vol. 253, no. 24, issue of december 25. pp. 90534062, 19% ... · role of the lysine residues of...

Vol. 253, No. 24, Issue of December 25. pp. 90534062, 19% Prmted m C’S.A.

Role of the Lysine Residues of Plasma Lipoproteins in High Affinity Binding to Cell Surface Receptors on Human Fibroblasts*

(Received for publication, May 11, 1978)

Karl H. Weisgraber, Thomas L. Innerarity,$ and Robert W. Mahley$j

From the Laboratory of Experimental Atherosclerosis, National Heart, Lung, and Blood Institute, National Institutes of Health, Bethesda, Ma&land 20014

The low density lipoprotein (LDL) cell surface receptors on human fibroblasts grown in culture bind specific plasma lipoproteins, initiating a series of events which regulate intracellular cholesterol metabolism. Specificity for the interaction with the receptors re- sides with the protein moieties of the lipoproteins, spe- cifically with the B and E apoproteins of LDL and certain high density lipoproteins (HDL, HDL1), respectively. It was previously established that the amino acid a&nine is a functionally significant residue in or near the recognition sites on the B and E apoproteins and that modification of this residue abolishes the ability of these apolipoproteins to bind to the receptor. The present study indicates that lysine residues are also involved in the lipoprotein-receptor interaction. Chem- ical modification of 15% of the lysine residues of LDL by carbamylation with cyanate or 20% by acetoacetylation with diketene prevents the LDL from competitively displacing unmodified lz51-LDL from the high affinity receptor sites or from binding directly to the receptor. Moreover, quantitative reversal of the acetoacetylation of the lysine residues of LDL by hydroxylamine treatment regenerates the lysyl residues and re- establishes greater than 90% of the original binding activity of the LDL. The reversibility of this reaction establishes that the loss of binding activity which follows lysine modification is not due to an irreversible alteration of the LDL or HDL, but is probably due to an alteration of a property of the recognition site associ- ated with specific lysine residues. While acetoacetylation and carbamylation neutralize the positive charge on the c-amino group of lysine, reductive methylation selectively modifies lysine residues of LDL and HDL, without altering the positive charge, yet abolishes their ability to bind to the receptor. Preservation of the charge but loss of binding activity following reductive methylation of the lipoproteins suggests that the specificity of the recognition site does not reside simply with the presence of positive charges but depends on other more specific properties of the site determined by the presence of a limited number of the lysine (and a&nine) residues. The precise role of lysine remains to be defined, but its function may be to establish and maintain the conformation of the recognition site or the alignment of reactive residues, or both, or to chemically react, through its e-amino group, with the recep-

* Portions of this work were performed under contract with Meloy Laboratories, Inc., Springfield, VA. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “aduertisement” in accord- ance with 18 USC. Section 1734 solely to indicate this fact.

$ Present address, Meloy Laboratories, Inc., Springfield, VA 22151. 8 To whom correspondence and reprint requests should be ad-

dressed at: Building 10, Room 5N204, National Institutes of Health, Bethesda, MD 20014.

tor (hydrogen bond formation would be such a possi- bility).

The low density lipoprotein (LDL) cell surface receptors on human fibroblasts have been shown to bind not only the apo- B-containing LDL’ (1,2) but also certain specific high density lipoproteins (HDL) which contain the arginine-rich (apo-E) apoprotein (3-6). It has been established that the protein moiety is the determinant for lipoprotein binding and that either the B (LDL) or E (HDL,) apoprotein is responsible for high affinity binding to the cell surface receptor (7). Further- more, the importance to the binding reaction of positively charged regions on the lipoprotein is suggested by the observation that heparin can displace LDL from the receptors, presumably by binding to cationic sites on the lipoproteins (8). It now appears that both the B and E apoproteins contain a similar positively charged domain or structural sequence which confers binding specificity on the lipoproteins (7,9).

We have shown previously that modification of the amino acid arginine abolishes the ability of specific lipoproteins to bind to the cell surface receptors (7). Reaction of the guanido group of arginine with the highly selective reagent 1,2-cyclohexanedione produces a stable complex and totally inactivates LDL and HDL, after modification of 40 to 50% of the arginyl residues of apo-B or apo-E. The observation that the amino acid arginine was functionally significant in the recognition site of these lipoproteins led to speculation that the B and E apoproteins contained similar structural sequences, common to both apoproteins which were involved in the high affinity binding (7, 9). Moreover, the E apoprotein, which is rich in arginine, enhances the binding activity of HDL, lo- to lOO- fold over the LDL activity, suggesting that increased binding activity may be directly correlated with an increased number of arginine-rich domains (4).

To determine whether other amino acid residues may be functionally significant in the lipoprotein recognition site, different amino acid residues have been selectively modified and the effects of such modifications on binding activity have been measured. This paper reports data from those mods- cation studies which show that a limited number of lysyl residues are also involved in the binding of LDL and HDL, to the cell surface receptors of cultured human fibroblasts. Var- ious types of lysine modification, including carbamylation and acetoacetylation with diketene (which neutralize the positive charge) and reductive methylation (which preserves the charge on lysine), abolish binding activity. The importance of

’ The abbreviations used are: LDL, low density lipoproteins; HDL, high density lipoproteins; apo-E, the arginine-rich apoprotein; DME medium, Dulbecco’s modified Eagle’s medium; DNFB, dinitrofluorobenzene.

9053

by guest on June 8, 2018http://w

ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

9054 Modification of Lysine Residues of LDL and HDL,

lysine, as well as arginine, in the interaction of specific lipoproteins containing the B and E apoproteins is discussed.

EXPERIMENTAL PROCEDURES

Materials-Dulbecco’s modified Eagle’s medium (Catalogue No. 430-2100), fetal calf serum, trypsin/EDTA solution, Dulbecco’s phosphate-buffered saline (Catalogue No. 450-1300), potassium penicillin G, and streptomycin sulfate were purchased from GIBCO (Grand Island, NY). Sodium [““Iliodide (carrier-free) in NaOH was obtained from AmershamSearle (Arlington Heights, IL). Certified reagent grade potassium cyanate, formaldehyde solution (37%, w/w), sodium borate, and hydroxylamine hydrochloride were obtained from Fisher Scientific Co. (Fair Lawn, NJ). Sodium borohydride, N-acetylimidazole, diketene, and dinitrofluorobenzene were purchased from Sigma Chemical Co. (St. Louis, MO). Constant boiling hydrochloric acid was obtained from Pierce Chemical Co. (Rockford, IL).

Zsolation of Plasma Lipoproteins-Human LDL (d = 1.02 to 1.05) were prepared from plasma obtained from a healthy fasted male subject by ultracentrifugation in a 60 Ti rotor at 59,006 rpm for 18 h and recentrifuged at d = 1.05 for an additional 16 h. Canine HDL, were isolated from the d = 1.006 to 1.02 fraction obtained from the plasma of foxhounds which had been maintained on a semisynthetic diet containing cholesterol and coconut oil as previously described (10). Human lipoprotein-deficient serum was prepared by ultracentrifugation at 59,000 rpm for 48 h at d = 1.215 to remove the lipoproteins. The d > 1.215 fraction was dialyzed fist against 0.15 M

NaCl, 0.01% EDTA, pH 7.0, then against phosphate-buffered saline; the volume was adjusted to the original serum volume by adding phosphate-buffered saline; and the serum was sterilized by filtration and stored at -20°C until needed. The cholesterol content of the lipoprotein-deficient serum was determined by gas chromatography to be less than 2 pg/ml.

Lipoprotein Characterization-The lipoproteins were character- ized by paper electrophoresis, negative-staining electron microscopy, and sodium dodecyl sulfate-polyacrylamide gel electrophoresis as previously described (11, 12). Geon-Pevikon block electrophoresis of the modified lipoproteins was performed as reported (13). Lipid analyses included total (14) and esterified (15) cholesterol, triglyceride (16), phospholipid (17), and protein (18).

Zodination-Human LDL were iodinated by the iodine monochlo- ride method of McFarlane as modified by Bilheimer et al. (19). The labeled lipoproteins were dialyzed at 4°C for 48 h against 0.15 M NaCl, 0.01% EDTA, pH 7.0, and sterilized by filtration. The “:‘I-LDL were used for up to 4 weeks after iodination but were redialyzed immedi- ately before use.

Cells in Culture-Human libroblasts from a normal human pre- putial specimen were maintained as monolayer cultures in 75-cm’ flasks in a humidified incubator (5% CO*) at 37°C and were used between the fourth and tenth passages as described (7). Confluent monolayers were dissociated from the flasks by treatment with 0.5% trypsin, 0.02% EDTA. The cells were transferred to four 75-cm’ flasks for maintenance or to 60-mm Petri dishes (9 x IO4 cells/dish) for use in experiments.

All experiments were performed using procedures previously described (4). The stock cultures were transferred to Petri dishes 7 days prior to the experiment. After 5 days of growth on DME medium supplemented with 10% fetal calf serum, each monolayer was washed twice with DME medium containing 5% (v/v) lipoprotein-deficient serum, and then 3 ml of DME medium containing 10% lipoprotein- deficient serum was added. The experiments were begun 48 h later on Day 7 when the cells were nearly confluent.

Assays for Binding, Internalization, and Degradation-The binding (surface-bound plus internalized) and degradation assays were performed at 37°C utilizing the methods of Goldstein and Brown (20) with minor modifications (4). Binding studies were performed at 4°C according to Goldstein et al. (8) except that 4-(2-hydroxyethyl)-l- piperazineethanesulfonic acid buffer (25 mM; pH 7.4) was used (4).

Assay for the Incorporation of [‘4C]Oleate into Cholesteryl Es- ters-The method for the measurement of the incorporation of [‘?]oleate into cholesteryl esters by the cell monolayers has been described previously (4, 21). Cellular cholesterol [14C]ester content was determined after thin layer chromatographic isolation.

Acetoacetylation of Lipoproteins-The LDL (2 to 5 mg/ml of protein) were diluted with 0.3 M sodium borate buffer, pH 8.5, to 1.5 times the original volume. The diketene reagent (0.06 pmol/p) was prepared by dissolving 50 4 of diketene (redistilled and stored at -20°C) in 10 ml of 0.1 M sodium borate buffer, pH 8.5. For each

milligram of LDL protein to be modified, 0.04 to 1.29 pmol of diketene was added and the mixture was allowed to stand for 5 min at room temperature. The reaction was stopped by dialysis against 0.2 M carbonate/bicarbonate buffer, pH 9.5, for 4 to 6 hat room temperature and then overnight at 4’C. The modified LDL were dialyzed for an additional 16 h against 0.15 M NaCl, 0.01% EDTA, pH 7.0, prior to use in the binding and degradation assays. The amount of diketene reagent added to the LDL was determined empirically by the effect it had on the LDL binding activity, and the extent of acetoacetylation was quantitated by amino acid analysis (see below) or by the FeCl., method of Marzotto et al. (22) using an acetoacetylglycine calibration curve. Approximately 20% of the lysine residues of LDL were modified by the addition of 0.43 to 0.56 gmol of diketene, a level which abolished binding activity. The diketene procedure was based on a previously described method (22).

Reversal of Lysine Diketene Modification-Treatment with hydroxylamine reversed the acetoacetylation and regenerated the lysyl residues on LDL. The method, described previously (22), was modified as follows. Hydroxylamine (3 M, pH 7.0) was added to the modified LDL (2 to 10 mg) to give a final concentration of 0.5 M. The mixture was allowed to react for 16 h at 37°C. Excess reagents were removed by dialysis at 4°C against 0.15 M NaCl, 0.01% EDTA, pH 7.0, for 24 h. The 16-h incubation with hydroxylamine was necessary to reverse the modification and regenerate greater than 90% of the original activity.

Carbamylatzon of Lipoproteins-The LDL (2 to 5 mg/ml of protein) in 0.15 M NaCl, 0.01% EDTA, were diluted with 0.3 M sodium borate buffer, pH 8.0, to 1.5 times the original volume. Potassium cyanate (20 mg/mg of LDL protein) was added to the lipoprotein solution, and the mixture was incubated at 35°C for various lengths of time (5 min to 2 h), depending on the desired extent of carbamylation. Excess reagents were removed by dialysis at 4°C against 0.15 M NaCl, 0.01% EDTA, pH 7.0, for 36 h before the lipoproteins were used in the binding and degradation assays. This procedure was based on a previously described method (23). The extent of carbamylation was determined by amino acid analysis as described below.

Reductive Methyl&ion of Lipoproteins-The LDL (2 to 10 mg/ml of protein) in 0.15 M NaCl, 0.01% EDTA, pH 7.0, were diluted with 0.3 M sodium borate buffer, pH 9.0, to 1.5 times the original volume. Reductive methylation of 6 to 10 mg of LDL protein was performed at 0°C by addition of 1 mg of sodium borohydride (zero time for reaction sequence) followed by six additions over 30 min of 1 ~1 of 37% aqueous formaldehyde (additions were made at zero time and 6, 12, 18, 24, and 30 min). After the last addition of formaldehyde, the reaction mixture was chromatographed on Sephadex G-50 to stop the reaction, and then the modified LDL were dialyzed at 4°C against 0.15 M NaCl, 0.01% EDTA, pH 7.0, for 18 h. For more extensive modification of LDL, the 30-min reaction sequence was repeated with a total of 12 additions of formaldehyde (zero time and at 6-min intervals for 60 min) and 2 additions of borohydride (zero and 30 min). When a progressive increase in the level of modification was needed, as shown in Table I, aliquots of the reaction mixture were taken at various times throughout the formaldehyde addition sequence following the zero time addition of borohydride, and the reactions were stopped by chromatographing the aliquots on Sepha- dex. Two, four, or eight 1-d aliquots of formaldehyde were added at 6-min intervals, respectively, for the lo-, 20-, and 45-min modifications. The procedure was based on a previously described method (24). The extent of modification was determined by amino acid analysis as detailed below.

Acetylation of LDL with N-Acetylimidazole-N-Acetylimidazole (1 ml of 0.91 M) in phosphate buffer was added to 3 ml of LDL (3.3 mg/ml) in 0.005 M sodium phosphate, 0.1 M KCl, pH 7.4, at 0°C. The reaction was stopped by passing the reaction mixture over Sephadex G-50. The LDL were then dialyzed at 4°C against 0.15 M NaCl, 0.01% EDTA, pH 7.0, for 36 h prior to use. Under these conditions, a reaction time of 1 min was sufficient to abolish binding activity. This procedure was based on a method previously described (25).

Reduction and Alkvlation of LDL-LDL (2 ma/ml of nrotein) in 0.2 M phosphate buffer, pH 8.0; were reduced by treatment with 3 ~1 of P-mercaptoethanol at room temperature for 4 h. Alkylation of the reduced LDL was performed by adding iodoacetamide (7 mg) and allowing the mixture to stand for 30 min at 0°C in the dark. The excess reagents were removed by dialysis against 0.15 M NaCl, 0.01% EDTA, pH 7.0, at 4°C. To determine the extent of alkylation a parallel reaction with iodo[‘4C]acetic acid (7.5 pCi/pmol) was performed. Based on half-cystine levels of 0.047 pmol/mg of LDL protein (2 half-cystine residues/mol, assuming 250 residues) (26), alkylation


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

Modification of Lysine Residues of LDL and HDL, 9055

of 30% of the half-cystine groups was obtained. Alkylation without prior reduction with ,&mercaptoethanol resulted in 2% of the half- cystine groups being modified. Results with both procedures are consistent with results reported by Margolis and Langdon (26).

Amino Acid Analysis-The lipoproteins for amino acid analysis were dialyzed at 4°C against 0.01% EDTA in water, pH 7.0. The lyophilized lipoproteins (60 pg of protein in l-ml ampules) were delipidated with CHCL:MeOH (2:1, v/v) and the apoprotein residue was dried under vacuum. Hydrolyses were performed in a sealed nitrogen atmosphere for 22 h at 110°C. Analyses were performed on a Beckman model 121M amino acid analyzer (Beckman Instruments, Palo Alto, CA) using a sodium citrate or lithium citrate buffer program. Depending on the modification, various methods (detailed below) were used to quantitate the extent of modification.

It was necessary to determine the extent of diketene modification by an indirect procedure, because the acetoacetyl derivative of lysine was converted back to lysine under acid hydrolysis conditions. Treat- ment of the modified protein (containing both unreacted lysyl and acetoacetyllysyl residues) with dinitrofluorobenzene (DNFB) (27) converted the unreacted lysine residues into acid-stable dinitrophenyl derivatives. Acid hydrolysis converted the acetoacetyllysine to lysine, which was quantitated by amino acid analysis. This value represented the number of lysine residues acetoacetylated. Greater than 94% of the lysine residues of unmodified LDL-or HDL, reacted with the DNFB, and this value was subtracted from the results obtained with the modified lipoproteins to correct for the unreactive lysine (less than 1 residue). Duplicate determinations were performed on each sample and the controls as follows. Following the addition of 1 ml of absolute ethanol and 50 mg of NaHC03 to the apoprotein residue, 25 4 of dinitrofluorobenzene (DNFB) was added to each ampule and the mixture was allowed to stand in the dark at room temperature for 28 h. The mixture was centrifuged at 2,500 rpm for 10 min, the supernatant was removed, and the dinitrophenylated protein was washed once with absolute ethanol, three times with distilled water, and once with acetone (27). Hydrolysis and amino acid analysis were performed as described above.

Treatment of protein with cyanate (carbamylation) converts lysine to homocitrulline, which in turn is partially degraded back to lysine during acid hydrolysis (17 to 30% depending on the protein and extent of carbamylation) (28). The extent of carbamylation was determined by adding the number of homocitrulline residues determined by direct measurement to the number of free lysine residues obtained by the DNFB method. Both determinations were performed in duplicate on the same sample.

Quantitation of the extent of reductive methylation was based on the difference between lysine content in the modified versus the unmodified lipoprotein. Reductive methylation of lysine produced a mixture of mono- and dimethyllysine, with the latter predominating at the longer reaction times (65% was in the dimethyl form after a 30- min reaction). In order to resolve these derivatives from lysine on the analyzer, it was necessary to use a lithium citrate buffer program recommended for physiological fluids (Beckman Instruments). Du- plicate determinations were performed on each modified sample and on the unmodified control.

The quantitative results of the various modifications (Table I) were expressed in number of lysine residues modified per 250 amino acid residues for LDL (20.2 lysine residues/250 residues in the unmodified LDL) and per 290 residues for HDL, (15.8 lysine residues/290 residues).

RESULTS

Acetoacetylation of Lysine-Diketene has been described as an excellent reagent for the modification of lysine residues of various proteins and extremely useful because its modification is reversible and the lysine is regenerated (22, 29). The reaction, an acetoacetylation of amino groups, can be rela- tively specific for lysine depending on the conditions, but may also modify tyrosine and serine in some cases (22). However, the acetoacetylation of tyrosine and serine is rapidly and easily reversed by dialysis of the modified proteins in carbonate/bicarbonate buffer (22). By contrast, the modified lysine is stable not only in the carbonate/bicarbonate buffer but also under the conditions of the tissue culture experiments reported here. As shown in Table I, the number of diketene- modified lysine residues of human LDL was determined by

the amount of reagent added, as described under “Experimen- tal Procedures.” It was particularly important for these plasma lipoprotein studies that reversal of the acetoacetylation regenerated greater than 95% of the lysine residues which had been modified (Table I). Incubation of the diketene- modified lipoproteins with hydroxylamine for 16 h at 37°C reversed the modification, regenerated the lysine residues, and established that the lipoproteins had not been irreversibly changed (see below). Under the conditions described here, the only amino acid which was modified was lysine. This was confirmed using calorimetry to quantitate the acetoacetyl groups and then relating this value to the number of lysine residues modified (as determined by amino acid analysis) (22).

Reaction of 0.56 pmol of diketene/mg of LDL protein mod- fied 4.3 out of a total of 20.2 lysyl residues (calculated per mol, assuming 250 residues/mol, Table I) and completely elimi- nated the ability of the modified LDL to compete with ‘251- LDL for binding, internalization, and degradation by human fibroblasts (Fig. 1). Greater than 90% of the binding activity was restored to the modified LDL by hydroxylamine treatment to remove the diketene (Fig. 1). For untreated LDL, the competitive displacement curve (bottom curue) was typical for studies performed at 37°C. The effect of diketene on the binding activity was further investigated by directly modifying the ‘251-LDL and performing a direct binding assay at 37°C. The results were identical with those obtained from the competitive assay and indicated that modification of approximately 4 out of 20 residues of lysine prevented LDL from binding to the LDL cell surface receptor of fibroblasts.

Likewise, when the study was performed at 4”C, a temperature at which only binding and no internalization occurs (8), the modified LDL were incapable of displacing 12”1-LDL from the high affinity receptor sites, indicating a direct effect of diketene on the binding process (Fig. 2). Total binding activity was restored to the LDL after reversal of the modification (Fig. 2), and amino acid analysis revealed that most of the lysine groups had been regenerated by the hydroxylamine treatment. Data related to number of residues modified and alterations in the binding activity are compiled in Table I.

The reaction of human LDL with diketene, at concentrations of approximately 0.5 to 1 pmol of diketene/mg of LDL protein, totally abolished binding activity and modified 4 to 7 lysine residues (Table I), but did not alter the physical or chemical properties of the lipoproteins, except for their elec- trophoretic mobility. On paper electrophoresis (Fig. 3), the diketene-treated LDL migrated further toward the anode than did the native LDL, reflecting the neutralization of the positive charge on the modified lysyl residues. Following removal of the acetoacetyl groups by hydroxylamine treatment, the LDL migrated as a sharp P-band with migration identical with that of untreated LDL (Fig. 3). Untreated LDL, diketene- treated LDL, and regenerated LDL were the same size and had the same morphologic appearance by negative staining electron microscopy (Fig. 4). The chemical composition (Ta- ble II) and the apoprotein pattern on sodium dodecyl sulfate- polyacrylamide gel electrophoresis were likewise unchanged by the treatment. These results in combination with the binding assay data indicated that the reaction was mild and that most of the original activity could be restored to the LDL following reversal of the diketene modification of the lysyl residues.

HDL, were also modified by diketene treatment. Reaction of cholesterol-induced canine HDL,. with 3 pmol of diketene/mg of protein abolished all competitive binding (Fig. 5). Untreated HDL, and HDL, regenerated after hydroxylamine reversal of the diketene modification gave essentially identical results. The HDL, required more extensive treat-


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/


TABLE I Effect of chemical modification of human LDL on binding and internalization by the cell receptors of fibroblasts as a function of the

number of amino acid residues modified

Residue modiiied Structural modification

Native LDL

Diketene

0.04 am01 0.24 pmol 0.43 pm01 0.56 pmol 1.08 am01 1.29 pm01

0.56 pmol (reversed) 1.08 pm01 (reversed)

Carbamylation

5 min 15 min 30 min 60 min 75 min 90 min

120 min

Reductive methylation

10 min 20 min 30 min 45 min 60 min

Reduction/alkylation

Cyclohexanedioned Arg 4-5

LYS 0 0

II II -NH-C-CHz-C-CHa

LYS

LYS

0

II -NH-C-NH2

+ -NH-(CHs),

CYS 0

II -S-CH,- C --NH2

“Results from three independent studies (I to III) in which the number of residues of the specific amino acid (lysine, cysteine, or arginine) modified was determined by amino acid analyses and in which the change in binding activity (37’C competitive assay) was expressed as a percentage of the original binding/internalization activity of the unmodified LDL at 50 or 100 pg of protein/ml of media, e.g. with diketene modification (I) using 0.24 pmol of reagent, 1.5 residues of lysine were modified, and the modified LDL displaced only 49% as much lz51-LDL as compared to the unmodified LDL.

ment to abolish totally the binding activity (approximately 5 times as much diketene as required for LDL). Amino acid analyses revealed that 50% of the total lysine residues (-8 out of a total of 15.8 residues of lysine, assuming 290 amino acids/mol) had to be modified to prevent binding. The HDL, used in these studies were isolated by Geon-Pevikon block electrophoresis from the d = 1.006 to 1.02 ultracentrifugal fraction and contained the arginine-rich apoprotein (apo-E) as the only protein constituent. The results indicated that high affinity binding of both apo-B-containing LDL and apo- E-containing HDL, were similarly abolished by lysine modification.

Carbamylation of Lysine-Lysine residues of human LDL were also modified by treatment with potassium cyanate. This reagent has been shown to react with amino groups of various

0.8 1.5 1.8

0.7 1.1 1.8 1.9

3.0

1.7 61 1.7 5.5 19 2.9 8.2 3 5.8 9.7 0

13.0 0

0.6 100

95 49 28

3.6

98 76 1.9 48 2.4 10

(2.83Z0,’ 8

7.4

<lO

0 4.3 0 7.0 0

97

0.4 96

0 95 100

10

c:, 2.7 12 3.2 0

0

74 50

0

19 0

Data for the competitive degradation assays parallel exactly the results of the binding.

b Number of residues of lysine (Lys), cysteine-cystine (Cys), and arginine (Arg) was calculated assuming 250 amino acid residues/m01 of protein. Based on this assumption, there are 20.2 lysine residues, 2 residues of half-cystine, and 9 residues of arginine/mol of protein.

’ Two additional studies were performed in which the modification of 2.8 or 3.0 residues of lysine abolished binding activity.

d Calculated from previous work (7).

proteins (28, 30, 31), and, under the conditions employed, it reacted with the e-amino group of lysine (for discussion of the apparent umeactivity of the NH2-terminal group of LDL and HDL,, see “Reductive Methylation of Lysine” below). We have consistently found that modification of 3 out of 20.2 lysine residues abolished the binding activity of the modified lipoprotein (Table I). In the particular study shown in Fig. 6, this was accomplished by a 30-min reaction time. Our data indicated that carbamylation of 3 residues was required to abolish binding activity but did not suggest that all 3 of the residues were key or essential for binding. We observed in two different experiments that, while modification of 0.7 or 1.9 residues did not affect the binding activity, modification of 3 residues abolished 92 or 100% of the binding, respectively (Table I). That only 1 or 2 (and not all 3) lysyl residues may


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/


I ’ I ’ 1 ’ 1 ’ ’ I t I ’ I ’ I ’ 1 A BINDING

250 01 B DEGRADATION

-.

-MKEl’ENE -

20 40 60 Bo loo 20 40 60 Bo loo UNLABELED LIPOPROTEIN (up protein /ml)

FE. 1. Ability of untreated human LDL (a), LDL modified with of protein), and the unlabeled lipoproteins at the protein concentra- diketene (A), and LDL from which diketene had been removed (0) to tions indicated. The experiment was performed at 37°C for 5 h. LDL compete with human lz51-LDL for binding and internalization (A) (1 mg) was allowed to react with 0.43 ~01 of diketene in 0.1 M borate and proteolytic degradation (B) in a normal human fibroblast. On buffer, pH 8.5, at room temperature for 5 min. The acetoacetyl group Day 7, the media were replaced by media containing 10% human was removed from the LDL by incubation with hydroxylamine at lipoprotein-deficient serum, 5 pg/ml of human ‘?-LDL (82 cpm/ng 35°C for 16 h.

15

12.5

0 0

20 40 60 80 100

UNLABELED LlPOPROTElN (ug protein / ml) FIG. 2. Ability of untreated human LDL (O), LDL modified with

0.43 jnnol (0) or 1.29 pmol (0) of diketene, and LDL from which the acetoacetyl groups on lysine had been removed (A) to compete with native human “‘1-LDL for binding at 4Y! in normal human fibroblasts The modification of LDL and the removal of the acetoacetyl group was performed as described in Fig. 1. On Day 7, 2.0 &ml of ‘*‘I-LDL (82 cpm/ng of protein) and the unlabeled LDL were added. Incubation was for 2 h at 4%

be essential for binding is speculative, however, since we have been unable to isolate an inactive subfraction of the modified LDL which has only the key 1 or 2 residues carbamylated. Carbamylation not only abolished the binding activity of LDL but also prevented canine HDL, from competitively displacing human %LDL at 37°C (data not shown).

Carbamylation did not alter the chemical composition of the LDL (data not shown) or their particle morphology as determined by negative staining electron microscopy (Fig. 4). However, the migration of the carbamylated LDL on paper electrophoresis was accelerated, reflecting the fact that there

Chemical Medication Df Human LDL

Diketene

Diketene Reversed

Carbamybtion

Reductive Methylation

FIG. 3. Paper electrophoretograms of control (untreated) and chemically modified human LDL. The modified LDL shown lacked binding activity and were treated as follows: diketene, 0.56 qol; diketene reversed, 0.56 pmol of diketene reversed with hydroxylamine; carbamylation, 60 min, and reductive methylation, 30 min.

was a neutralization of the positive charge of the modified lysyl residues (Fig. 3).

Reductive Methylation of Lysine-Modification of lysine residues by reductive methylation represented an important additional procedure with which to probe the role of lysine in the recognition site since, unlike the other modifications, this reaction did not alter the charge of the lysine or the net charge of the lipoproteins and yet was effective in blocking the binding activity (Fig. 7). Methylation of approximately 6 lysyl residues abolished LDL binding activity (Table I). The order of reactivity of various amino groups has been shown to depend on the type of reagent used (29). Thus, it might be predictable that the modification of different numbers of lysyl


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/


TABLE II Chemical composition of control, diketene-modified, and diketene-

modified-reversed LDL

Control” Diketene” Diketene reversed”

Total cholesterol 40.8 40.3 41.8 Phospholipid 25.5 26.0 26.5 Triglyceride 11.2 9.8 7.6 Protein 22.5 23.8 24.1

a Per cent composition based on duplicate determinations.

residues would be required for each reagent to be effective (i.e. 3, 4, and 6 modified residues were required to abolish binding activity by carbamylation, diketene treatment, and reductive methylation, respectively). Treatment of LDL with borohydride, a reagent which might potentially reduce disul- fide linkages, had no effect on the binding activity (Fig. 7). As with the acetoacetylation and carbamylation, reductive methylation totally abolished the binding activity of canine HDL, (Fig. 8).

One of the consequences of lipoprotein uptake mediated by the high affinity binding process is a stimulation of cellular cholesteryl ester formation (21). As shown in Fig. 9, reductive methylation of LDL, which prevented receptor-mediated uptake, also prevented the incorporation of [?Z]oleate into the cellular cholesteryl esters.

Reductive methylation of 1251-LDL likewise prevented binding, internalization, and degradation of the LDL in the direct assays performed at 37’C. Furthermore, essentially the same amounts of unmodified and modified ‘251-LDL were bound and internalized or degraded by fibroblasts from a patient

FIG. 4. Electron micrographs of negatively stained control and chemically modified human LDL. A, control (untreated) LDL, B, diketene modified (0.56 ~01); C, diketene modified (0.56 pmol) and reversed (hydroxylamine); D, control LDL, and E, carbamylated (60 mm). x 61,500.

30

= 25 6 \ P - 20 52

HDLc 8 HDLc -DIKETENE

UNLABELED LIPOPROTEIN (ug protein/ml)

FIG. 5. Ability of untreated canine HDL, (O), HDL, modified with diketene (0, and HDL, with diketene modification reversed (x) to compete with human ‘*‘I-LDL (45 cpm/ng of protein) for binding at 4’C. Other conditions were described in Fig. 2.

with homozygous type II hypercholesterolemia. By compari- son to values obtained in normal human fibroblasts (37°C) for LDL at a concentration of 15 pg/ml of protein, 7 to 10% of either the untreated or the modified ‘251-LDL was bound and


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/


I ’ I - I ’ I tB ’

250

200

150

100

50

600

400

200

I I I I I

20 40 60 80 100 20 40 60 80 100

UNLABELED LIPOPROTEIN (ug protein /ml)

240

60

FIG. 6. Changes in the ability of human LDL to compete with human lz51- LDL for binding and internalization (A) and proteolytic degradation (B) as a function of time of reaction with potassium cyanate (as described under “Ex- perimental Procedures”). On the day of the experiment, the media was replaced with media containing 10% human lipoprotein-deficient serum, 5 pg/ml of human “‘1-LDL protein (155 cpm/ng), and the untreated and treated lipoproteins at the concentrations indicated. 0, original untreated LDL, A, reaction time 4 min; n , reaction time 8 min; l , reaction time 15 min; V, reaction time 30 min.

1600

1400

1200

1000

800

600

400

200

I, I, I I I I II,1 1 I I I1 I, I

20 40 60 80 I00 20 40 60 80 100

UNLABELED LIPOPROTEIN (ug poteinhnl)

FIG. 7. Ability of untreated LDL (O), LDL modified with 1,2- media was replaced by media containing 10% human lipoprotein- cyclohexanedione (CHD) (x), LDL modified by reductive methyla- deficient serum, 5 pg/ml of human ““I-LDL (49 cpm/ng of protein), tion (RED/CHz) (01, LDL modified with N-acetylimidazole (NAn and the unlabeled lipoproteins at the protein concentration indicated. (O), and LDL treated with borohydride (LDL + BH,) (A) to compete The LDL were treated or modified as described under “Experimental with human lZ51-LDL for binding and internalization (A) and prote- Procedures.” olytic degradation (B) in normal human fibroblasts. On Day 7, the

A BINDING I 0 I ’ I ’ I

6 I ’ I ’ I ’ I

DEGRADATION .

400 Fc----- - . - REOICH3

1800

- 1500

300 - 1200

- 900

- 600

- 300

I, I, I,. 89

IO 20 30 40 50 IO 20 30 40 50

UNLABELED LIPOPROTEIN (ug poteirdml)

FIG. 8. Ability of native canine HDL, (0) and HDL, modified by reductive methylation (RED/CHd (0) to compete with human ?-LDL (119 cpm/ng of protein) for binding and internalization (A) and degradation (B). Reductive methylation was performed as described under “Experimental Procedures.” Other conditions were the same as described in Fig. 1.

internalized and 1 to 2% was degraded. Modification of the that lysine was the only amino acid residue modified by this

lysyl residues appeared to have little effect on the nonspecific method and that the major product formed was dimethylly-

processes. sine. It appeared that the NH?-terminal residue of the apo-B Reductive methylation to modify lysine has been exten- of LDL, which potentially could have been modified by re-

sively studied by Means and Feeney (24). Our data indicated ductive methylation, was unavailable to react. Amino acid


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

Modification of Lysine Residues of LDL and HDL,

I II T II II I I

LDL .

IO 20 30 40 50 60 70 80 90 100

LIPOPROTEIN (ug protein/ml)

FIG. 9. Incorporation of [Wloleate into cholesteryl esters as a function of concentration of either native LDL (0) or LDL modified by reductive methylation (RED/CH,) (0). Cell monolayers were preincubated for 15 h with the indicated amount of lipoprotein and then pulse labeled for 2 h with 20 ~1 of 10 mM [Wloleate (4,000 cpm/nmol). Cholesterol [‘%]ester content was determined after thin layer chromatographic isolation.

TABLE III

Amino acid composition of human LDL

Amino acids Control Reductive methylation”

Lysine 20.1h 2.0 Lysine-CHs 0 18.4‘ Histidine 6.3 6.3 Arginine 8.6 8.6 Aspartic acid 26.8 27.8 Threonine 16.0 16.5 Serine 19.3 19.4 Glutamic acid 31.3 31.2 Proline 10.3 9.7 Glycine 12.4 12.4 Alanine 15.5 15.4 Valine 13.4 13.2 Methionine 4.4 4.4 Isoleucine 14.7 14.6 Leucine 29.7 30.1 Tyrosine 8.4 7.6 Phenylalanine 12.6 12.3

a LDL were reductively methylated using a 60-min reaction sequence as described under “Experimental Procedures.”

’ Amino acid residues calculated per mol, assuming 250 residues/ mol.

’ Mixture of approximately 75% c-N,N-dimethyllysine and 25% l - N-methyllysine.

2 I 1 I 1 I , Ei ,%

A BlNMNG 1 B

, - DEGRADATDN

- 2000 7 600

i; s 500 3 400

i 300

200 k

B 100

VI I1 11 I I l v, 10 181 I I I E

5 IO 15 20 5 IO 15 20

LDL bg potein In-4

FIG. 10. Binding and internalization (A) and degradation (B) of tein-deficient serum and the indicated amounts of iodinated LDL (39 untreated human “‘I-LDL (O), alkylated ““I-LDL (x), and cpm/ng of protein). Normal LDL were first iodinated and then reduced/alkylated ‘?-LDL (0) in normal human fibroblasts. On Day alkylated or reduced and alkylated as described under “Experimental 7, the media was replaced by media containing 10% human lipopro- Procedures.”

analysis of LDL following exhaustive reductive methylation,

which modified 18 of the 20 lysine residues, did not reveal a modification of any residue other than lysine (Table III). Furthermore, unreactivity of the terminal amino group of the apo-E HDL, may be explained by the observation that this residue was blocked to Edman degradation.” The chemical composition, apoprotein pattern on sodium dodecyl sulfate- polyacrylamide gel electrophoresis, and lipoprotein particle morphology by negative staining electron microscopy were essentially identical for untreated LDL, LDL modified by reductive methylation, and LDL treated with the individual reagents required for the methylation procedure. Further- more, the net charge on the lipoprotein was not altered by the modification of 6 to 8 lysyl residues since the e-amino group of lysine retained its positive charge. Mobility was the same for methylated and for untreated LDL on paper (Fig. 3) or on Geon-Pevikon block electrophoresis.

lysyl groups, and even low concentration of reactants and brief exposure often resulted in the modification of 18 or more of the lysyl residues. As with the other procedures reported above, there was no apparent denaturation of the LDL and, although this reaction was not extensively investigated, it appeared to be a suitable modification technique.

Reduction and Alkylation of Cysteine-Alkylation with acetoacetamide or reduction (mercaptoethanol) and alkylation of human LDL had no effect on the ability of the LDL to compete with ‘251-LDL for binding, internalization, or degradation (data not shown). Furthermore, as assessed by the direct assays, these modifications did not alter the binding activity (Fig. 10). These data, as well as previous studies (5), indicated that the sulfhydryl groups were not involved in the receptor binding mechanism.

DISCUSSION N-Acetylimidazole Modification of Lysine-An additional

modification, which also abolished binding activity, was acetylation of lysine with N-acetylimidazole (Fig. 7) (25). This reaction was so rapid that it was difficult to modify only a few

’ K. H. Weisgraber, T. L. Innerarity, and R. W. Mahley, unpub- lished observation.

Previously, we have reported that modification of 40 to 50% of the arginyl residues of either the B or E apoproteins of human and canine LDL or canine HDL,, respectively, abolishes their ability to bind to the cell surface receptors of fibroblasts in culture (7). For human LDL, binding activity was prevented by the reaction of cyclohexanedione with 4


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

Modification of Lysine Residues of LDL and HDL,

arginine residues out of a total of 9 (assuming 250 amino acid residues/m01 of protein). It is now reported that modification of 15470, or less, of the total lysine residues of LDL (3 lysines out of 20, assuming 250 amino acids/m01 of protein) also prevents LDL binding, internalization, and degradation by fibroblasts. It appears that the interaction of LDL with the receptor requires the presence of both arginine and lysine within the recognition site. The chemical modification of either residue could prevent the binding of LDL to the receptor in one of several ways. 1) Modification may directly interfere with the positive charge or the availability of the positive charge because of the presence of the modifying group. Ionic interaction between the positively charged region of the LDL and the negatively charged receptor has been postulated (8). 2) The modification may directly interfere with the chemical reactivity, independent of charge, of the guanido or e-amino group of arginine or lysine, respectively. In the active sites of certain enzymes, lysine has been shown to form a Schiff base with certain substrates (32, 33) or participate in hydrogen bond formation (34). 3) The conformation of the recognition site may be altered by the modification in one of two ways: (a) alteration of the spatial distribution of the positively charged residues in the recognition site, i.e. the distance between or the arrangement of the key lysine and arginine residues, or ( b) alteration of the conformation of the recognition site preventing the reaction of as yet unidentified residues with the receptor. The arginine and lysine may be important residues, not because of a direct reactivity with the receptor, but because they stabilize the conformation of the recognition site; thus, modification of these groups could alter the stabilizing forces, e.g. hydrogen bonding within the recognition site. Similar alternative explanations have been con- sidered by others in the elucidation of the nature of active sites of enzymes (for review, see Refs. 35 and 36).

Several procedures for the modification of lysine residues were applied in an attempt to clarify the nature of the role played by these residues in lipoprotein binding to the cell surface receptors. Modification of lysine residues by acetoacetylation (diketene) or carbamylation (cyanate) results in neu- tralizing the positive charge of the e-amino group. This could directly prevent the ionic or chemical interaction between key lysine residues and the receptor. In addition, binding could be prevented through steric hindrance because of the bulk of the modifying group, or by alterations in the conformation of the recognition site. However, reductive methylation does not alter the charge on the lysine and steric hindrance by one or two methyl groups on the e-amino group is not thought to be a factor (24); nonetheless, receptor binding activity is abolished. As discussed by Means and Feeney (24), reductive methylation retains not only the total charge on the protein but also the spatial distribution of the charges. However, it remains to be shown that the spatial distribution of positive charges is retained on lipoproteins. If the charge distribution is unchanged and if steric hindrance is not a factor, then the lack of binding activity could be secondary to an alteration in the conformation of the recognition site or to the unavailabil- ity of the e-amino group for chemical reactivity with the receptor. The importance of the conformation of the protein is also suggested by the observation that the isolated (lipid- free) arginine-rich apoprotein does not possess binding activity despite the fact that it has been shown that this apoprotein is responsible for binding to the receptor (4). However, the importance of conformation remains to be proven, and we are presently attempting to determine if conformational changes do occur with chemical modification.

In addition to LDL and HDL,, certain positively charged molecules, such as histones, polylysines, and protamine, and

a positively charged protein, platelet factor 4, have been shown to inhibit competitively the binding of LDL to the LDL receptors on the surface of fibroblasts (37). However, other equally positively charged molecules (spermine, putrescine, and spermidine) and proteins (lysozyme and avidin) do not inhibit LDL binding. Although platelet factor 4 does bind to the receptor, the interaction is unlike that of LDL. Platelet factor 4 lacks specificity for the LDL receptor, binding almost equally well to fibroblasts from patients with homozygous familial hypercholesterolemia, which lack LDL receptors, as to fibroblasts from normal subjects. It has been suggested that the binding of platelet factor 4 may not be to the receptor but to adjacent sites on the cell, and displacement of LDL occurs through steric hindrance (37).

The present study, in agreement with the data cited above, indicates that the interaction of LDL and HDL, with the receptor sites is not determined simply by the presence of positive charges but depends on other highly specific properties of the recognition site of the B and E apoproteins. This is demonstrated by the results of the reductive methylation of LDL which maintains the positive charge on lysine but abolishes all binding activity. By contrast, less specific ionic interaction of LDL with heparin is not interferred with by reductive methylation, whereas neutralization of the charge on lysine by acetoacetylation or carbamylation totally prevents heparin/manganese precipitation,” as well as binding activity.

It is concluded from this study that arginine and lysine are both functionally significant residues involved in lipoprotein- receptor interaction, although their precise roles remain to be determined. Other key residues have not yet been identified, but it has been shown here that cysteine-cystine modification does not alter binding activity; furthermore, neutralization of the negatively charged carboxyl residues of glutamic and aspartic acids does not interfere with binding activity.4

Acknowledgments-We thank Ms. Barbara Kahler and Mrs. Wanda Le Bleu-Biswas for their excellent technical assistance and Mrs. Kathleen S. Holcombe for editorial assistance. We are indebted to Mrs. Barbara Torain and Dr. George Glenner at the National Institute of Arthritis, Metabolism and Digestive Diseases for provid- ing assistance with the amino acid analyses. We are appreciative to Miss Carolyn Groff and Mrs. Exa Murray for typing this manuscript.

REFERENCES

1. Brown, M. S., and Goldstein, J. L. (1976) Science 191, 150-154 2. Goldstein, J. L., and Brown, M. S. (1977) Annu. Reu. Biochem.

7.

8.

9.

46, 897-930 Mahley, R. W., and Innerarity, T. L. (1977) J. Biol. Chem. 252,

3980-3986 Innerarity, T. L., and Mahley, R. W. (1978) Biochemistry 17,

1440-1447 Innerarity, T. L., Mahley, R. W., Weisgraber, K. H., and Bersot,

T. P. (19781 J. Biol. Chem. 253.6289-6295 Mahley, R. W., Weisgraber, K. HI, Bersot, T. P., and Innerarity,

T. L. (1978) in High Density Lipoproteins and Atherosclerosis (Gotto, A. M., Jr., Miller, N. E., and Oliver, M. F., eds) pp. 149-176, Elsevier/North-Holland Press, Amsterdam

Mahley, R. W., Innerarity, T. L., Pitas, R. E., Weisgraber, K. H., Brown. J. H., and Gross. E. (19771 J. Biol. Chem. 252. 7279-7287

Goldstein, J. L., Basu, S. K., Brunschede, G. Y., and Brown, M. S. (1976) Cell 7,85-95

Mahley, R. W., and Innerarity, T. L. (1978) in Sixth International Symposium on Drugs Affecting Lipid Metabolism (Kritchev- sky, D., Paoletti, R., and Holmes, W. L., eds) pp. 99-127, Plenum Press, New York

’ T. L. Innerarity, K. H. Weisgraber, and R. W. Mahley, manuscript in preparation.

4 K. H. Weisgraber, T. L. Innerarity, and R. W. Mahley, unpub- lished data.


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/


10. Mahley, R. W., Innerarity, T. L., Weisgraber, K. H., and Fry, D. L. (1977) Am J. Pathol. 87,205-226

11. Mahley, R. W., Weisgraber, K. H., and Inneraritv, T. L. (1974) Circ. Res. 35, 722-733

12. Mahlev. R. W.. Weiseraber. K. H.. Inneraritv. T.. Brewer. H. B.. Jr., and Assmann, G. (1975) Biochemistry”i4,2817-2823

13. Mahley, R. W., and Weisgraber, K. H. (1974) Biochemistry 13, 1964-1969

14. AbeII, L. L., Levy, B. B., Brodie, B. B., and Kendall, F. E. (1952) J Biol. Chem. 195,357-366

15. Sperry, W. M., and Webb, M. (1950) J. Biol. Chem. 187,97-106 16. Fletcher, M. J. (1968) Clin. Chim. Acta 22, 393-397 17. Zilversmit, D. B., and Davis, D. K. (1950) J. Lab. Clin. Meal. 35,

155-160 18. Lowry, 0. H., Rosebrough, N. J., Farr, A. L., and Randall, R. J.

(1951) J. Biol. Chem. 193, 265-275 19. Bilheimer, D. W., Eisenberg, S., and Levy, R. I. (1972) Biochim.

Biophys. Acta 260,212-221 20. Goldstein, J. L., and Brown, M. S. (1974) J. Biol. Chem. 249,

5153-5162 21. Goldstein, J. L., Dana, S. E., and Brown, M. S. (1974) Proc. Natl.

Acad. Sci. U. S. A. 71,4288-4292 22. Marzotto, A., Pajetta, P., Gazigna, L., and Scoffone, E. (1968)

Biochim. Biophys. Acta 154,450-456 23. Chen, C. C., Grossberg, A. L., and Pressman, D. (1962) Biochem-

istry 1, 1025-1030

24. 25.

26.

27. Wofsy, L., and Singer, S. J. (1963) Biochemistry 2, 104-116 28. Stark, G. R., and Smyth, D. G. (1963) J. Biol. Chem. 238,214-226 29. Lindsay, D. G., and Shah, S. (1969) Biochem. J. 115, 587-595 30. Stark, G. R., Stein, W. H., and Moore, S. (1960) J. Biol. Chem.

31. 32.

33.

34.

35.

36.

37.

Means, G. E., and Feeney, R. E. (1968) Biochemistry 7,2192-2201 Roosdorp, N., Wan,, B., and Sjbholm, I. (1977) J. Biol. Chem.

252,3876-3880 Margolis, S., and Langdon, R. G. (1966) J. Biol. Chem. 241,

477-484

235, 3177-3181 Stark, G. R. (1965) Biochemistry 4, 1030-1036 Grazi, E., Meloche, H., Martinez, G., Wood, W. A., and Horecker,

B. L. (1963) Biochem. Biophys. Res. Commun. 10,4-10 Fischer, E. H. (1964) in Structure and Activity of Enzymes

(Goodwin, T. W., Harris, J. I., and Hartley, B. S., eds) pp. 111-120, Academic Press, London

Gorecki, M., and Shalitin, Y. (1967) Biochem. Biophys. Res. Commun. 29, 189-193

Singer, S. J. (1967) in Advances in Protein Chemistry (Anfinsen, C. B., Jr., EdsaIl, J. T., and Richards, F. M., eds) Vol. 22, pp. l-54, Academic Press, New York

Stark, G. R. (1970) in Advances in Protein Chemistry (Anfinsen, C. B., Jr., Edsall, J. T., and Richards, F. M., eds) Vol. 24, pp. 261-308, Academic Press, New York

Brown, M. S., Deuel, T. F., Basu, S. K., and Goldstein, J. L. (1978) J. Supramol. Struct. 8,223-234


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

K H Weisgraber, T L Innerarity and R W Mahleysurface receptors on human fibroblasts.

Role of lysine residues of plasma lipoproteins in high affinity binding to cell

1978, 253:9053-9062.J. Biol. Chem.

http://www.jbc.org/content/253/24/9053Access 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/253/24/9053.full.html#ref-list-1

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


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/content/253/24/9053

http://www.jbc.org/cgi/alerts?alertType=citedby&addAlert=cited_by&cited_by_criteria_resid=jbc;253/24/9053&saveAlert=no&return-type=article&return_url=http://www.jbc.org/content/253/24/9053

http://www.jbc.org/cgi/alerts?alertType=correction&addAlert=correction&correction_criteria_value=253/24/9053&saveAlert=no&return-type=article&return_url=http://www.jbc.org/content/253/24/9053

http://www.jbc.org/cgi/alerts/etoc

http://www.jbc.org/content/253/24/9053.full.html#ref-list-1

http://www.jbc.org/

vol. 253, no. 24, issue of december 25. pp. 90534062, 19% ... · role of the lysine residues of...

Documents