Monoclonal Antibodies against Bovine Milk Lipoprotein Lipase

14 downloads 88 Views 3MB Size Report
M. F. Rhode, D. P. Via, and L. C. Smith, unpublished experi- ments. assayed for ..... LaRosa, J. C., Levy, R. I., Herbert, P., Lux, S. E. & Fredrickson,. D. S. (1970) ...
Vol. 260, No. 2, Issue of January 25, p : 893-698,1985 rlnted m U.S. A.

THE JOURNAL OF BIOLOGICAL CHEMISTRY 0 1985 by The American Societyof Biological Chemists, Inc.

Monoclonal Antibodies against Bovine Milk Lipoprotein Lipase CHARACTERIZATION OF AN ANTIBODY SPECIFIC FOR THE APOLIPOPROTEIN C-I1 BINDING SITE* (Received for publication, December 1, 1983, and in revised form, June 1, 1984)

John C. VoytaS, David P. Via, Paavo K. J. Kinnunen, James T. Sparrow$, Antonio M. Gotto, Jr., and Louis C. Smith7 From the Departments of Biochemistry and Medicine, BaylorCollege of Medicine and The Methodist Hospital, Houston, Texas 77030

Ten murine monoclonal antibodies have been produced that are specific for bovine milk lipoprotein lipase. One monoclonal antibody, bLPL-mAb-7, inhibited completely the apolipoprotein C-I1 (apo-C-11)-dependent enzymic hydrolysis of trioleoylglycerol in a phospholipid-stabilized emulsion, but had noeffect on the hydrolysis of the water-soluble substrate p-nitrophenylacetate. Four times more bLPL-mAb-7 was required to achieve 50%inactivation of lipoprotein lipase activity when the enzyme was preincubated with excess apo-C-11. Disruption of the binding of a dansyllabeled apo-C-I1peptide to lipoprotein lipase by bLPLmAb-7 was demonstrated by resonance energy transfer, both in the presence and absence of lipid. This antibody thus appears to recognize the apo-C-I1 binding site of lipoprotein lipase. In addition, bLPL-mAb7 also inhibited the lipoprotein lipase activity of human post-heparin plasma.

Lipoprotein lipase catalyzes the hydrolysis of triacylglycerol and phospholipid in triacylglycerol-rich lipoproteins to form low density lipoproteins (1, 2). This process occurs mainly at the surface of the capillary endothelium (3)and requires apolipoprotein C-I1 for maximal rate of triglyceride hydrolysis (4,5). Endothelial cells apparently do not synthesize lipoprotein lipase but acquire the enzyme from other cells. Lipoprotein lipase is secreted by adipocytes (6),rat heart cells (7), cultured human monocytes (8, 9), a murine macrophage cell line (lo), and smooth muscle cells (11). Inculture, the binding of lipoprotein lipase to the surface of endothelial cells presumably involves heparan sulfate receptors (12, 13). Several distinct characteristics of lipoprotein lipase have been identified. These include hydrolysis of acylglycerols (14), direct interaction with apo-C-11’ both in the presence and *Support for this workwas provided by the Robert A. Welch Foundation Q-343, Human Health Services Grants HL-15648 and HL-27341, the Finnish State Medical Research Council, and the Office of Research and Development, EnvironmentalProtection Agency, under Grant R-80-8773. Additional funding was provided by Baylor College of Medicine. 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 accordance with 18 U.S.C. Section 1734 solelyto indicate this fact. $ Predoctoral Fellow of the Robert A. Welch Foundation. Present address: Dept. of Surgery, Childrens Hospital Medical Center, Boston, MA 02115. § Established Investigator of the American Heart Association. ll To whom reprint request should be addressed The Methodist Hospital, Department ofMedicine, MS A601, 6565 Fannin Street, Houston, T X 77030.

absence of lipid (15), andbinding of lipid with no requirement for apo-C-II(l6). Based on the currentunderstanding of other multifunctional proteins (17), it is reasonable to assume that each of the functional properties of lipoprotein lipase is associated with different structural domains of the enzyme: i.e. a catalytically active site, a substrate binding site, a lipid associating site, on apo-C-I1 binding site, anda heparan sulfate binding site. Previous immunological studies of bovine milk lipoprotein lipase have addressed questions of tissue and species crossreactivity with polyclonal antibodies (18,19). Recently, Shirai et al. (20) and Olivecrona and Bengtsson (21) have described Fab fragments of polyclonal antibodies that inhibit triglyceride hydrolysis by bovine milk lipoprotein lipase. Monoclonal antibody methodology (22) provides a different and more informative immunological approach that has the potential to characterize distinct regions of lipoprotein lipase. This report describes the production and characterization of 10 monoclonal antibodies directed against bovine milk lipoprotein lipase and the identification of 1 monoclonal antibody that inhibits apo-C-I1enhancement of triglyceride hydrolysis. MATERIALS ANDMETHODS

Preparation of Monoclonal Antibodies-Male BALB/c mice, 6-8 weeks old, were immunized intraperitoneally with 50 pg of purified bovine milk lipoprotein lipase (23) which had been emulsified with Freund’s complete adjuvant. Four weeks later, the mice were boosted intraperitoneally with a saline solution-of 50 pg ofjipoprotein lipase. On the fifth day after theboost, the mice were killed to prepare single cell suspensions of splenocytes. Spleen cells and hypoxanthine/aminopterin/thymidine-sensitivemyeloma cells were mixed at a ratio of 81 and centrifuged. After washing to remove serum components, 1 ml of 50% polyethylene glycol was added to thecell pellet to achieve fusion (24). Spleen cells from mice 1 and 2 were fused with the myeloma line Sl94/5.XXO.BU.l (25) and mouse 3 with P3X63AG8.653 (26). Both myeloma lines are non-producers of immunoglobin chains. Hybrids were screened for the secretion of lipoprotein lipase specific antibody by ELISA (27) using horseradish peroxidase-conjugated anti-mouse IgG and IgM (Cappell). Positive wells were cloned twice by limiting dilution. High titer ascites fluids were prepared by intraperitoneal injection of lo7hybridoma cells into individual pristane-primed mice. The purity of the material used for immunization and for screening is shown in Fig. 1. The material used for immunization showed no detectable contaminants (go%) lipoprotein lipase was used for screening procedures. Purification of Antibodies-Ascites fluid (7-14 ml) was applied to a 2.7 X 200-cm column of Sephacryl S-200 and eluted at a flow rate of 0.5 ml min” with phosphate-buffered saline, pH 7.8, containing



The abbreviations used are: apo-C-11, apolipoprotein C-11; bLPLmAb, monoclonal antibodies against bovine milk lipoprotein lipase; apo-C-II-DNS(70-78), a synthetic peptide containing residues 60 to 78 ofapo-C-I1with a single dansyl fluorophore on the amino-terminal residue; ELISA, enzyme-linked immunoassay.

893

Lipoprotein Antibodies Monoclonal Lipase

894

A

B

O93K

31 K

FIG. 1. Bovine milk lipoprotein lipase, 15 wg, was electrophoresed on 10%polyacrylamide SDS gels. Gels were fixed and then stained with Coomassie Blue for protein detection. A, bovine milk lipoprotein lipase used for monoclonal antibody immunization. R, bovine milk lipoprotein lipase usedfor monoclonal antibody screening. The arrow indicates the position of the lipoprotein lipase band. Molecularweight standards were phosphorylase b (93,000), bovine serum albumin (66,000), ovalbumin (45,000), and carbonic anhydrase (31,000). 0.01% NaN3. Total protein concentrations of ascites fluids varied between 10 and 45 mg ml-', with immunoglobulin representing 4080% of the protein recovered from the column. Aliquots of column fractions were examined by electrophoresis on 7.5% SDS gels under nonreducing conditions (28). The IgG peak fractions werepooled, sterilized by filtration, and frozen in 3-ml aliquots. The IgM peak fractions were precipitated with 50% ammonium sulfate, dialyzed against phosphate-buffered saline, and frozen. In all cases, a 1.0 mg ml" solution of immunoglobin was assumed to have an A m of 1.4 (29). The concentrations of immunoglobin in the ascites fluids were calculated from the S-200 elution profiles. Fab fragments were generated by digestion of bLPL-mAB-7 with papain for 2 h at 37 "C a t a 50:l (w/w) enzyme/antibody ratioin the presence of0.01 M 2mercaptoethanol (30). After alkylation with 0.014 M iodoacetamide and overnight dialysis, fragments were purified by Sephacryl S-200 chromatography. Subtypingof Monoclonal Antibodies-Purified monoclonal antibodies were screened for IgC,, IgG2., IgG2b, and IgG3 heavy chain specificity by ELISA. Falcon Microtest I1 plates were coated with bovine milk lipoprotein lipase by treating the plates with 50 pllwell of a 10 pgml-I solution of the enzyme for 1 h a t room temperature. The coated plates were incubated with serial dilutions of purified monoclonal antibody for 1h in phosphate-buffered saline containing 0.25% bovine serum albumin and then with a mixture of rabbit anti-mouse IgG subtype specific antibody (1:lOOO) (Miles) and horseradish peroxidase-conjugated goat anti-rabbit IgG (1:lOOO) (Cappell) for 2 h. After washing, substrate consisting of0.02% 2,2-azino-di-(3-ethylbenzthiazoline sulfonic acid) and 0.02% H202 in0.1 M citrate, pH 4.0, was added to each well. After 30 min, the absorbance at 414 nm was determined with a Titertek Multiscan (Flow). Western Blot Analysis of Monoclonal Antibodies-Lipoprotein lipase, 5 pgllane, was electrophoresed on 10%polyacrylamide slab gels

andthen electrophoretically transferred to nitrocellulose paper (Schleicher and Schuell) a t 300 mA for 14 h at 4 "C in a Bio-Rad TransBlot apparatus. Approximately 80% of the lipase was transferred with a buffer containing 25 mM Tris, pH 8.0, 192 mM glycine, and 20% (v/v) methanol. After blocking nonspecific binding sites with bovine serum albumin, each lane was incubated for 12 h at room temperature with purified monoclonals (1.5-0.100 mg m1-l) or ascites fluid (undiluted to 1:lO dilution). Strips were exhaustively washed and thenincubated with a 1:lOOO dilution of horseradish peroxidaseconjugated anti-mouse IgG or anti-mouse IgM for 24 h at 4 "C. After extensive washing, the strips were developed with 0.0025% o-diansidine in 0.01 M Tris, 0.01% H202,pH 7.4, for 1h a t room temperature. Determination of RelatiueAssociation Constants of the bLPLd b s - T h e relative titers of the bLPL-mAbs for lipoprotein lipase were determined using ELISA. For these assays, Falcon Microtest I1 plates were coated with bovine milk lipoprotein lipase, 50 pl of 10 pg ml" in PBS, pH 7.2, for 1 h a t room temperture. Approximately 20 ng of lipase is bound to each well as determined using '251-lipoprotein lipase. The plates were incubated with 1:5 serial dilutions of the purified bLPL-mAbs and subsequently incubated sequentially with horseradish peroxidase-conjugated anti-mouse immunoglobin (Cappell) and substrate solution as described above. The reciprocal of the concentration of bLPL-mAb which gave50% of the maximal response was used as the apparent K, value. Assays for Lipoprotein Lipase Activity-Liprotein lipase activity was determined with a phospholipid-stabilized emulsion of trioleoylglycerol as described previously (15). Release of [2-3H]glycerolfrom tri01eoyl[2-~H]glycerol was quantified. Other experimental details are described in the figure legends. Hydrolysis of p-nitrophenylacetate was used as an index of apo-C-I1 independent enzyme activity (31). The lipoprotein lipase-dependent increase in absorbance a t 405 nm of p-nitrophenol at a substrate concentration of 0.5 pmol ml" in 20 mM Tris, pH8.5, was measured on a Cary 15 spectrophotometer. The specificity of the monoclonal antibodies for the heparin binding site of lipoprotein lipase was tested asfollows. Lipoprotein lipase, 0.7 ng, was preincubated with individual bLPL-mAbs, 0.1 pg, for 15 min a t 0 "C in 0.5 ml of 0.1 M NaCI, 0.166 M Tris, pH 8.4, containing 300 p~ bovine serum albumin. Heparin-agarose (Pierce), 100 pl of a 50% suspension (v/v), was then added to each tube. The samples were shaken on ice for an additional 15 min. The tubes were then centrifuged a t 200 rpm to pellet the agarose and a 0.4-ml aliquot of the buffer was removedand added to a substrate:apo-C-I1 mixture to assay enzyme activity as previously described (15). Others Methods-Lipoprotein lipase was purified from bovine milk as previously described (23). Apo-C-I1 waspurified by the method of Jackson et al. (32) from apo-C peptides kindly supplied by Dr. Karl Weisgraber, Gladstone Foundation, San Francisco, CA. The synthesis of apo-C-11-DNS(60-78) and measurement of the association of this peptide with lipoprotein lipase has been described (15). Protein concentrations were determined by the method of Lowry et al. (33). Human post-heparin plasma was obtained by intravenous injection of 10 units kg-' of heparin into volunteers and by collection of blood after 10 min (34). RESULTS

The origin, the antibody concentration in ascitesfluid, and the subtype specificity of the 10 monoclonal antibodies to bovine milk lipoprotein lipase are summarized in Table I. The concentrations of immunoglobin in the murine ascites fluid ranged from 3.8 mg ml" to 28.6 mg ml-I. Four of the bLPLmAbs were identified as IgM molecules. Secretion of specific IgM by hybridomas was shown by the characteristic low levels of IgG by S-200 chromatography of these ascites fluids, by SDS-gel patterns andby the positive immunoreactivity of the IgM fractions, but not in IgG fractions, toward lipoprotein lipase. Each of the IgG monoclonal antibodies consisted of a single IgG heavy chain subtype suggesting a monoclonal origin. All ofthe mAbs showed immunoreactivity toward purified bovine milk lipoprotein lipase by ELISA. Most of the association constants ranged from lo6 M" to 7.7 X 10' "I. All of the monoclonal antibodies were capable of removing enzyme activity from solution using indirect immune precipitation.2 D. P. Via, unpublished observations.

895

Antibodies Monoclonal Lipase Lipoprotein TABLE I Characteristics of the bLPL-mAbs b;g

Immunoglobin

Myeloma line

Fusion

*gi2Solid phase association

concentrationb mg rnl"

1 3 5 7 8 9 10 11 12 13

1 1 1

constants M"

17.8 3.2 X 7.7 3.3 x 6.8 2.2 X 9.5 7.7 x 3.9 2.9 X 12.3 4 . 5 X 28.6