HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

J. Biol. Chem., Vol. 277, Issue 6, 4334-4342, February 8, 2002

This Article Abstract Full Text (PDF) All Versions of this Article: 277/6/4334    most recent M105421200v1 Submit a Letter to Editor Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Shen, Y. Articles by Olivecrona, G. Search for Related Content PubMed PubMed Citation Articles by Shen, Y. Articles by Olivecrona, G. Social Bookmarking                      What's this?

Functional Analyses of Human Apolipoprotein CII by Site-directed Mutagenesis

IDENTIFICATION OF RESIDUES IMPORTANT FOR ACTIVATION OF LIPOPROTEIN LIPASE^*

Yan Shen^§, Aivar Lookene^¶, Solveig Nilsson, and Gunilla Olivecrona

From the  Department of Medical Biosciences, Umeå University, Umeå SE-90 187, Sweden and ^¶ National Institute of Chemical Physics and Biophysics, Tallinn, Estonia

Received for publication, June 12, 2001, and in revised form, November 13, 2001

	ABSTRACT

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Apolipoprotein CII (apoCII) activates lipoprotein lipase (LPL). Seven residues, located on one face of a model -helix spanning residues 59-75, are fully conserved in apoCII from ten different animal species. We have mutated these residues one by one. Substitution of Ala⁵⁹ by glycine, or Thr⁶² and Gly⁶⁵ by alanine did not change the activation, indicating that these residues are outside the LPL-binding site. Replacement of Tyr⁶³, Ile⁶⁶, Asp⁶⁹, or Gln⁷⁰ by alanine lowered the affinity for LPL and the catalytic activity of the LPL·apoCII complex. For each residue several additional replacements were made. Most mutants retained some activating ability, but replacement of Tyr⁶³ by phenylalanine or tryptophan and Gln⁷⁰ by glutamate caused almost complete loss of activity. All mutants bound to liposomes with similar affinity as wild-type apoCII, and they also bound with similar affinity to LPL in the absence of hydrolyzable lipids. However, the inactive mutants did not compete with wild-type apoCII in the activation assay. Therefore, we conclude that the productive apoCII·LPL interaction may be dependent on substrate molecules. In summary, our data demonstrate that residues 63, 66, 69, and 70 are of special importance for the function of apoCII, but no single amino acid residue is absolutely crucial.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Human apolipoprotein CII (apoCII)¹ is a 79-residue protein synthesized in the intestine and liver and secreted as a surface component of chylomicrons, very low density lipoproteins, and high density lipoproteins (1-3). ApoCII plays an important role in plasma lipid metabolism as an activator of lipoprotein lipase (LPL) (1, 4), clearly illustrated by the severe hypertriglyceridemia that is associated with genetic deficiency of apoCII (5). A number of natural mutations affecting the apoCII gene have been identified and characterized (6). Most mutations reported so far result in either very low levels of plasma apoCII or major structural changes in the protein, e.g. prematurely truncated forms. A Trp²⁶  Arg mutation was recently reported to lead to apoCII deficiency, probably because of a disturbed lipid interaction of the activator (7). This is the only example of a single point mutation in apoCII leading to loss of function, but the interpretation is not straightforward because the protein was not detectable in plasma.

The structure of full-length human apoCII in complex with SDS micelles was recently reported based on studies by NMR (8). The protein contains three regions with helical conformation spanning residues 16-36, 50-56, and 63-77. The lipase-activating region of apoCII has previously been localized to the C-terminal one-third of the sequence, from about residue 56, whereas the N-terminal two-thirds of the sequence is involved in lipid binding (9-11). In most studies it has been found that apoCII causes an increase in the catalytic efficiency of LPL and not an enhanced lipid affinity of the enzyme (10, 12). Despite a long history of studies on the effects of apoCII, the detailed mechanism for the activation remains unknown. It has been proposed that apoCII binds to LPL and induces conformational changes in the lipase or changes its orientation at the lipid-water interface. As a result of these changes, LPL may become more efficient in binding of individual substrate molecules or in release of lipolysis products (fatty acids) (13). It has also been proposed that apoCII, as an amphipathic molecule, may influence the structural organization of lipids such that substrate molecules become more available to the active site of LPL (8, 12). However, synthetic peptides corresponding to the C-terminal part, which does not bind to lipid, are able to activate LPL in vitro (14). The strongest evidence for activation of LPL via formation of a complex with apoCII come from data showing that LPL and apoCII bind to each other even in the absence of a lipid/water interface (15).

In the present study we have focused on the region comprising residues 56-79 in human apoCII. Compared with the corresponding region in apoCII from nine other animal species, there are only seven fully conserved residues: Ala⁵⁹, Thr⁶², Tyr⁶³, Gly⁶⁵, Ile⁶⁶, Asp⁶⁹, and Gln⁷⁰ (16). In the rest of the apoCII molecule there are only two other fully conserved residues, Leu¹⁵ and Ala³³, but both are outside the activating region. We have systematically evaluated the functional role of each of the seven conserved residues by using site-directed mutagenesis followed by analyses of the secondary structure by circular dichroism (CD) measurements and analyses of activating ability. We have also directly studied the interaction of the mutants with LPL using energy transfer measurements. The results suggest that the LPL-binding site in apoCII consists of mainly 4 residues located on one face of the third -helix located closest to the C terminus.

	EXPERIMENTAL PROCEDURES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Materials-- Restriction endonucleases were from Amersham Biosciences, Inc. T4 ligase was from Roche Molecular Biochemicals. Kanamycin was from Duchefa, Haarle, The Netherlands. Isopropyl--D-thiogalactoside (IPTG), dansyl chloride, and egg yolk phosphatidylcholine were from Sigma Chemical Co., St. Louis, MO. Low melting agarose and polyvinylidene difluoride filters (PVDF, 0.2 µm) were from Bio-Rad, Hercules, CA. JET-sorb Gel extraction kit was from Genomed. The expression vector pET-29a, S-protein-agarose, and BL21(DE3) competent cells were from Novagen, Inc., Madison, WI. Epicurian Coli BL21-codon Plus(DE3)-RIL competent Cells and the QuikChange site-directed mutagenesis kit were from Stratagene, La Jolla, CA. Penta His-tag antibody and Ni²⁺-NTA-agarose were purchased from Qiagen GmbH, Hilden, Germany. Factor Xa from bovine plasma was from New England BioLabs, Beverly, MA. Serva Blue G was from Serva, GmbH, Heidelberg, Germany. Transfer Membrane was from Pall Corp., Ann Arbor, MI. Dialysis tubes were from spectrum Medical Industries, Houston, TX. Low molecular weight marker was from Amersham Biosciences, Inc., Buckinghamshire, UK. For the analyses of LPL activity an emulsion with a composition corresponding to Intralipid 10% (soy bean triacylglycerols emulsified in egg yolk phospholipids) but containing in addition a trace amount of [³H]oleic acid-labeled trioleoylglycerol (kindly prepared by KABI-Fresenius, Uppsala, Sweden) was used. A BCA protein assay kit was from Pierce, Rockford, IL. LPL was purified from bovine milk as described previously (17). The stock solutions contained about 0.5 mg of LPL/ml 10 mM Bis-tris, pH 6.5, 1 M NaCl. 1,2-Di-O-hexadecyl-sn-glycero-3-phosphatidylcholine was from Larodan Fine Chemicals AB, Malmö, Sweden. Oligonucleotides were synthesized by Amersham Biosciences, Inc., Uppsala, Sweden. An antiserum against human apoCII, purified from plasma and coupled as a hapten to keyhole limpet hemocyanin, was produced in a rabbit by a standard immunization procedure (18). Computer-assisted analyses were accomplished with the programs Lasergene DNA Star Inc., Madison, WI and Wisconsin sequence analysis package, Genetics Computer Group (GCG). For secondary structure prediction of the mutants the programs at the NPS@(Network Protein Sequence Analysis) web sites were used (available at pbil.ibcp.fr/). Binding and activation curves were fitted using the program Fig. P (BIOSOFT, Cambridge, UK).

Plasmid Constructions-- Two plasmids were used for expression of recombinant human apoCII: pET29a-hapoCII and pET-histag-hapoCII.

cDNA for human apolipoprotein CII in the pUC18 vector (hapoCII, a kind gift from Dr. Steven Humphries, London) was connected to an EcoRI linker coding for the cleavage site for factor Xa (coding sequence for factor Xa cleavage: Ile-Glu-Gly-Arg-) attached to the first codon of the mature apoCII cDNA sequence. The whole DNA fragment was cloned into the EcoRI/XhoI site of the vector pET29a(+) to form the expression plasmid pET29a-hapoCII. The recombinant was verified by EcoRI/XhoI cleavage and DNA sequencing. The plasmid was transformed into E. coli BL21(DE3) to express the fusion protein S-tag-"factor Xa cleavage site"-hapoCII (117 amino acid residues in total). For production of pET-histag-hapoCII, the pET29a-hapoCII plasmid was cleaved with NdeI and EcoRI to delete the sequence encoding the S-tag and the thrombin cleavage site. The complementary oligonucleotides 5'-TATGCACCATCATCATCATCATG-3' and 5'-AATTCATGATGATGATGATGGTGCA-3', coding for an N-terminal 6xHis-tag, were ligated to the cleaved pET29a vector. The new construct resulted in a vector with the T7 RNA polymerase promoter and the lac operator followed by sequences coding for 6xHis-tag, factor Xa cleavage site, and human apoCII.

Site-directed Mutagenesis of Human ApoCII-- Mutant 63A was generated by overlap extension by using the polymerase chain reaction (PCR) and four different primers (Table I). All other mutants were generated using QuikChange site-directed mutagenesis kit. The basic procedure for generation of each mutant utilizes a template vector with apoCII cDNA (pET29a-hapoCII or pET-histag-hapoCII) and two synthetic oligonucleotide primers containing the desired mutation (Table I). The oligonucleotide primers, each complementary to opposite strands of the vector, were extended by Pfu Turbo DNA polymerase. All mutants were sequenced using an ABI 337 DNA sequencer and the Dye Terminator Cycle Sequencing Ready Reaction kit (PerkinElmer Life Sciences).

Expression and Purification of Wild-type and Mutated Human ApoCII Proteins-- pET29-hapoCII and its mutants (A59G, T62A, Y63A, G65A, I66A, D69A, and Q70A) were introduced into BL21(DE3)-competent cells. Expression, purification, and cleavage of the fusion proteins by factor Xa were made as described,² but in LB medium.

pET-histag-hapoCII ant its mutants (Y63A, Y63T, Y63L, Y63F, Y63W, I66A, I66L, I66V, I66F, D69A, D69S, D69N, D69E, Q70A, Q70A, Q70N, Q70E, L72A, S73A, and V74A) were transformed into Epicurian Coli BL21-Codon Plus(DE3)-RIL competent cell. The bacteria were grown at 37 °C in Terrific Broth containing kanamycin (50 µg/ml) to an A_600 nm of 1.0. At this point the temperature was lowered to 16 °C, and 1 h later expression of human apoCII and its variants was induced with 0.5 mM IPTG. On the next day, the bacteria were harvested by centrifugation and purified by Ni²⁺-NTA-agarose as described previously (19). Briefly, buffers containing 8 M urea, 0.1 M NaH₂PO₄, 0.01 M Tris-Cl were used to produce a stepwise gradient from pH 8.0 to 4.5. The fusion proteins that eluted at pH 7.2 were dialyzed for 24 h at 4 °C against ammonium bicarbonate and were then lyophilized.

High Performance Liquid Chromatography of Wild-type and Mutated His-tagged Human ApoCII Proteins-- High performance ion exchange was performed on an ÄKTA purifier 900 (Amersham Biosciences, Inc.) using a Mini Q PE 4.6/50 column. The fusion proteins eluted between pH 6.8 and pH 4.5 from Ni²⁺-NTA-agarose were dialyzed against 6 M urea, 0.01 M Tris, pH 8.2, and were filtered through a 0.2-µm filter (Gelman Sciences, Ann Arbor, MI) before they were applied to the column at a pressure of 5 MPa. Proteins were then eluted by a linear gradient of Tris-Cl (pH 8.2) from 0.01 to 0.15 M. Remaining proteins were washed out by 1 M NaCl and/or 6 M guanidinium chloride (GdmCl). Protein concentrations in all the fractions were measured by the BCA assay. Fractions were combined, and the proteins were analyzed by electrophoresis on SDS-Tricine gels and by N-terminal amino acid sequence analyses as previously described (16). Two forms of apoCII were detected by matrix-assisted laser desorption/ionization time-of-flight mass spectrometry on a homemade instrument at the National Institute of Chemical Physics and Biophysics, Tallinn, Estonia.

Proteins Analyzed by Tricine-SDS-PAGE and Western Blotting-- The expressed proteins were separated on Tricine-SDS-PAGE and were stained by 0.025% Serva-blue G in 10% acetic acid (19). The separated proteins were transferred onto a nitrocellulose filter. The filter was put in TBST (10 mM Tris-Cl, pH 8.0, 0.5 M NaCl, and 0.05% Tween 20) containing 5% bovine serum albumin and 3% gelatin for 30 min at 42 °C. Human apoCII and mutants were detected by incubation with the rabbit anti-human apoCII serum (dilution 1:100). After incubation overnight the filter was washed 3 × 10 min in TBST. Then the filter was incubated with goat anti-rabbit IgG (conjugated to alkaline phosphatase) diluted 1:3000 in TBST. Finally the filter was washed 3 × 10 min as before. Bound antibodies were detected by addition of substrate to alkaline phosphatase, and the reaction was stopped by rinsing the filter in H₂O. For blotting with the penta-His-tag antibody, the nitrocellulose filter was handled as recommended by the manufacturer.

Circular Dichroism-- Possible global changes of the secondary structure of some of the mutated apoCIIs were investigated by measurements of far-UV spectra at 25 °C on a CD6 spectrodichrograph (Jobin-Yvon Instruments SA, Longjumeau, France) in a 0.5-mm path-length cell. For this the protein concentrations were determined by absorbance at 280 nm.

Lipase Assays-- For measurement of LPL activity we used an assay mixture of 200 µl containing 2 mg of triacylglycerols from the lipid emulsion. The medium contained 0.1 M NaCl, 0.1 M Tris-Cl, 20 µg of heparin, and 12 mg of bovine serum albumin. The pH was 8.5. To study the activation, wild-type apoCII and its mutants were dissolved in 5 M urea, 10 mM Tris-Cl (pH 8.2). The protein concentration was in each case determined by the BCA protein assay. Five µl of the stock solution of apoCII, or the same volume of dilutions in 5 M urea, 10 mM Tris (pH 8.2), was added to the incubations. The reactions were started by addition of LPL and were stopped after incubation for 15 min at 25 °C by addition of organic solvents for extraction of the labeled free fatty acids (19). The lipase activity is expressed in units/mg of LPL, where 1 unit corresponds to release of 1 µmol of fatty acid per min. For studies of competition between the wild-type apoCII and mutants for activation of LPL, the experiments were carried out as described, except for the presence of 2.5 µM Q70E, Y63F, or Y63W in the incubation mixture together with wild-type apoCII.

Determination of Kinetic Parameters from Activation Curves-- Quinn et al. (20) have shown that the activation of LPL by apoCII can be described by a random mechanism. Based on this mechanism, the velocity of the reaction (v) at substrate saturation is expressed as,

(Eq. 1)

where L corresponds to LPL and C corresponds to apoCII.

This equation was used for determination of K_d and . K_d characterizes the apparent affinity of apoCII for LPL, and  indicates how much more active the LPL·apoCII complex was compared with LPL alone (-fold activation). When  is equal to 1, apoCII does not activate LPL. The kinetic constants were determined by using the program Fig. P.

Preparation of Liposomes-- Two different phospholipids were used in this study. One was 1,2-di-O-hexadecyl-sn-glycero-3-phosphatidylcholine, which cannot be hydrolyzed by LPL. This ether lipid was used for studies of the apoCII·LPL interaction. The other was egg yolk phosphatidylcholine, used for studies of the interaction between apoCII and lipids. For preparation of liposomes the lipids were first dissolved in chloroform and dried under nitrogen. The dry lipid was then dispersed in 5 mM Tris-Cl, 0.15 M NaCl, 0.02% NaN₃, pH 7.4, and sonicated without break in ice-cold buffer by an MSE Soniprep 150 equipped with a 5-mm probe. Egg yolk phosphatidylcholine was sonicated for 30 min and 1,2-di-O-hexadecyl-sn-glycero-3-phosphatidylcholine was sonicated for 1 h. The liposomes were kept for 2-3 h at 37 °C before they were centrifuged for 10 min at 3000 rpm at 4 °C. The supernatant was collected and dialyzed against 0.1 M Tris-Cl (pH 8.1). Any possible large aggregates were removed by filtration through a 0.2-µm filter. The concentrations of phospholipids in the final preparations were measured using an enzymatic kit from Wako Chemical GmbH, Germany.

Determination of Equilibrium Dissociation Constants by Dansylation of ApoCII-- The wild-type apoCII and the mutants (His-tagged, not cleaved) were dansylated according to a procedure described in a previous study (19). The relative increase of fluorescence at 490 nm upon excitation at 280 nm was measured after each addition of bovine LPL to the solution of dansylated apoCII in 0.1 M Tris-Cl (pH 8.1) with 10 µl of 1,2-di-O-hexadecyl-sn-glycero-3-phosphatidylcholine liposomes (final concentration 10 µg/ml). For the measurements, a Spex FluoroMax-2 fluorometer or a Shimadzu spectrofluorophotometer (model RF500) was used. The fluorescence background of apoCII was subtracted from the experimental values. Binding of wild-type apoCII to LPL was also studied in the absence of liposomes.

Binding of ApoCII to Liposomes-- Binding of apoCII to liposomes was measured by using the change of fluorescence of the single tryptophan at residue 26 as described in our previous study (19). The increase of fluorescence was attributed to binding of apoCII to the liposomes.

	RESULTS

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Expression and Purification of ApoCII and Mutants-- Comparison of the amino acid sequences of apoCII from the 10 different animal species shows that there are seven fully conserved residues in the C-terminal region (Fig. 1A). All were located on the same side of a model -helix spanning residues 59-75 (Fig. 1B). To study the role of the conserved residues for the function of apoCII we first made substitutions with alanine using the codons GCC, GCA, and GCT. Ala⁵⁹ was instead substituted with glycine. The seven mutants and the wild-type apoCII were expressed using the pET29a-hapoCII vector, generating S-tagged apoCII proteins that were purified on S-protein-agarose and cleaved by factor Xa (Fig. 2).

View larger version (22K): [in this window] [in a new window]   Fig. 1.   A, schematic view of functional domains of human apoCII. The seven fully conserved residues in the activating region (16) are shaded and numbered. B, helical wheel model of residues 59-75 in human apoCII. The seven fully conserved residues are shaded.

View larger version (31K): [in this window] [in a new window]   Fig. 2.   Analysis of S-tagged apoCII on Tricine-SDS-PAGE. The wild-type (WT) and mutant apoCII proteins were purified on S-protein-agarose and repurified by passage over the same column after cleavage by factor Xa. A, a gel stained by Serva Blue G; B, a Western blot of the same fractions as in A using rabbit anti-human apoCII serum.

To identify critical chemical groups or side chains, more mutants were generated at residues 63, 66, 69, and 70. These mutants were expressed using the pET-histag-hapoCII vector and purified on Ni²⁺-NTA-agarose. When the pH of the elution buffer was decreased to 6.8 or lower, a protein with an apparent molecular mass of 26 kDa was eluted in addition to apoCII (Fig. 3). For separation of this protein from apoCII, ion-exchange chromatography on a Mini Q column was used (Fig. 4). The contaminating protein bound stronger than apoCII to the column and was eluted by 1 M NaCl and/or 6 M GdmCl (Fig. 4, inset). N-terminal sequence analyses of the contaminant gave the sequence KVAKDLVVS, which was found to correspond to an Escherichia coli protein consisting of a domain homologous to FK506-binding proteins (21). This protein is rich in potentially metal-binding amino acids, such as histidine, cysteine, and acidic amino acids, explaining why it bound very strongly to the Ni²⁺-NTA-agarose. ApoCII was separated into two forms by the ion-exchange column (Fig. 4, inset). N-terminal amino acid sequence analysis showed that they both had the same sequence, MHHHHH. Mass spectrometry revealed that there was a 260-Da difference between the two forms. This corresponded to a lack of the last 2 glutamate residues at the C-terminal end (the theoretical difference is 258.2 Da). The expressed wild-type apoCII contained more than 75% of the short form. Because the two forms of apoCII showed similar ability to activate LPL (data not shown), we used a mixture of the two forms of each expressed mutant for the following experiments. In another set of experiments we compared in detail the ability of His-tagged apoCII with apoCII purified from human plasma with regard to activation of LPL. From these data we know that the His-tag at the N terminus does not interfere with the activation.³ Therefore, we decided to use the His-tagged recombinant proteins without further cleavage for the present functional studies.

View larger version (44K): [in this window] [in a new window]   Fig. 3.   Analysis of His-tagged wild-type apoCII on Tricine-SDS-PAGE. His-tagged, wild-type apoCII was eluted from the Ni2+-NTA-agarose column by a pH gradient from pH 8.0 to 4.5 in 8 M urea, 0.1 M Tris, and 0.01 M NaH2PO4. A, a gel stained with Serva blue G; B, Western blot of the same fractions as in A using the penta-His tag antibody.

View larger version (33K): [in this window] [in a new window]   Fig. 4.   Purification of His-tagged wild-type apoCII by ion-exchange chromatography and analysis on Tricine-SDS-PAGE. The fraction eluted from the Ni2+-NTA-agarose column at pH 6.8 (see Fig. 3) was applied on a Mini-Q column. The chromatogram shows the separation of the two forms of apoCII. Inset, the analysis of the fractions on Tricine-SDS-PAGE. Lanes: Mw, low molecular weight markers; a.s., applied sample; b.t., break through (material that was not bound to Mini-Q); fractions that were run on the gel are shown by numbers; concentrations of NaCl and GdmCl used for elution were 1 and 6 M, respectively.

Activation of LPL by ApoCII Mutants-- Substitution of Ala⁵⁹ by glycine and residues Thr⁶² and Gly⁶⁵ by alanine did not have much effect on the activating ability of apoCII (Fig. 5). In contrast, the mutants with alanine substitutions at position 63, 66, 69, and 70 showed a defective function (Fig. 5). For quantitative comparisons, the kinetic parameters K_d and  were determined from the activation curves. Under our experimental conditions the parameters for wild-type apoCII were K_d = 0.29 ± 0.07 µM and  = 6.2 ± 0.1. Corresponding data for all mutants are shown in Table II.

View larger version (14K): [in this window] [in a new window]   Fig. 5.   Activation of lipoprotein lipase by recombinant wild-type apoCII and mutants produced as fusion proteins with the S-tag. Purified and cleaved wild-type apoCII and mutants (shown in Fig. 2) were incubated with LPL (7.85 ng) in a total volume of 200 µl assay mixture. Each data point is the mean of triple determinations.

View this table: [in this window] [in a new window]   Table II Kinetic parameters for the interaction of apoCII mutants with LPL and with liposomes Kd and  were determined from the activation curves. Klip is the dissociation constant for the apoCII·liposome complex determined by Trp fluorescence.

To investigate whether the markedly reduced activation ability of some of the mutants was due to a major change of secondary structure, mutants Y63A, I66A, D69A, and Q70A were analyzed by CD measurements in the presence of liposomes of phosphatidylcholine (Fig. 6). No major differences were found compared with the spectrum for wild-type apoCII. All mutants retained the typical, largely helical structure represented by double minima at 208 and 222 nm.

View larger version (12K): [in this window] [in a new window]   Fig. 6.   Circular dichroism spectra of wild-type apoCII and alanine replacement mutants of apoCII in the presence of liposomes of egg yolk phosphatidyl choline. The concentration of phospholipids was 240 µM, and the concentration of apoCII was in all cases 4.7 µM. The experiments were performed in 0.1 M phosphate buffer, pH 7.4, at 25 °C All spectra are means of three scans. WT (); D69A (- · ·  -); Y63A (· · · · ·); I66A (- - -); Q70A (-  ·  - ·).

In addition to the alanine substitutions mentioned above, Tyr⁶³ was replaced by threonine (Y63T), leucine (Y63L), phenylalanine (Y63F), and tryptophan (Y63W) to investigate the importance of the hydroxyl group and/or the aromatic ring (Fig. 7A). The Y63F and Y63W mutants showed very low, though detectable, activation. Because the activation curves remained linear even at the maximally achievable apoCII concentrations, it was not possible to calculate K_d and  for these mutants. To elucidate whether the poor activation was caused by low affinity of apoCII for LPL, or by low activity of the mutant apoCII·LPL complexes, the mutants Y63W and Y63F were mixed with wild-type apoCII in the same assay (Fig. 8). The activity of wild-type apoCII was not affected even at a 20-fold excess of the inactive mutants. Therefore, we concluded that the low activation by mutants Y63F and Y63W was mainly due to a reduced affinity for LPL. In contrast, the Y63T and Y63L mutants were more efficient than Y63A (Fig. 7A). The Y63T bound to LPL with K_d = 1.73 µM and enhanced the catalytic activity 4-fold ( = 4.0, Table II). The corresponding values for the Y63L mutant were 3.14 µM and 4.2-fold.

View larger version (19K): [in this window] [in a new window]   Fig. 7.   Stimulation of lipoprotein lipase by His-tagged apoCII mutants at residues 63, 66, 69, and 70. Conditions for the incubations were the same as in Fig. 5. Each data point is the mean of triple determinations. A, stimulation of LPL by apoCII and mutants at residue 63; B, stimulation of LPL by mutants at residue 66; C, stimulation of LPL by mutants at residue 69; D, stimulation of LPL by mutants at residue 70. In all panels the fitted curve for wild-type apoCII is shown by a dotted line.

View larger version (11K): [in this window] [in a new window]   Fig. 8.   Competitive assay using wild-type apoCII and mutants with low activity. The incubations were carried out as described in Fig. 7 except for the presence of 2.5 µM of the mutants Q70E, Y63F, or Y63W in the incubation mixture together with the indicated amounts of wild-type apoCII. Filled circle, wild-type alone; open circle, in the presence of Y63W; filled square, in the presence of Y63F; open square, in the presence of Q70E.

Substitution of Ile⁶⁶ by leucine (I66L), alanine (I66A), or phenylalanine (I66F) caused significant changes in the activation properties of apoCII (Fig. 7B). The activation curve for I66L remained close to linear up to 3 µM. Therefore the K_d and  values for this mutant were determined with large deviation (Table II). Also I66A exhibited low affinity K_d = 3.27 µM, and the maximal activation was about 3-fold ( = 2.9). I66V was the most active among the mutants at position 66. It appeared to bind to LPL even more tightly than wild-type apoCII (K_d = 0.11 µM). However, I66V could not activate LPL to the level reached with wild-type apoCII ( = 3.9). At high concentrations (>3 µM) there was even a tendency for inhibition of LPL by this mutant (Fig. 7B).

Substitution of Asp⁶⁹ by hydrophilic residues such as serine (D69S), glutamate (D69E), and asparagine (D69N) did not lead to prominent changes in the activating properties (Fig. 7C). Effects were mainly seen on the K_d values, which were increased ~6-fold compared with wild-type apoCII. Mutants D69S and D69E had similar  values as wild-type apoCII, but for D69N it was somewhat lower ( = 4.9). The weakest binding affinity was seen with the D69A mutant (K_d = 3.40 µM).

Three mutants were generated at position Gln⁷⁰: Q70A, Q70N, and Q70E (Fig. 7D). The most active one was Q70N with K_d = 2.07 µM and  = 3.0. Both Q70A and Q70N showed lowered binding affinity to LPL and decreased maximal activation (Table II). In contrast to these mutants, the Q70E mutant was completely inactive and could not compete with the wild-type apoCII in the assay (Fig. 8). Thus, a negative charge at position 70 appeared to block binding of apoCII to LPL.

In addition to substitutions at the fully conserved residues, three mutants were generated at residue Leu⁷², Ser⁷³, and Val⁷⁴ by alanine replacement. These mutants bound to LPL with an apparent affinity comparable to that of wild-type apoCII, and they activated the enzyme to the same extent as the wild-type apoCII did (data not shown) Thus, the structures of these 3 residues did not appear to be essential for the function of apoCII.

Interaction between LPL and ApoCII in the Absence of Hydrolyzable Lipids-- The interaction of apoCII mutants with LPL was studied in the presence of liposomes of a non-hydrolyzable lipid (1,2-di-O-hexadecyl-sn-glycero-3-phosphatidylcholine). In these experiments the apoCII mutants were dansylated, and their association to LPL was measured by monitoring fluorescence energy transfer from LPL tryptophans to the dansyl groups of the modified apoCII molecules. Dissociation constants (K) were determined for the apoCII·LPL complexes. All studied apoCII mutants appeared to bind to LPL with an affinity comparable to that of wild-type apoCII. The values for the dissociation constants were 0.3-0.8 µM (Fig. 9). For wild-type apoCII this value was similar in the absence of liposomes (data not shown). Thus, in the absence of hydrolyzable lipids, the dissociation constants (K) did not vary as much as they did in the presence of the radiolabeled emulsion of triacylglycerols, where the K_d varied from 0.11 to 4.6 µM (Table II). Fig. 9 demonstrates that there was no correlation between the K_d and K values.

View larger version (7K): [in this window] [in a new window]   Fig. 9.   Lack of correlation between Kd and K values. The determination of the Kd and K is described under "Experimental Procedures". Kd originates from activation curves, whereas K was determined from studies of the direct interaction between apoCII and LPL using dansylated apoCII.

Interaction of ApoCII with Lipids-- To elucidate whether the poor activation of LPL by some mutants was caused by a reduced interaction with lipids, the mutants were titrated with liposomes of phosphatidylcholine. Formation of liposome·apoCII complexes was measured by monitoring fluorescence of the single tryptophan (Trp²⁶) of apoCII on excitation at 280 nm. We had previously shown that on association of wild-type apoCII with liposomes, the tryptophan fluorescence increased and shifted the emission maximum from 345 to 320 nm (19). In the present study, we detected similar changes in the fluorescence for all of the studied apoCII mutants. Consequently, even the inactive mutants were able to interact with lipids. Apparent dissociation constants (K_lip) calculated from the titration curves, ranged between 101 and 420 µM (Table II).

	DISCUSSION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

The starting point for the present study was our previous observation that there are only seven fully conserved residues in the C-terminal one-third of the apoCII molecule, the part to which the ability to activate LPL had previously been localized (16). This implied that the detailed structure of this region, limited to residues 59-70, was important and had therefore not changed during evolution. We generated a set of single-residue substitution mutants of human apoCII, mainly at the seven fully conserved sites. ApoCII was expressed either as a fusion protein with the S-tag or the 6xHis-tag at the N-terminal end. The expressed apoCII appeared in two forms, one of which was due to truncation of the last 2 amino acid residues at the C terminus (Glu⁷⁸-Glu⁷⁹). Although the short form in some cases constituted about 75% of the total apoCII protein, we did not further separate the individual forms, because both were shown to activate LPL. Based on sequence comparisons of apoCII from different animal species we had previously concluded that the negatively charged residues (Glu⁷⁸-Glu⁷⁹) in the C-terminal end of human apoCII are not important for the activation (16). This was further supported by the present findings with the recombinant, truncated forms.

Effects of Mutations-- There was little or no effect of some of the presently investigated mutations on the secondary structure of apoCII. We can, however, not absolutely exclude conformational effects for some of the other mutants, which were not produced in sufficient amounts for CD measurements. However, theoretical structure prediction using several available computer programs did not indicate major changes for any of the mutants studied here.

To investigate if the mutations affected formation of the apoCII·LPL complex, the catalytic power of the complex, or both, we derived two kinetic parameters: the apparent dissociation constant K_d and the activation factor . The aim was to test the hypothesis that the LPL interaction domain of apoCII contains two regions: one contributing mainly to activation, whereas the other is mainly responsible for binding of apoCII to LPL (22). For most of our mutants, however, both parameters were affected, indicating that discrimination between these functional parts of apoCII was not possible.

Tyr⁶³ appeared to be the most sensitive residue. Both K_d and  were reduced for all studied mutants. Thus, Tyr⁶³ is probably involved both in the binding to LPL and in the activation. Substitution by Phe, Trp, Ala, or Leu reduced the activation factor more than did substitution by Thr. This indicated that the hydroxyl group could be involved in formation of a hydrogen bond between residue 63 and LPL during activation, but the difference between Y63T and the wild-type was still very large. The importance of Tyr⁶³ was previously demonstrated by substitution with Trp and Gly in a synthetic apoCII peptide spanning residues 44-79. This caused reduced but not lost ability to activate LPL (23).

The mutations of Ile⁶⁶ suggested that at this position both shape and hydrophobicity of the side chain are essential properties. I66A and I66F were much less efficient than I66L and I66V, indicating that a hydrophobic aliphatic side chain is preferred. I66V demonstrated even stronger apparent binding affinity to LPL than wild-type apoCII. The maximal activation level for I66V was, however, somewhat lower than that of the wild-type apoCII. Thus, even minor structural changes at position 66 led to marked changes in the activation. It appears as if the steric fit of Ile⁶⁶ into a narrow hydrophobic pocket in LPL is important for full activation.

For residue Asp⁶⁹, mutants D69S, D69E, and D69N did not show very dramatic changes in the activating ability, suggesting that apoCII may tolerate some variation in hydrophilic residues at this position. The main effect was an increase in the K_d values. The D69S mutant was practically as efficient as the D69E mutant. Thus, a negative charge at this position did not seem to play a crucial role. In contrast to hydrophilic residues, substitution with alanine caused a more pronounced decrease in apparent binding affinity.

One of the most dramatic effects on activation was obtained by substitution of Gln⁷⁰ by glutamate, indicating that the corresponding binding region in LPL is probably negatively charged. Also the Q70N mutant was rather inefficient, demonstrating that the additional carbon in the side chain of the glutamine residue is probably important for the interaction with LPL. Substitution with alanine decreased both the binding affinity (K_d) and activation ability (). Thus, not only the amide group in the side chain of residue 70 but also the length of the carbon side chain was important for the function.

LPL Interactions-- To further characterize the properties of our mutants, we studied the apoCII·LPL interaction in the absence of hydrolyzable lipids. In the system with ether phospholipids, the mutants did not show significant differences with regard to LPL affinity. These results are in conflict with the apparent LPL affinities from the activation experiments (K_d), where several of the mutants showed little or no activation. The competition experiments showed that the low activity mutants could not compete with wild-type apoCII for binding to LPL, even at a 20-fold higher concentration. Thus, the apoCII·LPL complex observed in the absence of hydrolyzable lipid may not be the true functional complex that can hydrolyze lipids at optimal rate. One explanation could be that the lipid substrate is needed for induction of the right conformation of either LPL or apoCII and that some of the mutations affected this interaction. It is, unfortunately, not possible to study the interaction by fluorescence with a hydrolyzable lipid. The substrate would almost immediately be consumed by the high amounts of lipase needed for the experiments.

Recently it was reported that the synthetic apoCII fragment 39-62 activates LPL to the same extent as full-length apoCII (24). These data cannot be understood from our present and previous experiences. In our study, this region is intact in all affected mutants. Still, their ability to activate LPL was severely reduced or absent. Therefore, the proposed region cannot be crucial for activation, at least not as part of native apoCII. It is possible that the peptide, by nonspecific effects, causes an increase in LPL activity. From previous experiments the activation seems to be specific, and similar effects have not been found by other peptides. We cannot exclude that the LPL-binding site is extended toward the N-terminal end to also involve residues 59 and 62. These residues were only substituted by glycine and alanine, respectively, with little change in the activation. Considering the evolutionary conservation of these residues, and their location preceding the third -helix (8), it is possible that other substitutions would have had effects.

Lipid Interactions-- The investigation of the apoCII·liposome interaction showed that all mutants had lipid-binding affinities comparable to that of wild-type apoCII. Thus, the loss or reduction in the activation ability of some of the mutants was not due to reduced ability to interact with lipid/water interfaces. This is consistent with the previous observations that the N-terminal domain of apoCII is the part mainly responsible for binding to lipids (9-11). The mutations could, however, still affect details in the steric arrangement of apoCII at the lipid/water interface in ways that were not detectable by the methods used here to measure lipid interaction.

ApoCII-binding Site-- It is still not clear where the apoCII-binding site is located on LPL. Studies with lipase chimeras of LPL and hepatic lipase suggest that apoCII binds to the N-terminal folding region of LPL (25, 26). It was also suggested that apoCII interacts simultaneously with regions located in the N- and C-terminal domains of opposing subunits of the LPL dimer (27). The crystal structure of the related pancreatic lipase shows that its activator, colipase, binds to the loop covering the active site (28). It is, however, well known that apoCII and colipase differ both in structure and functions. Recently the binding site of apolipoprotein AI (apoAI) for lecithin-cholesterol acyltransferases (LCAT) was reported by Roosbeek et al. (29). In apoAI three strictly conserved arginine residues point out from the same side of an amphipathic -helix, which interacts with lipids on the other side. The corresponding binding site in LCAT for its activator has not yet been defined. It is clear, however, that this interaction has a different nature than that between apoCII and LPL, which does not appear to involve charged residues. Thus additional studies are required to identify the apoCII-binding site on LPL.

Conclusions-- Our results suggest that matching between Ile⁶⁶ and LPL is important for full activation. Introduction of other hydrophobic side chains like those of Phe, Ala, Leu, and Val led to decreased activation. The other clear result is that a negative charge cannot be tolerated in position 70. From our data we conclude that residues 66, 69, and 70 contribute, together with Tyr⁶³, to the LPL-binding site of apoCII. The lack of change in the activation ability by substitution of residues 72, 73, and 74 by alanine indicated that these residues are outside the LPL interaction site, even though residue 73 is situated on the same side of the -helix as residues 69 and 70 (8). Taken together, multiple residues (Tyr⁶³, Ile⁶⁶, Asp⁶⁹, Gln⁷⁰) are involved in formation of the LPL-binding site of apoCII, but no single amino acid was absolutely crucial for the activation.

	ACKNOWLEDGEMENTS

We thank Dr. Per-Ingvar Ohlsson for the N-terminal amino acid sequence analyses and Dr. Juhan Subbi for the analyses by mass spectrometry.

	FOOTNOTES

^* This work was supported by the Swedish Medical Research Council (Grants 727 and 12203), by the Swedish Royal Academy of Sciences, by the Estonian Science Foundation (Grant 4925), and by Donation Funds at the Medical Faculty, Umeå University.The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

^§ Present address: Cardiovascular Research Center, Cardiology Division, Massachusetts General Hospital, Harvard Medical School, Boston, MA 02129-2060.

To whom correspondence should be addressed. Tel.: 46-90-786-7762; Fax: 46-90-786-7840; E-mail: Gunilla.Olivecrona@medchem.umu.se.

Published, JBC Papers in Press, November 21, 2001, DOI 10.1074/jbc.M105421200

² J. Zdunek, G. V. Martinez, J. Schleucher, P. O. Lycksell, Y. Yin, S. Nilsson, Y. Shen, G. Olivecrona, and S. Wijmenga, manuscript in preparation.

³ Y. Shen, A. Lookene, H. Vija, and G. Olivecrona, unpublished data.

	ABBREVIATIONS

The abbreviations used are: apoCII, apolipoprotein CII; apoAI, apolipoprotein AI; CD, circular dichroism; GdmCl, guanidinium chloride; IPTG, isopropyl--D-thiogalactoside; LCAT, lecithin-cholesterol acyltransferase; LPL, lipoprotein lipase; Ni²⁺-NTA, nickel-nitrilotriacetic acid; Tricine, N-[2-hydroxy-1,1-bis(hydroxymethyl)ethyl]glycine.

	REFERENCES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

CiteULike

Complore

Connotea

Del.icio.us

Digg

Reddit

Technorati

This article has been cited by other articles:

				C. S. Gangabadage, J. Zdunek, M. Tessari, S. Nilsson, G. Olivecrona, and S. S. Wijmenga Structure and Dynamics of Human Apolipoprotein CIII J. Biol. Chem., June 20, 2008; 283(25): 17416 - 17427. [Abstract] [Full Text] [PDF]

				C. J. A. Moen, A. P. Tholens, P. J. Voshol, W. de Haan, L. M. Havekes, P. Gargalovic, A. J. Lusis, K. W. van Dyk, R. R. Frants, M. H. Hofker, et al. The Hyplip2 locus causes hypertriglyceridemia by decreased clearance of triglycerides J. Lipid Res., October 1, 2007; 48(10): 2182 - 2192. [Abstract] [Full Text] [PDF]

				K. E. North, H. H. H. Goring, S. A. Cole, V. P. Diego, L. Almasy, S. Laston, T. Cantu, B. V. Howard, E. T. Lee, L. G. Best, et al. Linkage analysis of LDL cholesterol in American Indian populations: the Strong Heart Family Study J. Lipid Res., January 1, 2006; 47(1): 59 - 66. [Abstract] [Full Text] [PDF]

				A. Lookene, L. Zhang, V. Tougu, and G. Olivecrona 1,1'-Bis(anilino)-4-,4'-bis(naphtalene)-8,8'-disulfonate Acts as an Inhibitor of Lipoprotein Lipase and Competes for Binding with Apolipoprotein CII J. Biol. Chem., September 26, 2003; 278(39): 37183 - 37194. [Abstract] [Full Text] [PDF]

				T. L. McIlhargey, Y. Yang, H. Wong, and J. S. Hill Identification of a Lipoprotein Lipase Cofactor-binding Site by Chemical Cross-linking and Transfer of Apolipoprotein C-II-responsive Lipolysis from Lipoprotein Lipase to Hepatic Lipase J. Biol. Chem., June 13, 2003; 278(25): 23027 - 23035. [Abstract] [Full Text] [PDF]

				M. L. Misso, Y. Murata, W. C. Boon, M. E. E. Jones, K. L. Britt, and E. R. Simpson Cellular and Molecular Characterization of the Adipose Phenotype of the Aromatase-Deficient Mouse Endocrinology, April 1, 2003; 144(4): 1474 - 1480. [Abstract] [Full Text] [PDF]

				M. S. Kim, J. Shigenaga, A. Moser, K. Feingold, and C. Grunfeld Repression of Farnesoid X Receptor during the Acute Phase Response J. Biol. Chem., March 7, 2003; 278(11): 8988 - 8995. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) All Versions of this Article: 277/6/4334    most recent M105421200v1 Submit a Letter to Editor Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Shen, Y. Articles by Olivecrona, G. Search for Related Content PubMed PubMed Citation Articles by Shen, Y. Articles by Olivecrona, G. Social Bookmarking                      What's this?