein which becomes associated with lipid during absorption through the intestinal mucosa. A specu- lative interpretation of the aforementioned similarity in pep-

* Supported by a grant from the John A. Hartford Foundation, Inc. Preliminary reports of this work have appeared (1, 2).

$ Present address, Bio-Medical Research Division, Lawrence Radiation Laboratory, P.O. Box 808, Livermore, California 94551. This research was performed during the tenure of an Established Investigatorship of the American Heart Association.

tide maps led us to suggest that lipids in transit become associated with some of the more abundant soluble proteins of the mucosal cells and that the resulting lipoprotein com- plexes appear in the chylomicrons of intestinal lymph.

Thin layer methods of chromatography and electrophoresis were extensively employed. Their convenience and great sensitivity permitted detailed studies to be carried out on submilligram quantities of material.

Lymph chylomicrons appear to contain a small amount of pro- tein and thus may be included in the category of “lipoproteins” (3-6). The amount of protein present, its identity, and the locus of its biosynt.hesis continue to be unsettled points (7-10).

There is evidence, first reported by Bragdon in 1958, that the chylomicron protein is at least partly synthesized by the intestine (5). In 1959 Rodbell and Fredrickson (7) and Flodbell, Fred- rickson, and Ono (8) characterized three proteins which were as- sociated with lymph chylomicrons of the dog and human. One of these proteins was believed to be identical with the protein moiety of plasma high density lipoprotein. Indirect experiments have also suggested a relationship between high density lipopro- tein and chylomicrons (1 l-13). A second protein may have been the same as one found in the very low density (triglyceride- bearing) fraction of the plasma lipoproteins. These two proteins were believed to be synthesized by the intestine on the basis of data on incorporation of radioactive amino acids. Except for a single experiment in vitro with isolated mucosal cells of the dog, contribution of the proteins by the liver through its lymphatic drainage could not be excluded because the cannulated thoracic duct probably transported both hepatic and intestinal lymph. Biosynthesis of the major serum lipoproteins by the liver has been well established by experiments in viva and in vitro (14, 15).

More recently, clinical observations in patients with inborn errors of lipoprotein metabolism have raised provocative ques- tions concerning the role of lipoproteins in the intestinal absorp- tion of fat. Patients with Tangier disease, who appear to have a hereditary deficiency of (or- or high density lipoprotein, have no impairment of intestinal fat absorption (16). On the other hand, the deficiency of P-lipoprotein, which was not implicated in fat absorption by the biochemical experiments cited above, involves a severe impairment in the absorption of triglycerides of long chain fatty acids (17). The fat appears to enter the intestinal

1655

by guest on June 13, 2018http://w

ww

.jbc.org/D

ownloaded from

1656 Biosynthesis of Lipoprotein by Rat Intestinal Mucosa Vol. 241, No. 8

TABLE I Methods Incorporation of ‘T-leucine into lipoproteins by rat intestinal

mucosa cells Mucosal cells from oil-fed rats were incubated for 2) hours at

37” in a Dubnoff shaker under a gas phase of 95% 02-5yo COZ. The incubation medium contained 5 ml of Krebs-Ringer-bicar- bonate buffer (pH 7.5) (20), 5 mg of glucose, 3 mg of penicillin, 0.05 mg of streptomycin, and 10 pC of L-leucine-14C (specific ac- tivity, usually 240 mC per mmole). At the end of incubation, 1 mg of nonradioactive leucine was added. Control vessels were incubated, but label was added at the end, after the carrier leu- tine. Tissue and medium were homogenized and fractionated in the preparative ultracentrifuge (see the text).

Animal Preparation--Male Sprague-Dawley rats weighing 150 g were used for the isotope incorporation experiments in vitro. Larger animals were used for cannulation of the intestinal lym- phatic channels or for preparation of unlabeled protein in quan- tity. The rats were fasted for 18 hours and then were given pellets of chow which had been soaked in olive oil or corn oil. Ninety minutes a.fter the pellets had been eaten (during active fat ab- sorption), the rats either were operated upon under pentobarbital anesthesia or were killed by a blow on the head and cervical dislocation.

Collection of Zntestinal Chyle-The chylous main intestinal lymphatic channel was exposed and cannulated with PE-50 poly- ethylene tubing.1 After stable lymph flow was established, 35 PC of 14C-labeled Chlorella hydrolysate were administered into the stomach, followed by 1 ml of corn oil. At the end of the experiment blood was collected by cardiac puncture and the in- testinal mucosa was removed as described below.

Ultracentrifugal fraction

Lowest density lipoprotein.

Very low dell- sity lipopro- tein*...

Low density lipoprotein,&

High density lipoproteina.

Residual solu- uble protein.

-

II

Flotation conditions

lensity Time Force ~__ __- g/ml hr xg

1.005 0.5 26,000

1.00: 16 105,000

1.06: 20 105,000

1.21 40 105,000

1.21 Sedim- n ts

Experimental

‘rotein Specifi( activit:

~__ 7ng cpnth

0.516 771

1.63 88

1.03 20

1.01 21

3.61 85

;P -- %

Control

‘rot& Spqcilfc actndy

-- “g ~Wm/W

0.600 2.7

1.58 2.5

1.19 0

0.90 0

3.27 8

a In most subsequent experiments these three fractions were combined by proceeding directly to density 1.21 after the first fractionation step.

mucosa from the lumen l ut to remain largely within the cells, undergoing no detectatl: t)ransport via lymph or plasma chylo- microns (18). The foregoing summary of experimental and clini- cal observations suggests that the intestine synthesizes proteins which have a role in the absorption and circulatory transport of fat. However, they leave room for further exploration of the essential protein components of the chylomicron and the sources from which they arise.

We present here observations on the biosynthesis of lipoprotein carried out by isolated mucosal cells of the rat, and a comparison of this protein with that of int,estinal lymph chylomicrons. The protein appears to be concerned in the process of fat absorption. However, the nature of this protein differs significantly from that. of the lymph lipoproteins obtained in earlier experiment,s with whole animals of other species (7-9).

EXPERIMENTAL PROCEDURE

MO,t~ialS

r,-Leucine-‘4C, labeled uniformly or in the carboxyl group, and Chlorella protein hydrolysate, labeled uniformly with 14C, were purchased from New England Nuclear Corporation. Puromy- tin, actinomycin D, and chloramphenicol were gifts from Dr. M. Lamborg, Dr. A. Aisenberg, and Mr. J. T. Murphy, respectively. Crystalline bacterial proteinase (Nagarse) was obtained from the Enzyme Development Corporat.ion, New York. Trypsin (trichloracetic acid-treated) and pepsin were products of Worth- ington. Pronase, chymotr)psin, and ribonucleasc were obtained from Calbiochem.

Preparation of Intestinal Mucosal Cells-The small intestine was irrigated with cold Ringer’s or NaCl solution, and the duo- denum and jejunum were excised. The mucosal cells were ex- truded onto an ice-cold glass plate by scraping with a spatula along the serosal surface (19). This preparation appeared to consist largely of whole cells under phase contrast microscopy.

Incubation of Tissue and Isolation of Lipoproteins-Aliquots of the cell preparation were incubated as described in Table I. After 24 hours of incubation, very little labeled lipoprotein was present in the medium under any of the conditions tested. Ac- cordingly, the medium and tissue together were homogenized in a Potter-Elvehjem mortar with a motor-driven Teflon pestle for about 10 up-and-down cycles. The homogenate (4 ml) was added to 6-ml ultracentrifuge tubes. Water (2 ml) was layered on the top and the tubes were centrifuged in the Spinco prepara- tive instrument for 30 min at 26,000 X g without refrigeration.

The tubes were kept in ice except when in the centrifuge. With a syringe and hypodermic needle, 4 ml were transferred from the bottom of each tube (sedimented cell solids being avoided) to another tube for density adjustment. The upper 2 ml were mixed with lipid material adhering to the tube walls. Water (4 ml) was layered on top in the original tube, and cen- trifugation was repeated for 1 hour at 64,000 x g at a chamber temperature of 14” to remove a small amount of contaminating higher density protein from the lowest density fraction. The tubes were then cut with a Spinco tube slicer to yield a top fraction of approximately 2 ml containing all of the turbid lipoprotein of lowest density. The bottom fraction of this washing step was discarded.

To the bottom fraction (4 ml) of the original step were added 1.771 g of solid KBr and 1.45 ml of NaCl solution of density 1.005 (which was 1 mM in EDTA) to adjust the mixture to 6 ml with a solvent density of 1.21 by the method of Radding and Steinberg (14). The KBr was dissolved by mixing, and the tubes were centrifuged for 40 hours at 114,000 x g at 14”. The tubes were sliced to give top and bottom fractions of 2 ml and 4 ml, respec- tively.

Serum and lymph were fractionated by ultracentrifugation in an analogous manner.

Measurement of Incorporation of Radioactive Leucine-The top

1 We are indebted to Dr. M. Hori for skillful performance of the surgical procedure.

by guest on June 13, 2018http://w

ww

.jbc.org/D

ownloaded from

Issue of April 25, 1966 Hatch et al. 1657

fractions from the first and second centrifugation steps repre- sented lipoproteins of lowest density (corresponding roughly to lymph or serum chylomicrons) and other lipoproteins, respec- tively. The second bottom fraction (d > 1.21) represented residual soluble cellular proteins. The two top fractions and 2-ml aliquots of the second bottom fractions were treated with 10 ml of 10% trichloracetic acid solution (containing 0.1 mg of nonradioactive leucine carrier per ml) and were allowed to stand overnight at 4”.

Protein was determined on the fractions by the method of Lowry et al. (21). Turbidity and solid KBr, if present, were eliminated by centrifugation at 3000 rpm before the samples were read in the spectrophotometer. Readings were performed at 750 mp, or at 500 rnp for larger amounts of protein. Crystalline bovine serum albumin was used as the standard.

Aliquots of the suspension of precipitated proteins containing 1 mg or less of protein were filtered on Millipore filters (diameter 24 mm; pore size, 0.8 p) by the method of Loftfield and Eigner (22). The filters were washed with 20 ml of the above trichlor- acetic acid solution and then with 15 ml of chloroform-methanol (2:l) to remove lipids. The filters were dried in air and placed on the bottom of vials; then 5 ml of scintillation mixture (con- taining, per liter of toluene, 5 g of 2,5-diphenyloxazole and 0.3 g of p-bis-1,2’-(5’-phenyloxazolyl)benzene were added. The samples were counted in a Packard liquid scintillation counter. The results were expressed in terms of counts per min per mg of protein. The amount of protein was measured as described above because it proved impossible to obtain reproducible meas- urements directly upon the Millipore filters. As will be shown below, the amount of protein is overestimated in this way. A rather constant proportion of the material giving the color reac- tion apparently consists of peptides of relatively low molecular weight, which are soluble in trichloracetic acid and pass through the Millipore filter but are retarded by Sephadex G-25. The correction factor for the true protein concentration varies with experimental conditions from less than 0.5 to about 0.7, but would be fairly constant within a group of similar samples.

PuriJication of Lipoprotein and Residual Protein on Sephadex Columns-Preliminary experiments with thin layer chromatog- raphy of lowest density lipoprotein and residual protein on Sephadex G-25 (23) revealed evidence for the presence of low molecular weight peptide material in both fractions (24, 25). It was considered desirable to eliminate such peptides before treat- ment of the proteins with proteolytic enzymes. Each fraction was therefore passed through a column 0.9 cm in diameter which contained 7 g (dry weight) of Sephadex G-25 (coarse beads) with distilled water as eluent. Fractions were collected manually and observations were made of turbidity, protein or peptide, radio- activity, and bromide ion (when appropriate) in the eluate.

Digestion uith Proteolytic Enzymes-The material of the first peak from the Sephadex column was highly turbid (resembling chylomicrons) in the case of the lowest density lipoprotein frac- tion. The first peak of the residual proteins was optically clear. Solutions of these materials, containing 0.5 to 5 mg of protein, were lyophilized on glass beads approximately 75 ,U in diameter (Prism0 Safety Corporation, Huntingdon, Pennsylvania) in a procedure analogous to that presented by Gustafson (26). Lipids were extracted from the protein-laden beads by shaking three times with chloroform-methanol (2: 1, v/v). The beads were then incubated with the protease Nagarse (27) (20 to 50 pg per mg of protein) in water solution for 18 hours at 37” with

gentle shaking. No turbidity suggesting bacterial contamination was noted.

The pept,ides released by the enzyme were extracted from the beads with water, and the solution was concentrated under a stream of Nz gas. The peptides were finally obtained in a very small voIume of water. Any insoluble matter was centrifuged, and the yield of peptides was estimated by the method of Lowry et al. (21).

Peptide Mapping and Localization of Radioactive Peptides- Thin layers (0.375 to 0.50 mm) of Whatman Chromedia cellulose (without binder, type CC41) were prepared on glass plates (20 x 20 cm) with a Desaga apparatus (Brinkmann Instru- ments). After brief drying in air, the plates were heated for 30 min at 105”. An estimated 25 to 100 pg of peptides were applied to the plates in l-p1 portions with drying in air until the total aliquot was added. The plates were then sprayed with the buffer for electrophoresis (guarding the sample application zone) and placed in a cell (E-C Apparatus Corporation, Philadelphia). Wicks of Whatman G F/B glass fiber paper were applied to the ends of the plate, and a clean glass plate or the cell cover was placed on top, separated by spacers from the thin layer plate. Coolant at 15” was circulated through the lower cooling plate. A regulated potential of 800 volts (16.5 volts per cm with a cur- rent of approximately 25 ma) was applied for 20 to 40 min. The usual electrophoretic buffer consisted of pyridine-acetic acid-water (1:10:189) at pH 3.5 (28). In some instances, pyridinium-acetate buffer at pH 6.5 and triethylamine-formate buffer at pH 9.5 were used. The volatile buffers were subse- quently removed by a stream of air.

After the electrophoretic run had been completed, ascending chromatography was performed in the second dimension in I-butanol-acet.ic acid-water (4: 1: 1). When the buffer had evap- orated, the plates were sprayed with 0.4% ninhydrin in 10% aqueous 2-propanol containing 5% (v/v) collidine, and the color was developed by warming the plates in a gentle flow of steam (29).

Peptides containing radioactive leucine were located by scrap- ing the cellulose in the ninhydrin-positive spots and appropriate blank areas with a razor blade. The samples were transferred to counting vials and overlaid with 5 ml of toluene scintillation mixture. The net counts per unit of plate area were recorded on a diagram made from the original plate. Slight quenching by ninhydrin-peptide colors was not significant.

Total Hydrolysis and Amino Acid Composition of Lipoprotein Protein-The material of the first, turbid peak from the Sephadex column treatment of the lowest density fraction was extracted with organic solvents by the method of Peterson (9) to remove lipids. The protein residue was transferred to a glass ampoule with water and 88% formic acid, and the solvents were removed on a rotary evaporator. For about 1.5 mg of protein, 5 ml of redistilled constant boiling HCl were added. The mixture was frozen and evacuated with an oil pump for 5 min; partial t,haw- ing was allowed. The ampoule was then sealed under vacuum and heated in a sand bath at 110” for 40 hours.

The ampoule was opened, and its contents were transferred to a tube in which three extractions with 5 ml of hexane were carried out for removal of residual lipid. The HCl was removed on the rotary evaporator and several additions of water were Iikewise removed. The residue was taken up in a small volume of water.

A sample of protein hydrolysate was dissolved in citrate buffer

by guest on June 13, 2018http://w

ww

.jbc.org/D

ownloaded from

1658 Biosynthesis of Lipoprotein by Rat Intestinal Mucosa Vol. 241, x0. 8

30-

CPM(x IO31

mg Protein 20-

TOTAL LYMPH 2

COLLECTED

( ml 1 1

HOURS -2 0 2 4 6 8 IO 12 14 16

FIG. 1. Incorporation of amino acids into the ultracentrifugal lipoprotein fractions of lymph in a rat with cannulated intestinal lymphatic channels. After continuous lymph flow was estab- lished during fat absorption, 35 &l of ‘Glabeled ChZoreZZa protein hydrolysate were administered into the stomach at the time indi-

cated by the arrow. Curve 1 is the time course of specific radio- activitv of lvmoh chvlomicron orotein; Curve 2, verg low density lipoprotein of lymph” (d < 1.005) ; Cur& 5, low ‘plus- high density lipoprotein of lymph (d 1.005 to 1.21); Curve 4, residual proteins sedimenting at d 1.21.

TABLE III

Incorporation of IT-leucine into lipoproteins by mucosal cells from fasted or fed rats

Rats were fasted for 18 hours and were killed either at once or 90 min after eating a pellet of chow soaked in olive oil. Incuba- t.ion conditions were as described in Table I.

TABLE II Incorporation of ‘V-labeled Chlorella hydrolysate amino acids

into lymph, serum, and intestinal mucosa fractions in vivo

Intestinal lymph (1.4 ml) was collected from an operated rat between 30 and 60 min after intragastric administration of 35 PC of uniformly labeled amino acid mixture. At the end of the col- lection period, serum and a homogenate of intestinal mucosal cells were obtained and fractionated as indicated below. Pro- tein content and radioactivity were determined as in “Methods.”

-_ Fasted rats Fed rats

Protein Protein

w

Specific activity

~pm/W w

Specific activity

cP+CS

0.318 12,150 0.270 8,560

1.36 2,003 1.58 1,710

5.18 8,020 4.53 6,970

Ultracentrifugal fraction

-

r

Flotation conditions

Density Tiie FCWe

Protein Specific adioactivity

hr xg w CP~/W

0.5 26,000 0.115 16 105,000 0.117

Sediments 9.89

23,400 2,760

115

0.5 26,000 0.140 3,210 Sediments 20.0 308

0.5 26,000 0.070 9,100 16 105,000 4.68 70

40 105,000

Sediments

6.82 250

23.9 1,176

Ultracentrifugal fraction

Lymph Chylomicrons Very low density.. Residual soluble

protein Serum

Chylomicrons. Bottom fraction..

Mucosa Lowest density

lipoprotein.. Very low density.. Low and high den-

sity. Residual soluble

protein.

Lowest density lipoprotein (f hour; 26,000 X 9; d = 1.005).

Other lipoprotein (40 hours; 105,000 x 9; d = 1.21).

Residual soluble protein (sedi- ments at d = 1.21).

g/ml

1.005 1.005

1.21

1.005 1.005

1.005 1.005

1.21

1.21

at pH 2.2 and subjected to chromatography on a Beckman/ Spinco amino acid analyzer.2 Another sample, equivalent to 0.1 mg, was applied to a plate of silica gel G (20 X 20 cm; 0.5-mm

layer thickness), which had been prepared by drying in air at room temperature (30). This was subjected to ascending chro- matography in a first dimension solvent of 1-butanol-acetic acid-water (4:l :l) and a second dimension solvent of phenol- water (75:25) with 20 mg of KCN (31). The amino acid pat- tern was demonstrated with half-strength ninhydrin spray, and

2 We are indebted to Dr. S. C. Hartman for this analysis. -

by guest on June 13, 2018http://w

ww

.jbc.org/D

ownloaded from

1659 Issue of April 25, 1966 Hatch et al.

TABLE IV Effect of inhibitors on leucine incorporation

Incubation conditions were as described in Table I, except for the presence of inhibitory compounds. Lipoprotein fractions were isolated as indicated in Table II.

Puromycin Experiment A

Experiment B

Experiment C

Chloramphenicol Experiment A

Experiment B

Actinomycin D

Ribonuclease Experiment A

Experiment Ba

2,4-Dinitrophenol Experiment Aa

Concentration Activity Inhibition

c?ww %

0 28,500 0.4 x lo- M 12,170

0 771 0.8 x lo- M 10

0 670 2 x lo- M 0

0 529 6 x lo-5 M 696

0 416 6 x 1o-5 M 300

0 3,260 2 x 1o-6 M 3,300 1 x 1o-5 M 2,610

0 2 fig/ml

0 2 fig/ml

0 2 x 1o-4 M

414 310

31 12

31 2

Lowest density lipoprotein

57

99

100

0

28

0 20

25

61

93

Other lipoprotein Residual soluble protein

Activity

6,570 2,350

50 1.5

143 3

226 215

97 28

590 636 529

26 31

136 41

136 40

- Inhibition Activity Inhibition

%

64

97

98

5

71

0 10

0

70

71

13,500 2,320

85 0

229 17

1,042 Gl6 323 146

2,110 2,090 1,095

104 110 122 82

122 31

%

83

100

92

41

55

1 48

0

33

75

(1 Total counts incorporated are reported for these experiments because protein data were unreliable.

the amino acid spots were transferred separately from the plates to small test tubes with as much resolution as possible. The amino acids were estimated in a semiquantitative fashion by carrying out the ninhydrin color reaction according to Moore and Stein (32) and removing the silica gel particles by centrifu- gation before reading in the spectrophotometer. Standard amino acid mixtures were treated similarly, and the recoveries of individual acids were used in calculating the composition of the protein hydrolysates. Ten amino acids and three pairs (Gly- Ser, Leu-Ile, Phe-Trp) could be resolved by this method (cys- teine and methionine were not tested). Acceptable agreement with the amino acid analyzer was obtained.

In a separate experiment, the peptide digestion mixture pre- pared with Nagarse was subjected to thin layer electrophoresis at pH 4.5 in pyridine-acetic acid-1-butanol-water (1: 1:2:36) (33). The peptide locations were demonstrated on a marginal strip with ninhydrin, and the remaining portions of the cellulose, containing the acidic, neutral, and basic peptides, were separately

scraped from the plate, eluted with 5% aqueous 2-propanol, and subjected to total acid hydrolysis as described above. The

amino acid mixtures were chromatographed in two dimensions and measured after elution.

Lipid and Ribonucleie Acid Content-The lowest density lipo- protein from fasted rats was isolated after Sephadex column

chromatography. This was extracted with chloroform-methanol (2: 1) by the method of Albrink (34) ; the purified chloroform layer was concentrated and applied to thin layers of silica gel HR (Brinkmann) on microscope slides according to the method of Vahouny, Borja, and Weersing (35). Development was per- formed with hexane-ether-acetic acid (90: 10: 1) and chloroform- methanol-water (70:30: 1). After staining with 12 vapor, com- parison was made with pure standard lipids.

Ribonucleic acid was estimat,ed on aliquots of the Sephadex- treated lowest density lipoprotein and residual soluble protein by the method of Schneider (36) after alkaline hydrolysis (37).

RESULTS

Incorporation of Radioactive Amino Acids into Lymph Lipoproteins in Vivo

The incorporation of the amino acids of uniformly labeled Chlorella hydrolysate into the ultracentrifugal lipoprotein frac- tions of the lymph is presented in Fig. 1. During the first 2 hours after intragastric administration of the isotopic mixture, there was a rapid increase in the specific radioactivity of all flotatable fractions, with peak levels in the following decreasing order of magnitude: chylomicrons, d < 1.005 and d 1.005 to 1.21. The residual protein fraction fd > 1.21) never reached

by guest on June 13, 2018http://w

ww

.jbc.org/D

ownloaded from

1660 Biosynthesis of Lipoprotein by Rat Intestinal Mucosa Vol. 241, No. 8

as high a specific activity, although, of course, its protein con- serum proteins or to persistence of unlabeled chylomicrons formed tent was much higher than that of the lipoprotein fractions. before administration of the labeled mixture.

In a short term experiment, lymph (1.4 ml) was collected be- tween 30 and 60 min after administration of the labeled amino acids. At the end of the collection period, serum and a ho- mogenate of intestinal mucosal cells were obtained. The results are presented in Table II. In each tissue the chylomicron pro- tein had the highest specific activity. The high level shown by the lymph chylomicrons in comparison to the mucosa is explained by the collection period, which extended into an earlier phase when the mucosa probably possessed a higher radioactivity than when the tissue was removed. The lower level in the serum chylomicrons may be due to dilution with unlabeled

Incorporation of Radioactive Leucine into Protein Fractions of Intestinal Mucosa Cells In Vitro

Characteristics of System-The incorporation of labeled leucine into mucosal protein proceeded at a constant rate for at least 2 hours. The total amount of incorporation into all fractions was variable. Values ranged in different experiments from 0.01 to 0.5% of the added radioactivity. Incorporation was con- sistently higher in 150-g rats than in larger animals.

Glucose was required in the incubation medium when the rats had been fasted or fed pure olive oil by stomach tube, but was not required when they had been fed a pellet of oil-soaked chow. The requirements for penicillin and streptomycin were not clear, but it was felt that their presence helped to keep the incorpora- tion low in control vessels. These antibiotics were present in all the incubation experiments reported.

a b C d

+

Origin

.s i 5:

0 0 :S

a b c

r

:

eo * . .

ab c d e

FIG. 2. A, thin layer (0.37-mm) electrophoresis on potato starch granules in barbital buffer; pH 8.6, P/2 = 0.1, 600 volts, 20 ma, 2 hours. The paper was sprayed with ninhydrin in acetone, and color development was carried out at room temperature. Samples of 20 ~1 containing approximately 100 rg of protein were applied. a, lowest density lipoprotein; b, residual soluble pro- tein; c, lymph chylomicrons (lipid-extracted); d, human serum (reference marker). L3, thin layer descending chromatography on DEAE-cellulose in the Desaga B-N chamber. Phosphate buffer, pH 6.8, 0.02 M, with NaCl gradient increasing to 1 M. a, r-globulin; b, lowest density lipoprotein; c, residual soluble protein. C, thin layer descending chromatography on Sephadex G-100 (fine) in the B-N chamber. Phosphate buffer, pH 7.4, 0.05 M. a, hemoglobin; b, lowest density lipoprotein; c, residual soluble protein; d, r-globulin; e, thyroglobulin.

The distribution of incorporated radioactivity in the protein fractions derived from the successive ultracentrifugations is pre- sented in Table I. In other experiments the specific activity of the residual soluble protein fraction (d > 1.21) was higher than in this case, but the lowest density lipoprotein fraction nearly always exhibited the highest specific activity.

Little relationship existed between the incorporation of radio- active amino acid and the richness in absorbed lipid of the iso- lated mucosal cells, as estimated by their appearance or by measurement of triglyceride by a modification of the method of Van Handel (38). The specific radioactivities of lipoprotein and residual protein fractions also did not differ greatly when fed and fasted rats were compared (Table III). It was of in- terest that, when approximately equal amounts of mucosal tissue were used from fed or fasted rats, the same amount of protein was found in t,he ultracentrifugal fraction of lowest density. However, this fraction was markedly turbid when derived from fed rats and was optically clear when derived from fasted rats. Thin layer chromatography on silica gel of a chloro- form-methanol extract of the lowest density fraction from fasted rats revealed the presence of cholesterol and cholesterol esters, triglycerides, and lecithin. Small quantities of diglycerides and nonesterfied fatty acids were also tentat,ively identified. The glycerides showed overwhelming predominance in a similar extract from an oil-fed rat.

The lowest density lipoprotein and residual protein fraction each contained RNA, approximately 37, of the amount of pro- tein. From 7 to 15% of the protein (Lowry reaction) of lowest density lipoprotein and lymph chylomicrons was soluble in chloroform-methanol (2 : 1).

E$eet of Inhibitors on Amino Acid Incorporation-Data on the effects of several compounds known to inhibit various aspects of cell metabolism are presented in Table IV. The results are consistent with those obtained in certain other mammalian systems in which protein biosynthesis occurs within intact cells (39, 40). However, no attempt has been made at a thorough study of experimental design or inhibitor concentrations.

Studies on Homogeneity of Lipoprotein and Residual Protein Fractions

It would not be expected that the lipoprotein or residual pro- tein fractions isolated from cellular homogenates in the manner

by guest on June 13, 2018http://w

ww

.jbc.org/D

ownloaded from

Issue of April 25, 1966 Hatch et al. 1661

described would have been found to consist of single homogeneous proteins because no techniques of high resolving power were employed. It was important, however, to learn something about the degree of heterogeneity to provide a basis for inter- pretation of the succeeding studies of peptide mapping and amino acid composition. In order to minimize the technical complica- tions resulting from a large amount of lipid in the lowest density lipoprotein, these fractions were isolated from fasted animals and were studied by electrophoresis and chromatography after

purification by Sephadex column chromatography. The studies are summarized in Fig. 2.

The lowest density lipoprotein and the residual soluble protein exhibited similar behavior in several systems. Thin layer elec- trophoresis on starch granules gave the best resolution available (Fig. 2A). The lowest density lipoprotein had one major com- ponent and a minor one which moved only a short distance toward the cathode (perhaps owing to electroendosmosis), and another minor component moving rapidly toward the anode.

FIG. 3. Peptide maps on thin layers of cellulose obtained after stance, the first (horizontal) dimension represents movement extraction of lipids and digestion with Nagarse. A, lowest den- during electrophoresis at pH 3.5 in pyridine-acetate buffer, and sity lipoprotein fraction; B, lymph chylomicrons; C, residual the second (verlical) dimension represents movement during soluble protein fraction sedimenting at d = 1.21. In each in- ascending chromatography in 1-butanol-acetic acid-water (4: 1:l).

by guest on June 13, 2018http://w

ww

.jbc.org/D

ownloaded from

1662 Biosynthesis of Lipoprotein by Rat Intestinal Mucosa Vol. 241, No. 8

The residual soluble protein appeared to have the same three tion of molecular weights of 110,000 for lowest density lipoprotein components, but in nearly equal quantities, plus a minor cathodal and 200,000 for residual soluble protein (42, 43). It is possible band. Lymph chylomicrons from which lipids had been ex- that the Sephadex zones represent stable aggregates of smaller tracted revealed two components, one apparently corresponding monomers, since we have observed a reversible aggregation of to the fast anodal fraction of mucosa and the other having ap- the lowest density fraction (after extraction of lipid) when the proximately the mobility of the main slow band of the lowest ionic strength was lowered by electrodialysis, and some aggrega- density lipoprotein. tion with precipitation during storage at 4”.

Thin layer chromatography on DEAE-cellulose (Fig. 2B) showed a single mobile component (of slightly different mobility) for the two mucosal fractions. Some of the residual soluble protein remained at the origin, possibly corresponding to the prominent anodal band seen in electrophoresis. Column chro- matography of the lowest density lipoprotein on DEAE-Sephadex A-50 was carried out at pH 8.4 with a gradient of Tris-Cl buffer. The protein was eluted in a single peak beginning at a buffer concentration of about 0.22 M.

Mapping of Peptides Derived from Proteolytic Digestion of Protein Fractions

Disk electrophoresis in acrylamide gel columns (41) was rela- tively ineffective because most of the protein of the mucosal and lymph lipoproteins remained within the sample gel, ap- parently uninfluenced by the applied potential.

Thin layer chromatography on Sephadex G-100 (fine) (Fig. 2C) showed single zones for the two mucosal fractions. Comparison of their mobilities with those of pure proteins permitted estima-

Protein Digestion with Various Enzymes-Digestion of the lowest density lipoprotein and residual soluble protein fractions, after Sephadex column treatment, was carried out principally with Nagarse. On the basis of estimation of the peptide products with the reaction of Lowry et al. (21), this enzyme gave more extensive hydrolysis of both fractions than did several other proteases. The yield of soluble peptides recovered after delipida- tion was about 50%; significant mechanical losses occurred in the manipulation of small quantities. Part of the protein was an insoluble “core” which was removed by centrifugation. Further digestion of this “core” with Pronase or Nagarse gave a small additional yield of soluble peptides with practically the same peptide map as the original Nagarse digest.

TABLE V Amino acid composition of protein moiety of lowest

density lipoprotein The lowest density lipoprotein fraction was isolated from a

homogenate of rat intestinal mucosa by centrifugation for 3 hour at 26,060 X g (solvent density, 1.005). The top fraction in the tube was chromatographed on a column of Sephadex G-25 t.o remove contaminating peptide material. The first chromato- graphic peak contained 1.48 mg of protein. Lipids were removed by treatment with organic solvents, and the protein was digested with constant boiling HCl for 40 hours at 105’ in a sealed ampoule. Extraction with hexane was then carried out to remove a trace of residual lipid, and the acid was removed on a rotary evapora- tor. The residue was dissolved in 2.2 ml of citrate buffer at pH 2.2, and 1.0 ml was applied to each column of an automatic amino acid analyzer.

pmoles/ i I 1.48 mg

protein

Lysine. _. 0.291 9 Glycine. 0.386 12 Histidine.. 0.110 4 Alanine. . 0.323 IO Ammonia. 4.72 Cysteine Trace

Arginine. _. 0.220 7 Valine. _. . 0.330 11 Asparagine. 0.403 I3 Methionine” 0.100 4 Threonine.. 0.257 8 Isoleucine 0.242 8 Serine.......... 0.298 10 Leucine. 0.477 15 Glutamic acid.. 0.450 14 Tyrosine. 0.142 5 Proline. 0.235 8 Phenylalanine 0.249 8

Tot.al . 4.51d 146

0 Protein estimated by method of Lowry et ~2. (21). b In estimating number of residues, greatest weight was given

to eight amino acids which are most stable during acid hydroly- sis, and allowances were made for unstable or slowly released residues.

e Sum of methionine and isomeric sulfoxide peaks. d Excludes ammonia.

Somewhat smaller yields of soluble peptides were obtained by digestion of the lowest density lipoprotein fraction with trypsin in the presence of urea (7), with or without subsequent addition of chymotrypsin. Single experiments were also carried out with pepsin and papain, which gave similar results.

Behavior of Peptides in Two-dimensional Electrophoresis- Chromatography-There was a strong resemblance between the peptide maps of the lowest density lipoprotein and residual pro- tein fractions (Fig. 3, A and C). Both maps were unusual in that the majority of the peptides did not move at all in the elec- tric field, or moved only a distance consistent with electroendos- mosis. The electrophoretic immobility of most of the peptides was observed at pH 3.5, 4.5, 6.5, and 9.5. In the second (chro- matographic) dimension, the electrophoretically immobile pep- tides were resolved into about seven spots distinguishable on the basis of mobility or color after ninhydrin spraying, or both. None of the material remained adsorbed at the origin. The electrophoretically mobile peptides were resolved into small numbers of anodal and cathodal spots at pH 3.5, and their mo- bility varied appropriately with pH.

The final two-dimensional pattern consisted of a column of peptides that was resolved only in the chromatographic dimen- sion and a small number of peptides that moved in both dimen- sions. Thus large regions of the map were devoid of any de- tectable peptides. However, many experimental variations of these techniques failed to produce greater resolution of the pep- tide mixture. Under similar conditions, proteolytic digests of hemoglobin, ribonuclease, and albumin were well resolved over all regions of the map.

Amino Acid Composition of Lowest Density Lipoprotein Fraction

Total Amino Acid Composition-The amino acid analysis of the lowest density lipoprotein is presented in Table V. The presence of a significant amount of material tentatively identi- fied as hexosamine is noteworthy. The very large amount of NH, observed cannot as yet be explained; it is far in excess of that which could have been bound in amide linkage with the

by guest on June 13, 2018http://w

ww

.jbc.org/D

ownloaded from

Issue of April 25, 1966 Hatch et al. 1663

dicarboxylic amino acids, but a part may have been derived from degradation of hexosamine.

A portion of the foregoing lowest density lipoprotein was carried through the acid hydrolysis and then subjected to two- dimensional partition chromatography on a thin layer of silica gel G. Two unidentified ninhydrin-positive spots were noted: one was between aspartic and glutamic acids and the other w&s between histidine and proline. The latter may be hexosamine.

Amino Acid Composition of Separated Acidic, Neutral, and Basic Peptide Fractions-A sample of 400 pg of peptide digestion mixture from the lowest density lipoprotein was separated in the electrophoretic dimension only at pH 4.5. After a guide strip at the edge of the plate was colored with ninhydrin, the acidic, neutral, and basic peptide regions were scraped from the plate and eluted from the cellulose. The three fractions were hydrolyzed with HCl, and the amino acid mixtures were analyzed by the twodimensional thin layer technique described. Three noteworthy features were revealed: (a) the neutral peptides contained an essentially complete complement of amino acids; (b) 50% of the residues in the acidic peptides were dicarboxylic amino acids, with small amounts of several other amino acids, and 46% of the residues of the basic peptides were basic amino

acids, with only a few others present; (c) about 80% of the total leucine plus isoleucine content was in the neutral peptide frac- tion, with about 10% each in the acidic and basic peptide frac- tions. The incorporation of radioactive leucine was likewise limited almost entirely to the neutral peptides.

Peptide Maps of Rat Lgmph Chylomicrons and Serum Lipoproteins

Peptide maps were prepared for lymph chylomicrons (Fig. 3B) and chylomicrons, very low density, low density @I), and high density (cyr) lipoproteins of serum (Fig. 4). The pattern ob- tained from lymph chylomicrons was nearly, but not quite, identical with that of the lowest density lipoprotein of intestinal mucosa.

When our technique was applied to the serum lipoprotein fractions, the results were not so satisfactory nor reproducible as with the mucosa and lymph. Drawings of the peptide maps for each serum fraction are presented in Fig. 4. The chylomicron pattern had a column of electrophoretically immobile peptides which was complex in appearance (resembling a double column). There were two streaked zones moving toward the anode that were similar to those of mucosa and lymph, but additional anodal peptides were also present. The very low density lipoprotein

PEPTIDE MAPS OF RAT SERUM LIPOPROTEINS

iolvent front ---------

- Low Denstty Ltpoprot.etn r&t1

FIG. 4. Peptide maps on thin layers of cellulose obtained after represents movement during ascending chromatography in l- ultracentrifugal fractionation of rat serum, extraction of lipids, butanol-acetic acid-water (4: 1: 1). Dashed lines, continuous and digestion with Nagarse. In each map the first (horizontal) lines, and filled outlines represent increasing intensity of ninhydrin dimension represents movement during electrophoresis at pH 3.5 staining. Peptide colors differing from the usual violet are indi- in pyridine-acetate buffer, and the second (vertical) dimension cated in two instances.

.-------- ----

0-y - .-Origin

- Very Low Density Lipoprotein

High Density Lipoprotein (02) +

by guest on June 13, 2018http://w

ww

.jbc.org/D

ownloaded from

1664 Biosynthesis of Lipoprotein by Rat Intestinal Mucosa Vol. 241, No. 8

exhibited a simple pattern with over-all resemblance to those of the lowest density fraction and lymph chylomicrons. Close examination, however, showed a lack of correspondence of the individual peptides. Low density lipoprotein gave a map with much greater complexity in the anodal region. High density lipoprotein showed a larger number of individual peptides, with greater scatter over the plate, and relatively fewer zones did not move in the electrophoretic dimension.

DISCUSSION

The role of the intestine in lipoprotein metabolism is poorly understood. Biosynthesis of protein by the intestinal mucosa appears to be essential for lipid transport, since Sabesin and Isselbacher have recently reported that inhibition of protein synthesis in the rat prevents chylomicron formation and ali- mentary lipemia (44). With regard to the specificity of protein in the chylomicron, previously reported evidence is not definitive. On the one hand, it is suggested that specific chylomicron pro- teins may be synthesized by the intestinal mucosa (7, 8) ; on the other hand, nonspecific adsorption of proteins into chylomicrons from the environment is favored (9).

In the studies reported here, active biosynthesis of mucosal proteins has been demonstrated in viva and in vitro by means of the incorporation of labeled amino acids and the effects of in- hibitors. The fraction of highest specific activity was found to be associated with lipid during its absorption across the intes- tinal mucosa. This protein fraction was not homogeneous, but exhibited several unusual properties in addition to its affinity for lipids: (a) a predominance of apolar amino acids (Table V) similar to that which is present in two other lipoproteins, struc- tural protein of the mitochondrion, and myelin proteolipid (45) ; (b) resistance to fractionation procedures involving ionic prop- erties; (c) considerable solubility in organic solvents; and (d) peculiarities of the peptide map that are related to the first three properties.

Our studies again fall short of settling the dilemma about the specificity of the protein which becomes associated with lipid during transport. Further comment depends almost entirely upon a speculative interpretation of the starch electrophoretic patterns and the unusual peptide maps observed for various fractions. The key to this interpretation is the notion that patterns obtained from the residual protein fraction presumably represent the most abundant soluble proteins of the mucosal cells. There is an incomplete but strong resemblance among the electrophoretic patterns (Fig. 2A) and the peptide maps (Fig. 3) of (a) the residual protein fraction, (b) the lowest density lipoprotein, and (c) the intestinal lymph chylomicrons. It is possible that this resemblance means that lipids in transit through the mucosa become associated with some of the more abundant soluble proteins of the cells and that the resulting lipoprotein complex then appears in the intestinal lymph. One would par- ticularly wish to understand the relationship between the pro- teins we have described and the endoplasmic reticulum of the mucosa, for this cell organelle is believed to play an important role in intestinal fat absorption (46-48).

Thin Layer Methodology

A brief comment is appropriate on the usefulness of thin layer methodology in this work. In the mapping and measurement of radioactivity of peptides obtained from proteolytic digestion

and in the semiquantitative assay of amino acid composition, it has been feasible to work with amounts ranging between 25 and 100 pg of material. Without the availability of these methods it would have been necessary to pool samples from substantial numbers of animals, and the procedures would have become much more cumbersome.

Acknozoledgments-The authors acknowledge helpful discus- sions of various aspects of this work with Drs. J. M. Buchanan, V. M. Ingram, E. Haber, S. C. Hartman, K. J. Isselbacher, and R. B. Loftfield. Dr. G. I’. Canellos assisted in the initial phase of this work, and Miss C. E. Anderson participated during a summer under the Advance Studies in Science Program of Thayer Academy and the National Science Foundation. The typescript was prepared by Mrs. B. L. Jacobson.

1.

2.

3.

4. 5. 6. 7.

8.

9. 10. 11.

12.

13.

14.

15. 16.

17.

18.

19.

20.

21.

22.

23. 24.

25.

26. 27. 28

REFERENCES

HATCH, F. T., HAGOPIAN, L. M., RUBENSTEIN, J. J., AND CANELLOS, G. P., Circulation, 28, 659 (1963).

HATCH, F. T., Aso, Y., HAGOP~AN, L. M., A& RUBENSTEIN, J. J., Abstracts of the 148th Meeting of the American Chemical Society, Chicago, September 1964, p. 43C.

LINDGREN, F. T., AND NICHOLS, A. V., Plasma Proteins, 2, 1 (1960).

DOLE, V. P., AND HAMLIN, J. T., Physiol. Rev., 42, 674 (1962). BRAGDON. J. H.. J. Lab. Clin. Med.. 62. 564 (1958). SCANU, A:, AND GAGE, I. H., J. Expk Med., iO9, i39 (1959). RODBELL, M., AND FREDRICKSON, D. S., J. Biol. Chem., 234,

562 (1959). RODBELL, M., FREDRICKSON, 1). S., AND ONO, K., J. Biol.

Chem., 234, 567 (1959). PETERSON, M. L., Thesis, The Rockefeller Institute, 1960. ISSELBACHER, K. J., AND BUDZ, D., Nature, 200, 364 (1963). SCANU, A., AND HUGHES, W. L., J. Biol. Chem., 236, 2876

(1960). FURMAN, R. H., HOWARD, R. P., LAKSHMI, K., AND NORCIA,

L. N., Am. J. Clin. N&r., 9, 73 (1961). LEVY, R. I., LEES, R. S., AND FREDRICKSON, D. S., J. Clin.

Invest., 44, 1068 (1965). RADDING. C. M.. AND STEINBERG. D.. J. Clin. Invest.. 39.

1560 (1960). I , , ,

MARSH, J. B., J. Biol. Chem., 238, 1752 (1963). FREDRICKSON, D. S., ALTROCCHI, P. H., AVIOLI, L. V., GOOD-

MAN, D. S., AND GOODMAN, H. C., Ann. Internal Med., 66, 1016 (1961).

SALT, H. B., WOLFF, 0. H., LLOYD, J. K., FOSBROOKE, A. S., CAMERON, A. H., AND HUBBLE, D. V., Lancet, 2, 325 (1960).

ISSELBACHER, K. J., SCHEIG, R., PLOTKIN, G. R., AND CAUL- FIELD, J. B., Medicine, 43, 347 (1964).

DAWSON, A. M., AND ISSELBACHER, K. J., J. Clin. Invest., 39, 150 (1960).

UMBREIT, W. W., BURRIS, R. H., AND STAUFFER, J. F., Mano- metric techniques, Ed. 3, Burgess Publishing Company, Minneapolis, 1957, n. 149.

LOWRY, 0: H.,‘RosEBRouGH, N. J., FARR, A. L., AND RANDALL, R. J.. J. Biol. Chem.. 193. 265 (1951).

LOFTFIELD, R. B., AND EIGNER,‘E. k., Acta Chem. Stand., 17 (suppl.), 117 (1963).

DETERMANN, H., Ezperientia, 18, 430 (1962). BANASZAK, L. J., AND MCDONALD, H. J., Biochemistry, 1, 344

(1962). ASHWORTH, L. A. E., AND GREEN, C., Biochem. J., 89, 561

(1963). GUSTAFSON, A., Biochim. Biophys. Acta, 84, 223 (1964). HAGIHARA. B., Enzvmes, 4, 195 (1960). NAUGHTON, M!. A.,SANGER, F., HARTLEY, B. S., AND SHAH,

D. C., Biochem. J., 77, 149 (1960). 29. LEWIS, P. R., Biochem. J., 62, 330 (1952). 30. BRENNER, M., NIEDERWIESER, A., AND PATAKI, G., in E.

by guest on June 13, 2018http://w

ww

.jbc.org/D

ownloaded from

Issue of April 25, 1966 Hatch et al. 1665

STAHL (Editor), Thin-layer chromatography, Academic Press, Inc., New York, 1965, p. 394.

31. BRENNER, M., NIEDERWIESER, A., AND PATAKI, G., in E. STAHL (Editor), Thin-layer chromatography, Academic Press, Inc., New York, 1965, p. 399.

32. MOORE, S., AND STEIN, W. H., J. Biol. Chem., 211, 893 (1954). 33. DINTZIS, H. M., Proc. Natl. Acad. Sci. U. S., 47, 247 (1961). 34. ALBRINK, M. J:, J. Lipid Res., 1, 53 (1959). 35. VAHOUNY, G. V., BOFLIA, C. R., AND WEERSING, S., Anal.

Biochem., 6, 555 (1963). 36. SCHNEIDER, W. C., Methods Enzymol., 3, 680 (1957). 37. FLECK. A., AND MUNRO, H. N., Biochim. Biophys. Acta, 66,

571 (196i). 38. VAN HANDEL, E., Clin. Chem., 7, 3 (1961). 39. FRUTON, J. S., Proteins, 1, 189 (1963).

40. BROCKMAN, R. W., AND ANDERSON, E. P., Ann. Rev. Biochem., 32, 497, 1963.

41. ORNSTEIN, L., AND DAVIS, B. J., Disc electrophoresis, Distilla- tion Products Industries, Rochester, New York, 1962.

42. WHITAKER, J. R., Anal. Chem., 35, 1950 (1963). 43. ANDREWS, P., Biochem. J., 91, 222 (1964). 44. SABESIN, S. M., AND ISSELBACHER, K. J., Science, 147, 1149

(1965). 45. HATCH, F. T., Nature, 206, 777 (1965). 46. PALAY, S. L., AND KARLIN, L. J., J. Biophys. Biochem. Cytol.,

6, 373 (1959). 47. SJ~STRAND, F. S., in A. C. FRAZER (Editor), Biochemical

problems of lipids, Vol. 1, Elsevier Publishing Company, New York, 1963, p. 91.

48. STRAUSS, E. W., J. Cell Biol., 17, 597 (1963).

by guest on June 13, 2018http://w

ww

.jbc.org/D

ownloaded from

Frederick T. Hatch, Yoshiro Aso, Lillian M. Hagopian and Joel J. RubensteinBiosynthesis of Lipoprotein by Rat Intestinal Mucosa

1966, 241:1655-1665.J. Biol. Chem. 

  http://www.jbc.org/content/241/8/1655Access 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/241/8/1655.full.html#ref-list-1

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

by guest on June 13, 2018http://w

ww

.jbc.org/D

ownloaded from