1030

Lipoprotein(a) and Apolipoprotein(a) in aNew World Monkey, the Common Marmoset

(Callithrix jacchus)Association of Variable Plasma Lipoprotein(a) Levels

With a Single Apolipoprotein(a) Isoform

Heng-Chang Guo, Jean-Baptiste Michel, Yves Blouquit, and M. John Chapman

In an earlier report (Chapman et al, Biochemistry 1979;18:5096-5108), we suggested that thecommon marmoset may represent an important model for the study of human plasmalipoprotein metabolism. We now extend the interest of this monkey model to the study oflipoprotein(a) (Lp[a]) and apolipoprotein(a) (apo[a]). Density gradient ultracentrifugalfractionation of marmoset plasma revealed a bimodal distribution of Lp(a), with one peak ofconcentration occurring in association with very low density lipoproteins (VLDLs) and a secondin the density range 1.040-1.080 g/ml. The dense Lp(a) subspecies displayed physicochemicalproperties (chemical composition, particle size, and electrophoretic mobility) that closelyresembled those of its counterpart in humans and baboons but that were distinct from those oflow density lipoprotein (LDL). Furthermore, the particle size of marmoset Lp(a) was invariant(31 nm) over the density interval 1.040-1.080 g/ml, whereas that of LDL decreased progres-sively with an increase in density (—26-25.2 nm). Use of polyclonal and monoclonal antibodiesto human apo(a) and of a polyclonal antibody to marmoset Lp(a) allowed immunologicidentification of a single apo(a) isoform in the marmoset whose size was similar to that of apoB-100; apo(a) and apo B-100 were associated in Lp(a) particles by a disulfide linkage. The totalprotein mass of apo-Lp(a) was estimated to be 800,000 or more by electrophoresis in sodiumdodecyl sulfate-polyacrylamide-agarose gels. The amino acid compositions of marmoset andhuman apo(a) resembled each other but were distinct from those of the corresponding forms ofapo B-100. Immunologic evidence is provided for a high degree of cross reactivity betweenapo(a) in marmosets, baboons, and humans, supporting the idea of the existence of a markeddegree of structural homology between these proteins. In addition, electroimmunoblotting ofmarmoset apo(a) and marmoset plasminogen showed that these proteins shared certainepitopes in common, suggesting that marmoset apo (a) may possess kringle-like structuralfeatures. Finally, despite possession of a single apo(a) isoform, marmoset Lp(a) levels variedover a 100-fold range (0.5-49 mg/dl plasma). Considered together, our present findings suggestthat the common marmoset monkey constitutes a unique model in which to study the regulationof apo(a) gene expression and the posttranslational processing of apo(a), as well as factors thatmodulate the synthesis, intravascular metabolism, and cellular catabolism of Lp(a). (Arterio-sclerosis and Thrombosis 1991;ll:1030-1041)

Since its original detection in human serum in1963,1 lipoprotein(a) (Lp[a]) has been impli-cated in both atherogenesis and thrombogen-

esis (for a review, see Reference 2). The precisemolecular mechanisms involved in the expression of

From INSERM Unite 321 (H.-C.G., MJ.C), "Lipoproteins andAtherogenesis," Pavilion Benjamin Delessert, Hopital de la Pitie,Paris; INSERM Unite 36 (J.-B.M), "Vascular Pathology and RenalEndocrinology," Paris; and INSERM Unite 91 (Y.B.), "MolecularGenetics and Hematology," Hopital Henri Mondor, Cr6teil, France.

Supported by the Faculte de M6decine Universite Paris VI(Pitie-Salpetriere), the Fondation pour la Recherche Medicale,

such activities by this intriguing lipoprotein particle,however, remain only partially resolved.

Progress in our understanding of the pathophysi-ology of a specific disease state in humans frequentlyresults from investigations in defined and appropri-

INSERM, and the Commissariat d'Energie Atomique (grant No.SA 9550/CG). H.-C.G. was supported by a Research Fellowshipfrom the Fondation pour la Recherche Medicale.

Address for reprints: Dr. M. John Chapman, INSERM U.321,Pavilion Benjamin Delessert, Hopital de la Pitie, 75651 ParisCedex 13, France.

Received October 10, 1990; revision accepted March 25, 1991.

by guest on Novem

ber 8, 2017http://atvb.ahajournals.org/

Dow

nloaded from

Guo et al Marmoset Lipoprotein(a) 1031

ate animal models. In the case of Lp(a), the avail-ability of suitable animal species in which to under-take extensive studies of atherogenesis and ofthrombotic phenomena is presently restricted to thenonhuman primates. Indeed, biochemical, genetic,and immunologic evidence has documented the oc-currence of Lp(a) in species of both great and lesserapes such as the chimpanzee, orangutan, and gibbonand in numerous species of Old World monkeyincluding the baboon, rhesus monkey, Erythrocebuspatas, and pig-tailed and stump-tailed monkeys.3"11

Of these species, attention to date has been primarilyfocused on the baboon (genus Papio) and the rhesusmonkey (Macaca mulatto)*'5-1-11

Despite the close phylogenetic relation of OldWorld monkeys to humans, use of the former forlipoprotein and atherosclerosis research has tendedto be restricted to specialized primate centers, giventhe difficulties encountered in their handling, breed-ing in captivity, and the high cost of colony mainte-nance. Such limitations have favored the recentdevelopment of colonies of smaller species of nonhu-man primates, in particular the common marmoset,Callithrix jacchus. 12~14 This New World monkey, amember of the family Callithrichidae, breeds readilyin captivity, is conveniently handled and bled in viewof its small size (as much as =400 g in adults), and isrelatively easy and inexpensive to maintain.12"14 In-deed, it was such characteristics that originallyprompted us to propose this species as an appropri-ate and promising model for lipoprotein research,15

in addition to our observation that both the qualita-tive and quantitative aspects of lipoprotein and apo-lipoprotein profile in C. jacchus closely resemblethose found in normolipidemic humans.1516 Morerecently, the susceptibility of the marmoset to diet-induced atherosclerosis, as originally described byDreizen et al,17 has been confirmed18 and in additionhas been extended to demonstrate the occurrence ofboth foam cell lesions and proliferative intimalplaques in monkeys hyperresponsive to diets supple-mented with cholesterol and saturated fat.18

To date, both biochemical and immunologic studieshave failed to reveal the presence of circulating Lp(a)in the plasmas of New World monkeys, including thecommon marmoset.31518 As plasma Lp(a) levels mayvary over a wide range both in humans and in nonhu-man primates,2-4'7-91119 it seems plausible to suggestthat the apparent absence of Lp(a) in New Worldspecies may result at least in part from the chanceselection of animals with low Lp(a) concentrations(<5 mg/dl) and in part from a low degree of immu-nologic cross reactivity of the monkey Lp(a) particleswith antisera to their human counterparts.3-4

In light of the foregoing considerations and withthe aim of extending the marmoset model to studiesof the atherogenic and thrombotic role of Lp(a), weundertook further investigation of the possible occur-rence of this cholesterol-rich low density lipoprotein(LDL)-like particle in C. jacchus. We now describethe purification, physicochemical characterization,

and plasma density profile of Lp(a) and of its specificprotein component apolipoprotein (apo)(a) in thecommon marmoset monkey. Both the lipoprotein par-ticle and its distinctive apo(a) showed marked homol-ogy with their human counterparts. The marmosetprotein is, however, distinguished by its lack of poly-morphism and presents as a unique size species whosecirculating concentrations varied widely between mon-keys. The results support and extend our earliercontention that the marmoset constitutes a highlyvaluable experimental animal model for comparativestudies of the role of lipoproteins in atherosclerosis.

MethodsAnimals and Diets

The marmosets used in this study form part of abreeding colony established at INSERM U.36, andthe investigations described herein were made onblood samples taken from animals within this facility.Animals in our colony are genetically diverse. Thus,part of the original breeding population was derivedfrom the CIBA-GEIGY colony in Basel, Switzerland,and the remainder from Shamrock Farms in the UK.Marmosets were caged in pairs, and adult animalsweighed approximately 350 g; the temperature wasmaintained between 19°C and 23°C, and the relativehumidity was kept between 60% and 70%. Theanimals were fed a diet that consisted of 20% pro-tein, 7.5% fat, 52% carbohydrate, 3% cellulose, 5.5%mineral mixture, and 12% moisture; a mixture ofvitamins was also included in this diet. The diet wasmade up as a pelleted preparation (UAR, Villemois-son-sur-Orge, France). Supplements of fruit (apples,oranges, or bananas) were offered daily.

Blood SpecimensBlood samples (as much as 1 ml) were routinely

taken from restrained, unanesthetized trained ani-mals that had been fasted overnight for approxi-mately 15 hours. Blood was withdrawn into glasstubes containing Na2EDTA (final concentration, 1mg/ml) by direct needle puncture of the femoralvein, and plasma was rapidly separated by low-speedcentrifugation at 4°C. Lipoprotein fractionation wascommenced within 3-4 hours of plasma separation,during which time the plasma samples were cooledon ice at 4°C, and NaN3 and sodium merthiolate(final concentrations, 0.01% and 0.001% wt/vol, re-spectively) were added. Lipoprotein studies wereperformed either on aliquots of individual plasmasamples (as much as =0.5 ml) or on pooled samplesfrom six or more monkeys (=3 ml or more).

Lipoprotein Isolation by DensityGradient Ultracentrifugation

Marmoset lipoproteins were fractionated on thebasis of their hydrated densities by the isopycnicultracentrifugal density-gradient procedure de-scribed for human and marmoset serum lipoproteins

by guest on Novem

ber 8, 2017http://atvb.ahajournals.org/

Dow

nloaded from

1032 Arteriosclerosis and Thrombosis Vol 11, No 4 July/August 1991

by Chapman et al15-20 and subsequently modified.21

Ultracentrifugation was performed in an SW41 rotorof a Beckman L8-55 ultracentrifuge at 40,000 rpm(56xl07g/min) f°r 48 hours at 15°C; no brake wasused at the end of the run. On completion ofcentrifugation, 20 0.5-ml fractions were collectedsuccessively from the meniscus of each tube byaspiration with a narrow-bore Pasteur pipette; thebottom fraction, essentially devoid of lipoproteins,was removed in a volume of approximately 2 ml. Ourgradient fractionation did not take into account theposition of visible lipoprotein bands, but their posi-tions were noted relative to the number of thevarious fractions. The background density of eachlipoprotein gradient subfraction was determined byreference to the density profile obtained from controlgradients.20

Lipoprotein fractions were exhaustively dialyzed inSpectrapor tubing (Spectrum Medical Industries, LosAngeles, Calif.; M, cutoff of 12,000) for 48 hours at 4°Cagainst a solution containing either 0.1 M NaCl and0.01% EDTA (pH 7.4) or 0.005 M NH^HCOj, 0.04%EDTA, and 0.01% NaN3 (pH 7.4), at which timesamples were destined for apolipoprotein analysis.

Isolation of Lipoprotein(a)Lp(a) was isolated from pooled marmoset plasma

samples (=50 ml) essentially according to the proce-dure of Ehnholm et al.22 After selection of thedensity interval over which the majority of Lp(a) wasdistributed in the marmoset (see "Results"), theplasma was brought to a solvent density of 1.050 g/mlby addition of solid KBr23; centrifugation was thenperformed in the 50 Ti rotor of a Beckman L8.55ultracentrifuge at 45,000 rpm for 24 hours at 10°C.After centrifugation the top fraction (~2 ml) wasremoved, and the lower fraction was brought to adensity of 1.100 g/ml with solid KBr. After centrifu-gation under the same conditions as before but for 48hours, the top fractions (~2 ml) containing lipopro-teins of d=1.050-1.100 g/ml were removed by aspi-ration with a narrow-bore Pasteur pipette andpooled. The pooled d=1.050-1.100 g/ml subfractionswere then concentrated to a volume of approximately4 ml by use of a collodion thimble membrane undervacuum (Schleicher and Schuell, Darmstadt, FRG);the final protein concentration was in the range 2-5mg/ml.

Lp(a) was purified from the d=1.050-1.100 g/mllipoproteins by gel-filtration chromatography24 on anagarose column (Bio-Gel A-5m; Bio-Rad, La Jolla,Calif.) with dimensions of 0.9x80 cm. This columnwas eluted under the conditions as described byKrempler et al.24 Fractions that eluted in distinctpeaks were pooled (see Figure 3), and their puritywas evaluated by chemical analysis and electropho-resis in agarose gel (see "Results").

Purification of Apolipoprotein(a)Marmoset apo(a) was separated from Lp(a) by a

modification of the methodology of Armstrong et al25

and Laplaud et al.26 Dithiothreitol (final concentra-tion, 20 mM) was added to 1 mg native Lp(a) in afinal volume of 1 ml, and the solution was incubatedat 37°C for 1 hour. The mixture was then adjusted toa solvent density of 1.100 g/ml with solid KBr andcentrifuged at 40,000 rpm for 48 hours at 10°C in a Ti50.3 Beckman rotor. After completion of centrifuga-tion, apo(a) was found as a translucent pellet; thispellet was solubilized in a solution of 20 mM tris(hy-droxymethyl)aminomethane (Tris), 1 mM EDTA, 10mM dithiothreitol, and 2% sodium dodecyl sulfate(SDS) (pH 7.2). Possible contamination of apo(a)preparations by marmoset apo B-100 was excluded byfurther purification by immunoaffinity chromatogra-phy. Apo(a) preparations were loaded onto anti-LDL-Sepharose columns, and apo(a) was eluted inthe unbound fraction; such preparations were dia-lyzed against 0.5% HCOOH and subsequently lyo-philized before chemical analysis.

Plasminogen PurificationThe d>l.2A g/ml bottom fraction of marmoset

plasma was taken after density gradient fractionationand dialyzed against a buffer containing 0.1 MNa2HPO4, 0.01% NaN3, and 0.01% EDTA (pH 7.5).Plasminogen was then separated from this fraction byaffinity chromatography on lysine-Sepharose (Phar-macia Fine Chemicals, Uppsala, Sweden) by use ofthe protocol of Deutsch and Mertz.27

Monoclonal and Polyclonal AntibodiesMonoclonal antibodies to human apo(a) {M,

=570,000) were produced from hybridoma cells asdescribed earlier.28 A monoclonal antibody to humanapo B-100 (1.8. C4) was produced from hybridomacells that were kindly provided by Thomas L. Inner-arity. The characterization of this antibody has beenreported earlier.29 A polyclonal antibody to marmo-set apo(a) was produced by subcutaneous injection of10 /Mg apo(a) emulsified in Freund's complete adju-vant into a mouse. The same amount in Freund'sincomplete adjuvant was similarly injected at 2 and 4weeks after the initial immunization. After immuno-logic testing of the antiserum, the mouse was bledand the serum was separated. In addition to itsspecificity for marmoset apo(a) (see Figure 6), thispolyclonal antibody reacted with both human andbaboon apo(a) upon immunoblotting (see Figure 6for baboon apo[a]).

Chemical AnalysisThe protein content of lipoprotein and protein

preparations was determined by the method ofLowry et al30 with the use of bovine serum albumin(BSA) (Sigma Chemical Co., St. Louis, Mo.) as theworking standard. Total and free cholesterols wereestimated with the enzymatic kit of Biomerieux(Charbonnieres-les-Bains, France). Ester cholesterolwas calculated as the difference between total andfree cholesterol; cholesteryl ester was then calculatedas ester cholesterol times 1.67.20 Total triglyceride

by guest on Novem

ber 8, 2017http://atvb.ahajournals.org/

Dow

nloaded from

Guo et al Marmoset Lipoprotein(a) 1033

and phospholipid contents were also determinedwith enzymatic kits (Biomerieux).

Amino acid analysis of acid hydrolysates of puri-fied marmoset apo(a) was performed on a BiotronikLC 6001 autoanalyzer. Hydrolysis of 15-30 /j,g lyoph-ilized apo(a) was effected under the conditions out-lined earlier.16

Electrophoresis of Native LipoproteinsAgarosegel. Aliquots (2-5 /AI) of whole plasma and

of native lipoprotein fractions were electrophoresedfor 40 minutes on agarose gel films (Universal elec-trophoresis film agarose, catalog No. 470100, Cor-ning, Palo Alto, Calif.) with use of the Corning ACIsystem. On completion of electrophoresis, the sheetswere stained for lipid with oil red O. This procedureis essentially the same as described by Noble.31

Pofyacrylamide gradient gels. To assess lipoproteinheterogeneity by identification of specific particlesize species and to determine lipoprotein particlediameters, continuous gradient gel electrophoresis ofnative lipoproteins was performed in a Pharmaciaelectrophoresis apparatus (GE-2/4 LS) loaded withgels containing a 2-16% gradient (PAA 2/16; Phar-macia). Approximately 15 /ig lipoprotein protein wasapplied to each well, and electrophoresis was per-formed at 125 V for 12 hours at 4°C in a Tris-boratebuffer (0.09 M Tris, 0.08 M boric acid, 0.003 MEDTA, pH 8.35).32 Gels were stained overnight in asolution containing 25% CH3OH, 10% acetic acid,and 0.1% Coomassie brilliant blue R-250; destainingwas effected with a solution containing 25% CH3OHand 10% acetic acid. A set of standard proteins withknown hydrated diameters (thyroglobulin, 170 A;ferritin, 122 A; catalase, 104 A; lactic dehydrogenase,81 A; BSA, 71 A; high molecular weight electropho-resis calibration kit, Pharmacia) together with latexbeads (mean diameter, 38 nm; the kind gift of A.V.Nichols) were run in duplicate on each slab asreference markers. From the migration distances ofthe different lipoprotein subfractions and those ofthe proteins contained in the calibration kit, it waspossible to calculate the Stokes' diameters of lipo-protein particles with the Stokes-Einstein equation.33

Apolipoprotein ElectrophoresisAgarose-acrylamide gels and electroimmunoblotting.

Serum (20 /il) and lipoprotein samples (2-20 ngprotein) were first lyophilized and subsequently de-lipidated in ethanol/diethyl ether (3:1, vol/vol) at 4°Covernight. After they were dried under N2, apolipo-protein samples as well as those of purified plasmin-ogen (5-15 ng protein) were solubilized in runningbuffer (see below) containing 5% SDS and 20 mMdithiothreitol; they were then heated at 95°C for 10minutes. Samples were loaded onto polyacrylamideslab gels made from 2.75% acrylamide and 0.07%bisacrylamide and containing 0.5% agarose (Bio-Rad) and 0.2% SDS.34-35 A Tris-glycine buffer (pH8.3) containing 0.2% SDS was used as a runningbuffer in both the upper and lower buffer chambers.

Electrophoresis was performed at a constant voltageof 180 V at 4°C for approximately 2 hours. Oncompletion of electrophoresis, protein bands wereelectrophoretically blotted onto a nitrocellulosesheet. A transfer buffer containing 25 mM Tris, 192mM glycine, and 4% CH3OH (pH 8.3) was used in aTransblot cell (Bio-Rad); electrophoretic transferwas performed for 2 hours at a current of 500 mA.The nitrocellulose was initially treated by incubationin a solution of 3% BSA in 10 mM Tris-150 mMNaCl (TBS) at pH 7.4 for 1 hour at room tempera-ture. The nitrocellulose was then incubated for 4hours at room temperature with 10 fig/ml of amonoclonal antibody to apo(a) or apo B-100 in TBScontaining 3% BSA (vol/vol). After washing for 24hours with TBS containing 0.1% BSA, the blots wereincubated with an iodine-125-labeled conjugatedsheep anti-mouse immunoglobulin G (Amersham) at20°C for 4 hours. The nitrocellulose was washed for24 hours with TBS containing 0.1% BSA. Radioac-tively labeled antibodies were visualized by exposure(-80°C) to x-ray film (Kodak Ortho G Film) for24-48 hours and subsequent film development.

Enzyme-Linked Immunoassay for Lipoprotein(a)Our procedure was based on the methodology of

Engwall and Perlmann36 and was described earlier,28

with the exception that the antigen was either mar-moset plasma (at an appropriate dilution) or isolatedlipoprotein fractions. In addition, monoclonal anti-body 39A1 (in biotin-labeled form) was employed asthe second antibody instead of antibodies 14A12 and53A9. For calibration purposes, marmoset Lp(a)purified as described above was used; the workingconcentration range of our assay was 0.02-1 pg Lp(a)lipoprotein mass.

Dot Immunobinding AssaysFor the initial detection of Lp(a) in lipoprotein

subfractions isolated by density gradient ultracentrif-ugation, we employed a dot immunobinding assayas detailed elsewhere37 and adapted from thatof Hawkes et al.38

Immunologic Quantification of ApolipoproteinsA quantitative estimate of apo B, A-I, and E

concentrations in density gradient subfractions de-rived from marmoset plasma was performed bynephelometric assays based on the use of the BNAnephelometer analyzer (Behringwerke AG, Mar-burg, FRG). These assays were performed in accor-dance with the manufacturer's protocol, employing afixed-time nephelometric methodology with mul-tipoint calibration. Monospecific rabbit antisera tohuman apo A-l and apo B (Behring) and a mono-specific goat antiserum to human apo E prepared inour laboratory were used. The corresponding puri-fied human apolipoproteins (i.e., apo A-I, 165 mg/dl;apo E, 4.6 mg/dl; and apo B, 123 mg/dl) were used asstandards and were supplied as Immunoneph refer-ence standard by Immuno AG. The assay for apo B

by guest on Novem

ber 8, 2017http://atvb.ahajournals.org/

Dow

nloaded from

1034 Arteriosclerosis and Thrombosis Vol 11, No 4 July/August 1991

101.013 1.024

Density (g/ml)1.055 1.120 1.205

1I

O)

5 10 15

Density gradient fraction number

FIGURE 1. Plot showing representativedensity (g/ml) gradient distribution ofLp(a), total cholesterol, apo B, apo A-I,and apo E in marmoset plasma. Choles-terol, apolipoproteins, and Lp(a) were as-sayed in density gradient subfractions (xaxis) as outlined in "Methods." Lipid andapolipoprotein concentrations on the loga-rithmic-scale y axis are expressed in fig ormg/0.5 ml density gradient subfraction andwere derived from 3 ml plasma that wassubfractionated in each gradient. Choi, cho-lesterol, n; apo B, apolipoprotein B, A; apoA-I, O; Lp(a), lipoprotein(a), • ; and apo E,A. Mean solvent density of individual gradi-ent subfractions was determined as outlinedearlier20 and is indicated on the x axis at thetop of the figure.

was linear over the range 14-480 mg/dl, the apo A-Iassay over the range 18-580 mg/dl, and that of apo Efrom 0.5 to 4.5 mg/dl. Before immunoassay, sampleswere diluted 1:20. The immunologic cross reactivityof the major marmoset apolipoproteins (i.e., apo B,apo A-I, and apo E) with their counterparts inhumans was documented in our earlier studies15-16;furthermore, recent sequence data have confirmedthe close structural homology between human andmarmoset apo A-I and E.39

ResultsDensity Distribution of Lipoprotein(a), ApolipoproteinB, Total Cholesterol, Apolipoprotein A-I, andApolipoprotein E in Marmoset Plasma

Preliminary studies employing our polyclonal anti-body to human Lp(a) or monoclonal antibodies tothis same antigen28 revealed positive reactions withmarmoset plasmas in dot immunobinding assays(data not shown); equally, immunoblots of marmosetplasma electrophoresed in an agarose-polyacryl-amide gradient gel system and revealed with mono-clonal antibody 39A128 permitted tentative identifi-cation of a high MT counterpart of human apo(a) (seebelow). On the basis of these initial findings, weproceeded to fractionate marmoset plasma by densitygradient ultracentrifugation (see "Methods") and toapply our enzyme-linked immunosorbent assay formarmoset Lp(a) to the gradient subfractions.

The quantitative distributions of Lp(a), apo B,total cholesterol, apo A-I, and apo E were deter-mined as a function of density in three differentplasma pools (Figure 1). A peak in cholesterol, apoE, and apo B concentrations occurred concomitantlywith the presence of triglyceride-rich lipoproteins infraction 1 (d< 1.013 g/ml); a peak of Lp(a) concen-tration (=0.5 mg/0.5 ml) was also detected in thissubfraction. To exclude the possibility that Lp(a)reactivity in the d< 1.013 g/ml fraction might arisefrom bound plasminogen,40 we performed electro-phoresis of the total d< 1.013 g/ml protein content

after reduction in agarose-acrylamide gels (see"Methods") followed by electroimmunoblotting withour polyclonal antibody to human apo(a); this anti-body reacts strongly with plasminogen.28 A positiveblot revealed a diffuse band with Mr in the range500,000-600,000 but failed to detect plasminogen(Afr of 90,000). The Lp(a) immunoreactivity presentin marmoset triglyceride-rich lipoproteins thereforeappears to reflect the presence of apo(a).

The major broad peak in cholesterol and apo Blevels coincided with the distribution of LDLs overthe density range 1.024-1.055 g/ml; over this samerange, apo E levels varied from approximately 2-3/Ag/0.5 ml fraction. Apo A-I levels were at the lowerlimit of our detection method in these subfractions;by contrast, the peak in apo A-I concentrationsoccurring over the density range 1.055-1.120 g/mlclearly defined the distribution of high density lipo-protein (HDL).

A broad Lp(a) peak of d~ 1.040-1.076 g/ml over-lapped the density profiles of both LDL and HDL,the distributions of the latter being indicated by theprofiles of apo B and apo A-I, respectively. Further-more, low levels of Lp(a) (=0.1 mg/0.5 ml fraction)were found over the entire density range classicallyattributed to intermediate density lipoproteins andLDLs (d=1.013-1.065 g/ml). Immunologically reac-tive Lp(a) was, however, undetectable at gradientdensities greater than 1.120 g/ml, thereby suggestingthe absence of lipid-poor or free apo(a) (or apo[a]-B-100 complexes) in marmoset plasma.

The distinct density distributions of LDL andLp(a) in gradient subfractions were further con-firmed by immunologic identification of lipoproteinbands in nondenaturing gradient gel electrophoresis(Figure 2); such immunologic detection involved, onthe one hand, electrophoretic transfer and immuno-blotting and on the other hand, dot immunobinding,by employing monospecific antibodies to eitherapo(a) or apo B-100 in each case (see "Methods").Thus, strong Lp(a) immunoreactivity was found in

by guest on Novem

ber 8, 2017http://atvb.ahajournals.org/

Dow

nloaded from

Guo et al Marmoset Lipoprotein (a) 1035

Subtraction n° , . . 2 • • 3 - • 4 • - 5 • • • - 6 • • - 7 - 9 • 1 0 - 1 1 - 1 2 • - 1 3 • - 1 4 • 1 5

St(nm) i t i i / i i i i i i i i i i

Lp(a)

LDL

Lp(a) (diam.)

LDL (diam.)

Dot blot Lp(a)

FIGURE 2. Photograph showing results of nondenaturing polyacrylamide gradient gel electrophoresis and dot blot analysis oflipoprotein content of density gradient subfractions from marmoset plasma. Electrophoresis was performed in 2-16% gradient gelslabs on successive gradient fractions (n°=l-15; 10 ng protein/well) whose average densities (g/ml) are shown at top. Thecorrespondence between lipoprotein particle diameters and the positions of individual stained gel bands was calculated as detailedin "Methods"; migration positions and Stokes' diameters of latex beads (38.0 nm) and of each marker protein (St, nm) areindicated at left. Diameters of bands identified as Lp(a) and as LDL (arrows) by immunoblotting of the 2-16% gel are indicatedbeneath the gel photo; the second gel as shown was electrophoresed in parallel and stained with Coomassie brilliant blue R-250.In the lower part of the figure, positive dot immunoblots reveal the presence ofLp(a) in discrete subfractions; a monoclonal antibody(39A1) to human apo(a) was used for detection in this assay. Faintly positive dot immunoblots were also found for subfractions3-8 (not shown). Lp(a), lipoprotein(a); LDL, low density lipoproteins; apo(a), apolipoprotein(a).

dot blots of fractions 1 and 2 (d< 1.013 and 1.013-1.016 g/ml, respectively) and in the denser fractions9, 10, and 11 (d=1.048-1.076 g/ml). By contrast, apoB-100 immunoreactivity was detected across the en-tire density interval from 1.013 to 1.100 g/ml (datanot shown). Immunoblotting of electrophoreticallytransferred material from gradient gels showed thatLp(a)-reactive material in fractions 1 and 2 staineddiffusely, corresponding to particles of diameters 31nm or greater (data not shown). In fractions 9-11,however, Lp(a) was visualized as a single band,representing particles of 31.0 nm (Figure 2). Indeed,the invariant diameter of Lp(a) across the densityinterval from 1.048 to 1.076 g/ml distinguished itfrom LDL, which presented as one or more bandsper subfraction and which decreased progressively insize (from a major component of 29 nm diameter atd=1.019 g/ml to one of 25.2 nm at d=1.060 g/ml)with an increase in density. Considerable overlap inthe density distributions of LDL and HDL occurredin fraction 10 of d=1.055-1.065 g/ml, as reportedearlier15 and recently confirmed18; in addition, Lp(a)was a component of this subfraction (Figure 2).

Purification and Characterization ofMarmoset Lipoprotein(a)

Purified Lp(a) was obtained by pooling the centralfractions of the first peak eluted from the Bio-Gel

A-5m column (Figure 3). This material presented asingle band on gradient gel electrophoresis with aparticle diameter of 31 nm, closely resembling that ofpurified human Lp(a) (Figure 3, inset). The secondand third peaks from the agarose column were shownto contain primarily LDL and HDL, respectively, byelectrophoresis in agarose gels and in nondenaturingpolyacrylamide gradient gels (data not shown).

The chemical composition of marmoset Lp(a)markedly resembled that of its human counterpart(Table 1) but was slightly poorer in cholesteryl ester(32.4% compared with 37.8%). Marmoset Lp(a) andLDL isolated from the same plasma pool were dis-tinguished by the elevated cholesteryl ester contentand diminished proportions of both protein andphospholipid in the latter. As noted earlier,15 mar-moset and human LDLs were of similar chemicalcomposition. Purified marmoset Lp(a) and LDLwere also distinct in their electrophoretic mobility inagarose gel, with Lp(a) migrating to a pre-/3 positionand LDL to a /3 position (data not shown).

The apoprotein content of marmoset Lp(a) wasexamined by electrophoresis in SDS-agarose-acryl-amide gels (Figure 4). Native delipidated Lp(a)presented as a single band of MT more than 800,000 inthis system (Figure 4, lane 1); after reduction with 10mM dithiothreitol (Figure 4, lane 2), this band dis-appeared and was replaced by a single smaller com-

by guest on Novem

ber 8, 2017http://atvb.ahajournals.org/

Dow

nloaded from

1036 Arteriosclerosis and Thrombosis Vol 11, No 4 July/August 1991

60

= 40

s.a.

20

?6

I

0,0

160 200

ELUTION VOLUME (ml)

FIGURE 3. Elution profile for purification of marmoset Lp (a) from d=1.050—1.100 g/ml lipoproteins by gel-filtration chromatographyon Bio-Gel A-5m. Chromatography was performed essentially as described by Krempler et al,24 and the absorbance of the eluate wasmonitored at 280 nm (o). Lp(a) content of each fraction (fxg/ml) (•) was estimated by ELISA (see "Methods"). Inset at left shows thenondenaturing gradient gel electrophoretic analysis of marmoset Lp(a) (lane a) purified by this procedure and of human Lp(a) (lane b)separated by a similar method; lane c shows protein standards used for size calibration purposes. Gel was stained with Coomassie brilliantblue R-250. Fractions containing Lp(a) were pooled as indicated by the shaded bar. St(nm), Stokes' particle diameters; Lp(a),lipoprotein(a); ELISA, enzyme-linked immunosorbent assay; LDL, low density lipoprotein; HDL, high density lipoprotein.

ponent whose size (=550,000) was indistinguishablefrom that of marmoset apo B-100 (not shown). More-over, the molecular weight of the band correspondingto the reduced form of apo-Lp(a) was approximatelyhalf that of the nonreduced apoprotein. These find-ings clearly suggested that marmoset apo Lp(a) mightconsist of two disulfide-linked subunits of similar size.This question was further evaluated by immunoblot-ting of SDS-agarose (0.5%)-polyacrylamide (2.75%)gels in which reduced plasmas from several individualmonkeys had been electrophoresed (Figure 5). Whenapo(a) was specifically revealed by use of a mixture offive monoclonal antibodies to human apo(a) followed

TABLE 1. Chemical Composition of Lipoprotein(a) and LowDensity Lipoprotein in the Common Marmoset and in Humans*

by autoradiography (see "Methods"), a single bandwas detected in each of 13 plasmas (Figure 5, patternA, lanes 1-13); this component exhibited an Mr of

- > 800 K

- = 550 K

Component

Cholesteryl esterFree cholesterolTriglyceridePhospholipidProtein

Lp(a)

Marmosett

32.4±1.211.5±1.14.8+0.5

24.1+0.627.2±1.9

Human

LDL

Marmosett(mean weight %)

37.8+1.18.9+1.33.9+0.0

23.4+0.226.0+2.5

43.3 ±1.210.2±0.93.6±0.4

21.8±1.121.1±11.6

Human

42.9±0.79.4±0.84.4±4.6

22.6+0.720.7±1.4

LDL and Lp(a) were isolated by agarose column chromatogra-phy (Figure 3).

Lp(a), lipoprotein(a); LDL, low density lipoprotein.* Values in humans are taken from Reference 41; LDL are of

rf=1.02-1.05 g/ml and Lp(a) of d=1.05-1.08 g/ml.tData for marmoset lipoproteins are the mean±SD of triplicate

analyses of three separate preparations.

FIGURE 4. Electrophoretic patterns in 2.75% SDS-agarose-polyaaylamide gels of purified marmoset apo-Lp(a) in nativeand reduced forms. Lane 1 shows native apo-Lp(a); this sample(40 ijg protein) was solubilized in 5% SDS and running buffer(see "Methods") in the absence of a reducing agent. The samesample was electrophoresed in lane 2 but after solubiUzationunder reducing conditions (10 mM dithiothreitol). A series ofmolecular weight markers was electrophoresed in parallel forcalibration purposes; gels were stained with Coomassie brilliantblue R-250. The position to which marmoset apo B-100 migratesin this gel is indicated by the arrow at right. SDS, sodium dodecylsulfate; Lp(a), lipoprotein(a); apo, apolipoprotein.

by guest on Novem

ber 8, 2017http://atvb.ahajournals.org/

Dow

nloaded from

Guo et al Marmoset Lipoprotein(a) 1037

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17

Apo(a) — «** j**» — B100

FIGURE 5. Electroimmunoblots ofapo(a) and apo B-100 in plasmas from individual marmosets. Plasma (2 \d) was reduced withdithiothreitol, electrophoresed in sodium dodecyl sulfate-polyacrylamide (2.75%)-agarose (0.5%) gels, and electroblotted ontonitrocellulose. In pattern A, lanes 1-13, apo(a) was revealed by incubation with a mixture of five monoclonal antibodies to humanapo (a)28 followed by an iodine-125 -labeled anti-mouse immunoglobulin G. In pattern B, lanes 14-17, apo B-100 was revealedby incubation with a mouse monoclonal antibody (1.8.C4) to human apo B-100, followed by a mI-labeled anti-mouseimmunoglobulin G. Immunoblots in patterns A and B were visualized by autoradiography (see "Methods"). Note the difference inband intensities between individual marmoset plasmas in pattern A. apo(a), apolipoprotein(a).

approximately 550,000. Immunoblots from four ofthe same plasmas were reacted with a monoclonalantibody to apo human B-100 (Figure 5, pattern B,lanes 14-17) and equally revealed a single band(Mr==550,000) whose migration closely resembledthat identified as apo(a). These findings indicate thatmarmoset apo(a) and apo B-100 are of similar sizeand in addition that apo(a) is remarkably homoge-neous within and between individual monkeys, pre-senting as a single molecular weight species.

The apparent presence of a single size isoform ofapo(a) in C. jacchus might, however, result from thelimited immunoreactivity of our monoclonal antibod-ies to human apo(a) with their counterparts in non-human primates. We therefore proceeded to produce

a mouse polyclonal antibody to marmoset apo(a) (see"Methods") and to use this antibody in immunoblotsof both marmoset and baboon plasmas (Figure 6). Asmany as five distinct apo(a) isoforms were identifiedin the baboon (Figure 6, pattern A, lanes 1-6), whilethe same murine polyclonal antibody to marmosetapo(a) detected only a single isoform in the homol-ogous plasma (Figure 6, pattern B, lanes 7-13). Suchdata confirm the polymorphism of baboonapo(a).5'7-9 Control immunoblots were performedwith an unrelated monoclonal antibody (IgG.C7)directed against the cellular LDL receptor.42 Suchblots gave a uniform low background, attesting to thespecificity of the positive blots at Mr«550,000 ob-tained by use of either monoclonal antibodies to

1 2 3 4 5 6 7 8 9 10 1 1 1 2 13

- = 5 5 0 K

I

A BFIGURE 6. Photograph showing determination of apo(a) isoforms by electroimmunoblotting of baboon and marmoset plasmas.Plasma samples (2 fil) were reduced with 10 mM dithiothreitol and electrophoresed in sodium dodecyl sulfate-polyacrylamide(2.75%)-agarose (0.5%) gels (see "Methods"). Protein bands were electrophoreticatty transferred onto nitrocellulose andimmunoblotted initially with a mouse polyclonal antibody to marmoset apo(a) and subsequently with an iodine-125 -labeledanti-mouse immunoglobulin G. Apo(a) bands were finally visualized by autoradiography (see "Methods"). Pattern A, lanes 1-7,individual baboon plasmas; pattern B, lanes 8-13, individual marmoset plasmas, apo(a), apolipoprotein(a).

by guest on Novem

ber 8, 2017http://atvb.ahajournals.org/

Dow

nloaded from

1038 Arteriosclerosis and Thrombosis Vol 11, No 4 July/August 1991

TABLE 2. Comparison of Amino Acid Compositions of Apoli-poprotein(a) and Apoiipoprotein B-100 in the Marmoset andin Humans

1 2

Amino acid

LysHisArgAspThrSerGluProGlyAlaValMetIsoLeuTyrPheCys

Apo(a)

Marmoset

2.24.25.29.7

10.97.7

13.97.07.26.66.71.23.35.56.22.1

ND

Human

2.12.95.88.9

10.47.9

13.210.38.17.46.61.92.45.25.71.4

ND

Apo

Marmoset

7.02.03.2

10.36.78.9

12.73.86.26.65.2

l.i5.4

11.73.14.90.6

B-100

Human

8.02.63.4

10.76.68.6

11.63.84.76.15.61.66.0

11.83.45.10.4

Data are expressed as mol per 100 mol; data for marmoset apo(a)are representative of two analyses. Data in humans for apo(a) aretaken from Reference 26 and for apo B-100 from Reference 43;values for marmoset apo B-100 are derived from our earlier report.16

Apo, apoiipoprotein; ND, not determined.

human apo(a) (Figure 5) or a polyclonal antibody tomarmoset apo(a) (Figure 6).

Chemical Composition of Apolipoprotein(a)The amino acid compositions of apo(a) and apo

B-100 from the marmoset are compared with those ofhumans in Table 2. The amino acid profiles ofmarmoset and human apo(a) are remarkably alikeand may only be distinguished by minor differences(=1 mol %) in histidine, aspartic acid, glycine, ala-nine, and isoleucine. An exception was the propor-tion of proline (10.3% and 7.0% in humans andmarmosets, respectively).

The amino acid compositions of marmoset andhuman apo B-100 were characterized by elevatedproportions of lysine, aspartic and glutamic acids,and leucine; their enrichment in lysine and leucineserved to distinguish them from the homologous formof apo(a). By contrast, apo(a) in marmosets andhumans featured elevated contents of threonine andproline, amino acids that were poorly represented inapo B-100 in these two species.

Immunologic Cross Reactivity of Lipoprotein(a)and Plasminogen

The marked homology of apo(a) and plasminogenin humans and the rhesus monkey has been docu-mented immunochemically and by determination ofcDNA-derived protein sequence.21028-42 To examinethe potential relation between apo(a) and plasmino-gen in the marmoset, we purified the latter protein

- 550 K - — Apo(a)

90 K - I- plasminogenFIGURE 7. Photograph showing immunoreactivity of amonoclonal antibody to human apo(a) with marmoset apo(a)and plasminogen. After reduction by treatment with 10 mMdithiothreitol, marmoset plasma (an aliquot equivalent to 2 figapo[aj protein) was loaded into well 1, and purified marmosetplasminogen (2 /ig protein) was loaded into well 2. Electro-phoresis was then performed in sodium dodecyl sulfate-polyacrylamide (3%)-agarose (0.5%) slab gels, and theproteins were electroblotted onto nitrocellulose. Protein bandswere first reacted with monoclonal antibody 39A1 to (human)apo(a),28 followed by reaction with an iodine-125-labeledanti-mouse immunoglobulin G. Autoradiographic visualiza-tion was performed as described in "Methods." Molecularweights were estimated from a series of marker proteinselectrophoresed in parallel, apo(a), apolipoprotein(a).

from plasma by lysine-Sepharose chromatography.27

Subsequently, marmoset Lp(a) and plasminogenwere electrophoresed in SDS-polyacrylamide gel,and the protein bands were transferred electro-phoretically to nitrocellulose paper and immunoblot-ted (Figure 7). Our mouse monoclonal antibody toapo(a) (39A1) reacted positively with marmosetplasma to reveal a single high A/r band (apo[a]) and inaddition a second component with Mr =90,000 (Fig-ure 7, lane 1). When this same antibody was blottedto the purified homologous plasminogen (Figure 7,lane 2), a single band of Mr =90,000 was identified.Clearly, marmoset apo(a) and plasminogen are im-munochemically related.

Distribution of Plasma Lipoprotein(a) ConcentrationsBy application of an enzyme-linked immunoassay

for marmoset Lp(a) (see "Methods"), plasma levelswere determined in 43 male and 64 female marmo-sets in which the mean concentrations (±SD) were11.0+1.5 and 11.7+1.3 mg/dl, respectively. All indi-viduals possessed immunologically detectable Lp(a)by this procedure. The overall range in Lp(a) levelswas 0.5-49 mg/dl plasma, with a bimodal distribution

by guest on Novem

ber 8, 2017http://atvb.ahajournals.org/

Dow

nloaded from

Guo et al Marmoset Lipoprotein(a) 1039

10

J2CO

co

2

• males (n = 43)• females (n = 64)

10 20 30 40

Lp(a) concentration (mg/dl)

450

FIGURE 8. Frequency distribution ofplasma lipoprotein(a) (Lp[a]) concentrations(mg/dl) in 107 marmosets as determined by amonoclonal antibody-based sandwich en-zyme-linked immunosorbent assay procedure(see "Methods").

of peaks occurring at 3 and at 25 mg/dl, respectively.Five animals were identified with levels in excess of30 mg/dl. No sex differences were detectable.

DiscussionThe present studies have documented for the first

time the presence of a counterpart to human Lp(a) ina New World primate, the common marmoset. Sev-eral explanations may be proposed to account for theapparent absence of Lp(a) in New World monkeys ingeneral,3 and in the common marmoset in particu-lar3,i5,i8; these potentially include the use of humanpolyclonal antisera with poor reactivity to the mon-key lipoprotein, serendipitous selection of marmosetplasmas containing low Lp(a) concentrations, anduse of relatively insensitive immunologic and electro-phoretic detection methods.31518

Lp(a) in fasting marmoset plasma typically dis-played a bimodal density profile (Figure 1), withapproximately half of the total Lp(a) mass beingassociated with triglyceride-rich (d< 1.013 g/ml) andintermediate density (d=1.013-1.016 g/ml) lipopro-teins while the remainder presented as a more typicaldense lipoprotein species distributed over the densityinterval from 1.040-1.080 g/ml. The interaction ofLp(a) with triglyceride-rich lipoproteins and thepresence of apo(a) as a component of postprandiald< 1.006 g/ml lipoproteins in humans are now wellestablished.44-46 By contrast in the baboon, the onlynonhuman primate in which the physicochemicalproperties of Lp(a) have been detailed until now,5-7-9

neither Lp(a) nor apo(a) has been found as a con-stituent of triglyceride-rich particles. An explanationfor this species difference is lacking, especially be-cause apo B-100 is a component of VLDL in thebaboon,47 providing the potential for the interactionof its exposed lysine residues with lysine-binding sitesin baboon apo(a).5-8-10

As in the baboon, rhesus monkey, and hu-man,2-5711 a dense particle species is a significantcomponent of the plasma Lp(a) profile in the mar-moset monkey. The precise dimensions of the densityinterval over which the dense form of marmoset

Lp(a) was primarily distributed {d~ 1.040-1.080g/ml) closely resembled that in humans, baboons,and rhesus monkeys2'5711 and in addition, that in theEuropean hedgehog.26 A major influence on thedensity profile of Lp(a) is exerted by apo(a) isoformpattern.2748'49 In this context it is relevant that weobserved little variation in Lp(a) density profileamong the individual marmosets studied, a findingconsistent with the presence of a unique apo(a)isoform.

In addition to hydrated density, a high degree ofhomology was evident after comparison of otherphysicochemical properties of the dense Lp(a) spe-cies in the marmoset with those of its counterpart inthe baboon and in humans. Thus, the particle sizes ofmarmoset and baboon Lp(a)s as determined by non-denaturing polyacrylamide gradient gel electropho-resis were alike (30.8 nm and 31 nm, respectively) butsomewhat larger than values reported for humanLp(a) (=23.6-25.5 nm2); such apparent differencesmay, however, reflect to some degree the nature ofthe analytical methodology applied, which in the caseof the particles from humans, has primarily con-cerned ultracentrifugal procedures.49 Nonetheless,human, baboon, and marmoset Lp(a) particles areeach of greater diameter than that of the correspond-ing LDL (Figure 2; References 2 and 5). The overallchemical compositions of human, baboon, and mar-moset Lp(a) are alike and equally distinct from thoseof LDL (Table 1; References 5 and 41); in addition,Lp(a) particles in the latter species are all of pre-/3electrophoretic mobility.

The remarkable finding in the current work was ofa unique form of apo(a) in marmoset monkey Lp(a).This feature distinguishes the marmoset monkeyfrom all other species (humans, rhesus monkeys, andbaboons) in which apo(a) isoforms have been evalu-ated to date and in which multiple (as many as 11)forms have been detected.2'7'81119'28'49 The molecularweight of marmoset apo(a) was indistinguishablefrom that of apo B-100 (each derived from marmosetLp[a]) in the various electrophoretic systems used(Figures 3 and 4), and only immunoblotting proce-

by guest on Novem

ber 8, 2017http://atvb.ahajournals.org/

Dow

nloaded from

1040 Arteriosclerosis and Thrombosis Vol 11, No 4 July/August 1991

dures facilitated its identification (Figure 4). Thelinkage between apo(a) and apo B-100 in marmosetLp(a) appears to occur by means of disulfide bond(s),as disulfide reduction dissociated the high molecularweight protein complex (> 800,000) into a single bandof Mr~550,000 (Figure 3); similar behavior has beenreported for Lp(a) in other species.5-25-26'50

Evidence for the identity and nature of marmosetapo(a) in addition to that provided by its immuno-reactivity with polyclonal and monoclonal antisera tohuman apo(a) was derived by chemical analysis (Ta-ble 2); such analyses revealed marmoset apo(a) to beenriched (>10 mol%) in threonine and glutamic acidand deficient (=2 mol% or less) in lysine, methio-nine, and phenylalanine. Indeed, the contents oflysine, arginine, threonine, proline, leucine, tyrosine,and phenylalanine served to distinguish apo(a) fromthe homologous form of apo B-100 (Table 2); similardata have been described for apo(a) and apo B-100 inhumans and the hedgehog. 26>50-51 It is noteworthy thatthe proline content of marmoset apo(a) (7 mol%)resembles that detected in its human counterpart(8.1-11.4 mo^).46-50-51

Further insight into the structural organization ofmarmoset apo(a) was gained from examination of itsrelation with plasminogen. Earlier studies haveclearly established the common possession of krin-gles IV and V, together with a carboxyl-terminalprotease domain, in plasminogen and apo(a) in hu-mans and in the rhesus monkey; hybridization of arhesus monkey apo(a) cDNA probe with baboonapo(a) mRNA attests to a similar degree of homol-ogy between the baboon and rhesus proteins. On thebasis of the cross reactivity of our polyclonal andmonoclonal antibodies to human apo(a) with marmo-set apo(a) and marmoset plasminogen, a similaroverall structural relation clearly exists betweenthem, although protein sequencing data and identi-fication of specific apo(a) epitopes recognized by ourmonoclonal antibodies28 are required before thisrelation may be delineated further.

Finally, it is remarkable that Lp(a) concentrationsin our marmoset colony varied over a 100-fold rangedespite the apparent expression of a unique apo(a)isoform in C. jacchus. Indeed, an inverse relationbetween apo(a) isoform size and plasma Lp(a) con-centration exists in humans.52 More recently, Gavishet al53 have demonstrated that the variable number ofkringle IV domains encoded by the apo(a) gene is theprinciple factor in determining both apo(a) proteinsize and plasma Lp(a) level in humans. By contrast,Rainwater and colleagues7 did not observe an inverserelation between apo(a) isoform size and Lp(a) levelin baboons, and in addition they observed variabilityin the Lp(a) concentrations associated with a specificsingle isoform phenotype. Further interpretation ofthese fascinating—and apparently contradictory—findings may be premature, especially in light of thesignificant and apparently variable contribution ofcarbohydrate content to apo(a) isoform size.19

In conclusion, the detection of Lp(a) in the com-mon marmoset, considered together with the presenceof a single apo(a) isoform and variable plasma Lp(a)levels, makes this New World monkey an exciting andpromising model for future research on the expressionof the apo(a) gene, on the posttranslational processingof apo(a), and on the synthesis, intravascular metab-olism, and cellular degradation of Lp(a).

AcknowledgmentsWe are indebted to G. Debuch (CNRS, Villejuif,

France) for the kind gift of baboon plasmas, to ChantalBernard for technical assistance, and to GeorgesSalmon for careful maintenance of the animals. Theauthors gratefully acknowledge the invaluable help ofVeronique Soulier in manuscript preparation.

References1. Berg K: A new serum type system in man, the Lp system. Ada

Pathol Microbiol Scand 1963;59:369-3822. Utermann G: The mysteries of lipoprotein(a). Science 1989;

246:904-9103. Makino K, Abe A, Maeda S, Noma A, Kawade M, Takenada

O: Lipoprotein(a) in non-human primates: Presence andcharacteristics of Lp(a) immunoreactive materials using anti-human Lp(a) serum. Atherosclerosis 1989;78:81-85

4. Albers JJ, Adolphson JL, Hazzard WR: Radioimmunoassay ofhuman plasma Lp(a) lipoprotein. / Lipid Res 1977;18:331-338

5. Rainwater DL, Manis GS, Kushwaha RS: Characterization ofan unusual lipoprotein similar to human lipoprotein(a) iso-lated from the baboon, Papio sp. Biochim Biophys Ada 1986;877:75-78

6. Mahley RW, Weisgraber KH, Innerarity T, Brewer HB Jr:Characterization of the plasma lipoproteins and apoproteinsof the Erythrocebus patas monkey. Biochemistry 1976;15:1928-1933

7. Rainwater DL, Manis GS, Vandeberg JL: Hereditary anddietary effects on apolipoprotein(a) isoforms and Lp(a) inbaboons. / Lipid Res 1989;30:549-558

8. Hixson JE, Britten ML, Manis GS, Rainwater DL: Apoli-poprotein(a) (apo[a]) glycoprotein isoforms result from sizedifferences in apo(a) mRNA in baboons. / Biol Chem 1989;264:6013-6016

9. Rainwater DL, Manis GS: Immunochemical characterisationof lipoprotein(a) in baboons: Development of an assaydepending on two antigenically distinct proteins. Atherosclero-sis 1988;73:23-31

10. Tomlinson JE, McLean JW, Lawn RM: Rhesus monkeyapolipoprotein(a) sequence, evolution and site of synthesis.J Biol Chem 1989;264:5957-5965

11. Neven L, Khalil A, Pfaffinger D, Fless GM, Jackson E, ScanuAM: Rhesus monkey model of familial hypercholesterolemia:Relation between plasma Lp(a) levels, apo(a) isoforms, andLDL-receptor function. / Lipid Res 1990;31:633-643

12. Hearn JP, Lunn SF, Burden FJ, Pilcher MM: Management ofmarmosets for biomedical research. LabAnim 1975;9:125-134

13. Hearn JP: The common marmoset, in Hearn JP (ed): Repro-duction in New World Primates: New Models in Medical Science.Lancaster, NY, MTP Press, 1982, pp 183-215

14. Hiddleston WA: The production of the common marmoset,Callithrix jacchus, as a laboratory animal, in Chivers DJ,Lane-Petter W (eds): Recent Advances in Primatology. London,Academic Press, Inc, 1978, pp 173-182

15. Chapman MJ, McTaggart F, Goldstein S: Density distribution,characterization and comparative aspects of the major serumlipoproteins in the common marmoset (Callithrix jacchus), aNew World primate with potential use in lipoprotein research.Biochemistry 1979;19:5096-5108

16. Forgez P, Chapman MJ, Mills GL: Isolation, characterizationand comparative aspects of the major serum apolipoproteins,

by guest on Novem

ber 8, 2017http://atvb.ahajournals.org/

Dow

nloaded from

Guo et al Marmoset Lipoprotein(a) 1041

B100 and AI, in the common marmoset, Callithrix jacchus.Biochim Biophys Ada 1983;754:321-333

17. Dreizen S, Levy BM, Bernick S: Diet-induced atherosclerosisin the marmoset. Proc Soc Exp Biol Med 1973;143:1218-1223

18. Crook D, Weisgraber KH, Boyles JK, Mahley RW: Isolationand characterization of the plasma lipoproteins of commonmarmoset monkey: Comparison of effects of control andatherogenic diets. Arteriosclerosis 1990;10:633-647

19. Gaubatz JW, Ghanem KI, Guevara J, Nava ML, Patsch W,Morrisett JD: Polymorphic forms of human apolipoprotein(a):Inheritance and relationship of their molecular weights toplasma levels of lipoprotein(a). / Lipid Res 1990;31:603-613

20. Chapman MJ, Goldstein S, Lagrange D, Laplaud PM: Adensity gradient ultracentrifugal procedure for the isolation ofthe major lipoprotein classes from human serum. / Lipid Res1981;22:339-358

21. Hermier D, Forgez P, Chapman MJ: A density gradient studyof the lipoprotein and apolipoprotein distribution in thechicken, Gallus domesticus. Biochim Biophys Ada 1985;836:105-118

22. Ehnholm C, Garoff H, Simons K, Aro II: Purification andquantitation of the human plasma lipoprotein carrying theLp(a) antigen. Biochim Biophys Ada 1971;236:431-439

23. Radding CM, Steinberg D: Studies on the synthesis andsecretion of serum lipoproteins by rat liver slices. / Clin Invest1960;39:1560-1569

24. Krempler F, Kostner GM, Bolzano K, Sandhofer F: Turnoverof lipoprotein(a) in man. / Clin Invest 1980;65:1483-1490

25. Armstrong VW, Walli AK, Seidel D: Isolation, characteriza-tion and uptake in human fibroblasts of an apo(a)-free lipo-protein obtained on reduction of lipoprotein(a). J Lipid Res1985;26:1314-1323

26. Laplaud PM, Beaubatie L, Rail SC Jr, Luc G, Saboureau M:Lipoprotein(a) is the major apoB-containing lipoprotein in theplasma of a hibernator, the hedgehog (Erinaceus europaeus).J Lipid Res 1988;29:1157-1170

27. Deutsch DG, Mertz ET: Plasminogen-purification fromhuman plasma by affinity chromatography. Science 1970;170:1095-1096

28. Guo HC, Armstrong VW, Luc G, Billardon C, Goulinet S,Nustede R, Seidel D, Chapman MJ: Characterization of fivemouse monoclonal antibodies to apolipoprotein (a) fromhuman Lp(a): Evidence for weak plasminogen reactivity./L(/>idte.s 1989;30:23-37

29. Hui DY, Innerarity TL, Mahley RW: Defective hepatic lipo-protein receptor binding of B very low density lipoproteinsfrom type-Ill hyperlipoproteinemic patients. / Biol Chem1984;259:860-869

30. Lowry OH, Rosebrough NJ, Fair AL, Randall RJ: Proteinmeasurement with the Folin phenol reagent. J Biol Chem1951;193:265-275

31. Noble RR: Electrophoretic separation of plasma lipoproteinsin agarose gel. J Lipid Res 1968;9:673-700

32. Nichols AV, Krauss RM, Musliner TA: Non denaturingpolyacrylamide gradient gel electrophoresis. Methods Enzymol1986;128:417-431

33. Anderson DW, Nichols AV, Forte TM, Lindgren FT: Particledistribution of human serum high density lipoproteins. Bio-chim Biophys Ada 1977;493:55-68

34. Gabelli C, Stark DG, Gregg RE, Brewer HB Jr: Separation ofapolipoprotein B species by agarose-acrylamide gel electro-phoresis. / Lipid Res 1986;27:457-460

35. Laemmli UK: Cleavage of structural proteins during theassembly of the head of bacteriophage T4. Nature 1970;227:680-685

36. Engwall E, Perlmann P: Enzyme-linked immunosorbentassay, ELISA: III. Quantitation of specific antibodies byenzyme-labelled antiimmunoglobulin in antigen-coated-tubes. J Immunol 1972;109:129-135

37. Chapman MJ, Laplaud PM, Luc G, Forgez P, Bruckert R,Goulinet S, Lagrange D: Further resolution of the low densitylipoprotein spectrum in normal human plasma: Physicochem-ical characteristics of discrete subspecies separated by densitygradient ultracentrifugation. / Lipid Res 1988;29:442-458

38. Hawkes R, Niday E, Gordon J: A dot-immunobinding assayfor monoclonal and other antibodies. Anal Biochem 1982;119:142-147

39. Crook D, Weisgraber KH, Rail SC Jr, Mahley RW: Isolationand characterization of several plasma apolipoproteins ofcommon marmoset monkey. Arteriosclerosis 1990;10:625-632

40. Gilmore NJ, Moroz LA: Plasminogen-binding lipoprotein:Isolation and characterisation of a plasma very low densitylipoprotein which cochromatographs with plasminogen onlysine-sepharose. Thromb Res 1983;31:863-874

41. Gaubatz JW, Heideman C, Gotto AM, Morrisett JD, DahlenGH: Human plasma lipoprotein(a): Structural properties.J Biol Chem 1983;258:4582-4589

42. Beisiegel U, Schneider WJ, Goldstein JL, Anderson RGW,Brown MS: Monoclonal antibodies to the low density lipopro-tein receptor as probes for study of receptor-mediated endocy-tosis and the genetics of familial hypercholesterolemia. / BiolChem 1981;256:11923-11931

43. Kane JP, Hardman DA, Paulus HE: Heterogeneity of apolipo-protein B: Isolation of a new species from chylomicrons. ProcNatlAcad SciUSA 1980;77:2465-2469

44. Ye SQ, Trieu YN, Stiers DL, McConathy WJ: Interaction oflow density lipoprotein2 and other apolipoprotein B-contain-ing lipoproteins with lipoprotein (a). / Biol Chem 1988;263:6337-6343

45. Armstrong VW, Zimmeck H, Ruppelt JH, Nustede R, SchleefJ, Seidel D: Quantification of plasma lipoprotein Lp(a) infasting and postprandial lipoproteins of d<1.006 g/ml, inWindier E, Greten JW (eds): Intestinal Lipid and LipoproteinMetabolism. Munich, Zuckschwerdt Verlag, 1989, pp 141-145

46. Bersot TP, Innerarity TL, Pitas RE, Rail SC, Weisgraber KH,Mahley RW: Fat feeding in humans induces lipoproteins ofdensity less than 1.006 that are enriched in apolipoprotein(a)and that cause lipid accumulation in macrophages. / ClinInvest 1986;77:622-630

47. Chapman MJ: Animal lipoproteins: Chemistry, structure andcomparative aspects. / Lipid Res 1980;21:789-853

48. Guo HC, Chapman MJ, Bruckert E, Farriaux JP, de GennesJL: Lipoprotein Lp(a) in homozygous familial hypercholester-olemia: Density profile, particle heterogeneity and apolipopro-tein(a) phenotype. Atherosclerosis 1991;31:69-83

49. Fless GM, Rolih CA, Scanu AM: Heterogeneity of humanplasma lipoprotein(a): Isolation and characterization of thelipoprotein subspecies and their apoproteins. / Biol Chem1984;259:11470-11478

50. Fless GM, Zummallen ME, Scanu AM: Physicochemicalproperties of apolipoprotein(a) and lipoprotein (a-) derivedfrom the dissociation of human plasma lipoprotein (a). J BiolChem 1986;261:8712-8718

51. Gaubatz JW, Chari MV, Nava ML, Guyton JR, Morrisett JD:Isolation and characterization of the two major apoproteins inhuman lipoprotein (a). J Lipid Res 1987;28:69-79

52. Utermann G, Kraft HG, Menzel HJ, Hopferwieser T, Seitz C:Genetics of the quantitative Lp(a) lipoprotein trait: I. Rela-tion of Lp(a) glycoprotein phenotypes to Lp(a) concentrationsin plasma. Hum Genet 1988;78:41-46

53. Gavish D, Azrolan N, Breslow JL: Plasma Lp(a) concentrationis inversely correlated with the ratio of kringle IV/kringle Vencoding domains in the apo(a) gene. / Clin Invest 1989;84:2021-2027

KEY WORDS • apolipoprotein(a) isoform • monoclonalantibodies • immunoblotting • enzyme-linked immunosorbentassay • density gradient ultracentrifugation

by guest on Novem

ber 8, 2017http://atvb.ahajournals.org/

Dow

nloaded from

H C Guo, J B Michel, Y Blouquit and M J Chapmanapolipoprotein(a) isoform.

(Callithrix jacchus). Association of variable plasma lipoprotein(a) levels with a single Lipoprotein(a) and apolipoprotein(a) in a New World monkey, the common marmoset

Print ISSN: 1079-5642. Online ISSN: 1524-4636 Copyright © 1991 American Heart Association, Inc. All rights reserved.

Avenue, Dallas, TX 75231is published by the American Heart Association, 7272 GreenvilleArteriosclerosis, Thrombosis, and Vascular Biology

doi: 10.1161/01.ATV.11.4.10301991;11:1030-1041Arterioscler Thromb Vasc Biol. 

http://atvb.ahajournals.org/content/11/4/1030World Wide Web at:

The online version of this article, along with updated information and services, is located on the

  http://atvb.ahajournals.org//subscriptions/

at: is onlineArteriosclerosis, Thrombosis, and Vascular Biology Information about subscribing to Subscriptions:

  http://www.lww.com/reprints

Information about reprints can be found online at: Reprints: 

document.Permissions and Rights Question and AnswerFurther information about this process is available in theis being requested is located, click Request Permissions in the middle column of the Web page under Services.Clearance Center, not the Editorial Office. Once the online version of the published article for which permission

can be obtained via RightsLink, a service of the CopyrightArteriosclerosis, Thrombosis, and Vascular Biology Requests for permissions to reproduce figures, tables, or portions of articles originally published inPermissions:

by guest on Novem

ber 8, 2017http://atvb.ahajournals.org/

Dow

nloaded from