Dissociation- Enhanced Lanthanide Fluorescence

CLIN.CHEM.38/6, 853-859 (1992)

Dissociation- Enhanced Lanthanide Fluorescence Immunoassay of Lipoprotei n(a) in Serum G. Jurgens,

A. Hermann,

Lipoprotein(a),

D. Aktuna,

and W. Petek

a human serum iipoprotein structuraily

related to low-density iipoprotein (LDL), contains in addition to apoiipoprotein B (apo B) apolipoprotein(a) [apo(a)], a glycoprotein with a strong homoiogy to plasminogen. Lp(a) is a risk factor for coronary heart disease and ischemic cerebrovascuiar disease. Several immunological techniques are used to quantify Lp(a) in human serum, including radioimmunoassays, rocket immunoelectrophoresis, and enzyme-linked immunosorbent assays. Only the last method is suitable for large-scale clinical studies. We describe another solid-phase immunoassay, based on the dissociation-enhanced lanthanide fluorescence system Delfia#{174} (Waliac Oy), and outline the technical details. A polyclonal antiserum directed against Lp(a) was used as the capture antibody. Two kinds of detection antibodies were applied, a polyclonal antiserum against apo B and the polyclonalantiserum against Lp(a).

The resultsagreeexcellentlywiththevaluesestimatedby This assay is easily established, measures Lp(a) in a wideconcentration range,and is suitable for screening large populations. rocket immunoelectrophoresis.

AddItIonalKeyphrasee:apolipoproteins

.

screening

fide bridge with apolipoprotein(a) contains isoforms

specific antigenic of apo(a), differing

[apo(a)J (12), and (13). There are six in molecular mass and

sites

genetics, resulting in Lp(a) glycoprotein phenotypes with different Lp(a) concentrations in plasma (14). Eleven polymorphic forms of apo(a) have been reported by others (15). Furthermore, a partial amino acid sequence of human apo(a) is homologous to plasminogen (16, 17), but apo(a) is not capable of cleaving fibrin (17). Amidolytic activity is associated with apo(a) (18, 19), but fibronectin acts as the natural substrate (19). In the search for the biochemical mechanisms responsible for the atherogenicity of Lp(a), some have suggested that Lp(a) may act as an interloper in the fibrinolytic system (20,21). Lp(a) possibly competes with the tissue plasminogen-activating factor for the binding sites on fibrin, yet is not able to cleave these sites or destabilize thrombi (22). Lp(a) also inhibits the activation of plasminogen, mediated either by streptokinase (23) or tissue plasminogen activator (24). In many of the clinical studies cited, Lp(a) was quantitatively determined by rocket immunoelectrophoresis. Lp(a) has also been quantified by several other techniques: radioimmunoassay (8, 25), enzyme-linked immunosorbent assay (26, 27), nephelometry (28), zonal imniunoassay (29), and radial immunodiffusion (5). Screening for Lp(a) is done with electrophoresis (30). A comparison of the characteristics of the immunological assays for Lp(a) was published recently (31). Here we present an Eu3-based fluorescence immunoassay (EuFIA) of Lp(a) used in the Delfia#{174} system. This is a quick, accurate, and reliable assay of normal and pathological concentrations of Lp(a) in large-scale studies.

the past two decades, many studies have that lipoprotein(a) [Lp(a)] is an important independent risk factor for atherosclerosis.1 Increased serum concentrations of Lp(a) correlate positively with angiographically documented coronary heart disease (1, 2) and indicate high risk for myocardial infarction, even at a young age (3). Moreover, individuals with increased concentrations of Lp(a) have a higher risk of ischemic cerebrovascular diseases (4-7). Lp(a) concentration is also a predictor of vein-graft stenosis after coronary artery bypass surgery; Lp(a) was detected immunochemically in coronary artery bypass vein grafts resected at re-operation (8, 9). Therapy with hydroxymethylglutaryl-CoA reductase inhibitors such as lovastatin or simvastatin has in some cases led to an increase in concentrations of Lp(a) in serum (10, 11). Monitoring Lp(a) concentrations during this therapy is

fluorometry was described in a review by HemmilA (32). Buffer reagents: Elution buffer, pH 7.7; per liter: Ti-is HC1, 50 mmol; sodium chloride, 9 g; sodium azide,

recommended.

0.5 g.

During

shown

Lp(a) is a low-density comprising

apolipoprotein

lipoprotein

(LDL)-like particle, B (ape B) linked by a disul-

Institute of Medical Biochemistry, Karl-Franzens Universitat Graz, Harrachgasse 21, A-8010 Graz, Austria. ‘Nonstandard abbreviations: Lp(a), lipoprotein(a); apo(a), apolipoprotein(a); LDL, low-density lipoprotein; ape B, apolipoprotein B; Eu-FIA, Eu3-based fluorescence immunoassay; and Eu3’DTPA, Eu3 chelate of W-(p-isothiocyanatobenzyl)-diethylenetri. amine-N’,N2,N-tetraacetic acid. Received July 29, 1991; accepted December 11, 1991.

Materials and Methods FluorescenceImmunoassay The principle of time-resolved

-

Assay buffer, pH 7.75; per liter: Tris HC1, 50 mmol; BSA, 5 g; bovine ‘y-globulin, 0.5 g; diethylenetriamine pentaacetic acid, 20 mol; sodium azide, 0.5 g. Wash buffer, pH 7.75; per liter of doubly distilled water: Tris (Sigma Co., Munich, F.R.G.), 0.6 g; Tween 20 (Sigma), 0.05 mL; sodium chloride, 0.9 g; sodium azide, 0.5 g; pH adjusted by adding HC1. Labeling buffer, pH 9.6; per liter: sodium carbonate, 1.58 g; sodium hydrogen carbonate, 2.92 g. Coating buffer, pH 9.6; per liter: sodium carbonate, -

CLINICALCHEMISTRY, Vol.38, No. 6, 1992 853

1.58 g; sodium hydrogen

carbonate,

2.92 g; sodium

azide, 0.2 g. Labeling the

antibody: A commercially available Delfia Eu-labeling kit (cat. no. 1244-302; Wallac Oy, Thrku, Finland) was used for labeling antibodies to Lp(a) and LDL (ape B) according to the manufacturer’s recommendations. We isolated the immunoglobulin

fraction of the rabbit antisera by diethylaniinoethylcellulose chromatography, followed by dialysis overnight against labeling buffer. We adjusted the protein concentration of the immunoglobulun fraction to 4 g/L by diluting with the labeling buffer (10-fold) and then measured the absorbance at 280 nm. The Eu3 chelate of N’-(p-isothiocyanatobenzyl)-diethylenetriamine-N’, N2,N3’3-tetraacetic acid (Eu3-DTFA) is used as the labeling reagent in this kit (33). Another Eu3 chelator, 4,7-bis(chlorosulfophenyl)-1,10 phenanthroline-2,9-dicarboxylic acid, described as an alternative label for time-resolved fluorometric applications (34), was not used here because the appropriate instrumentation for laser-excited solid-phase time-resolved fluorometric measurements (Cyber Fluor” Immunoanalyzer; 35) was not available. We labeled 1 mg of the immunoglobulin (250 pL) with 0.2 mg of Eu3-DT’FA dissolved in 250 p1 of the labeling buffer. We kept the solutions at room temperature for 20 h and then separated the Eu3-labeled immunoglobulin fraction from excess labeling reagent by gel ifitration on a 30 x 0.9 cm column packed with Sephadex G25 Superfine (Pharmacia Fine Chemicals AB, Uppsala, Sweden). Before use, the column was rinsed with 70 mL of the elution buffer. On this column, the monomers of the Eu3-1abe1ed iminunoglobulun fraction eluted between 8 and 13 mL. We monitored the absorbance of the eluate at 280 nm and pooled the peak fractions. To stabilize the labeled immunoglobulin fraction, we added a bovine serum albumiii solution (7.5 g/L) containing sodium azide (provided in the kit) to the antibody solution (1/75, by vol). Stored at 4#{176}C, the labeled antibodies could be used for at least a year. Measuring fluorescence: We measured fluorescence with a Model 1234 Delfia research fluorometer (Wallac Oy), which displays fluorescence intensity in counts per second. Samples were applied to a microtiter strip (Nunc Maxisorb, 8 x 12 wells, cat. no. 473709; Nunc, Roskilde, Denmark) at 200 pL/well (in duplicate) after 10000-fold dilution with enhancement solution; we compared the fluorescence of the sample with that of the Eu3 standard (1 nmol/L; stock standard, diluted 100-fold with the enhancement solution). The enhancement solution (Waliac Oy) contained 1 g of Triton X-100, 6.8 mmol of potassium hydrogen phthalate, 100 mmol of acetic acid, 50 pmol of tri-n-octylphosphine oxide, and 15 pxnol of 2-naphthoyltrifluoroacetone per liter of doubly distilled water. Calculating labeling efficiency: We calculated the concentration of Eu3 of the immunoglobulin solution as follows: 854 CLINICALCHEMISTRY,Vol.38, No.6, 1992

Eu3

(pniolJL)

=

Eu

counts x 10

counts for 1 nmol/L Eu

Eu

std.

The protein content of the immunoglobulin fraction was measured at 280 nm and calculated after subtracting the absorbance of the aromatic thiourea bonds (0.008 A per 1 pznol/L thiourea concentration): Protein

(1mol/L)

Yield (Eu3/IgG)

A,,, =

=

-

0.008 x Eu3 (pmol/L) 1.34

1000 X -

Eu3 (punol/L)/protein (pmol/L)

We calculated that 21 mol of Eu3 were attached to 1 mol of IgG in the anti-Lp(a) antibody and that 17 mol of Eu3 were attached to 1 mol of IgG in the anti-ape B antibody. These values are in the recommended range of 5-25 mol of Eu3 per mole of IgG. Coating the microtiter plates with antibody: We set up a two-step sandwich assay, using 1.25 ig of the capture antibody per well for coating the microtiter plates. The concentration of the antibody was 3.06 g of rabbit polyvalent immunoglobulins per liter. We diluted 80 p1 with 20 mL of the coating buffer, which was suffIcient to coat one microtiter plate (200 pL/weil). We incubated the plates for 14-18 h at room temperature and then washed them twice with wash buffer (250 p1/well). An additional blocking reagent (bovine serum albumin in sorbitol) was not needed, perhaps because the assay buffer used for the 10000-fold dilution of the serum samples contained bovine serum albumin and y-globuun. Applying the serum samples: We diluted serum samples and standards 10000-fold with the assay buffer with a diluter-dispenser. That is, we diluted 10 p1 of serum sample with 990 p1 of the assay buffer, then further diluted 10 p1 of this solution with 990 p1 of the same buffer and applied in duplicate 200 p1 of the diluted sample (containing 20 nL of the original serum) to each well. After shaking the microtiter plate containing the samples for 90 mm at room temperature, we washed the plate three times with wash buffer (250 ALfwe1l). Applying the Ew-labekd antibodies and measuring the fluorescence: We added 50 ng of the detection antibody [anti-Lp(a) or anti-ape B] in 200 p1 of the assay buffer to each well. After the wells were shaken for 60 miii at room temperature, we washed them six times with wash buffer (250 p1/well). As the final step, we

added the enhancement solution (200 pL/well), mixed the contents by shaking for 5 mm, and then incubated the samples for 10 miii (all at room temperature). The fluorescence of the wells was measured with a Delfia research fluorometer. We performed calculations with the MuLTICALC immunoassay data management program (Pharmacia; Wallac Oy) installed on a personal computer (MBC-27 MT1; Sanyo Electric Co. Ltd., Tokyo, Japan). Standardization: We used a commercially available

Lp(a) standard serum (lot no. 2900/090; Immuno AG, Vienna, Austria) with an Lp(a) concentration of 587 mgfL and containing the Si and S3 isoforms; 20 p1 was diluted with 960 p1 of the assay buffer to give an Lp(a) concentration of 12 mg/L. Additional twofold dilutions with the assay buffer gave Lp(a) concentrations of 6, 3, 1.5, 0.75, 0.375, and 0.1875 mg/L. These samples were further diluted by using 10 p1 of the solution and 990 p1 of the assay buffer, giving Lp(a) concentrations of 1200, 600, 300, 150, 75, 37.5, and 18.75 mg/L in a 10000-fold dilution. Linearity range: We diluted 50 p1 of the standard serum with 930 p1 of the assay buffer to give a concentration of 30 mg/L. Further dilutions with the assay buffer gave Lp(a) concentrations of 2500, 2000, 1500, 1000, 500, 250, 125, and 62.5 mg/L in a 10 000-fold dilution. Detection limit: For evaluating the detection limit, we estimated the value of the blank 10 times and assumed that a value threefold greater than the standard deviation from the mean could be measured confidently. Rocket Immunoelectrophoresis We poured 6 mL of 10 g/L agarose (Bio-Rad Labs., Richmond, CA) preparation containing anti-Lp(a) antiserum, 10 g/L, onto 70 x 70 mm glass plates. After the gel became rigid, we made 12 holes in it and applied 4 p1 of a serum sample to each. The electrophoresis was performed for 16 h at 1.6 V/cm with an LKB-2117 Multiphor II electrophoresis unit (Pharmacia LKB Biotechnology AB, Bromma, Sweden) supplied by an LKB2303 Multidrive XL 3.5 kV power supply. After the run, we soaked the gels for 2 h in saline, pressure-dried them on filter paper, and stained them with Serva-Blue G-250 (Serva Feinbiochemika GmbH & Co., Heidelberg, FRG). We applied the standard serum and five dilutions of it (1 + 0.5,1 + 1,1+2,1 +3,andi +6,byvol)toeachplate along with the serum samples being tested. A linear range of responses was achieved for serum samples containing Lp(a) concentrations of 84-580 mg/L. For the serum samples with an Lp(a) concentration 500 mg/L were diluted with electrophoresis buffer (10 g of Tris, 2.8 g of boric acid, and 0.65 g of EDTA per liter) before reevaluation with rocket immunoelectrophoresis. We calculated the concentrations of Lp(a) from the peak areas (height x 0.5 width) by extrapolating from the standard curve. We compared the mean value of two estimates on two different gel plates with the Eu-FIA result for each serum sample. Serum Samples The serum samples were selected without conscious bias from a consecutive series of patients attending the Department for Vascular Surgery of the University of Graz Medical School as part of a routine screening program. Although the serum samples were analyzed within four days after blood collection, we also investi-

gated the stability of the samples stored at 4 or -20 #{176}C. We stored pooled serum for three months under nonsterile conditions and measured the Lp(a) content monthly, with the following results: 475, 475, 480, and 490 mg/L for storage at 4#{176}C, and 459, 461, 462, and 453 mg/L for storage at -20 #{176}C. Antiserum We prepared a polyvalent anti-Lp(a) antiserum from rabbit (4, 6-8), which we applied along with anti-ape B antiserum from Behrung AG (Marburg, F.R.G.). The crude anti-Lp(a) antiserum cross-reacted with ape B and with plasminogen. The antibodies against ape B and plasminogen were removed from the immune serum by absorption with plasma from an Lp(a)-negative (1000 mgfL (38). For the Eu-FIA, we Table 2. Accuracy of Eu-F1A Recevery, % Sample type

Initial Lp(a) concn. mg/L

Rang.

Anfi-Lp(a) captu re antibody, anti-apo B detection antibody Serum 0.0 97.2-104.3 Serum 83.7 96.7-101.3 Serum 209.0 98.9-103.2 Serum 369.2 95.5-98.0

Serum F isoform Mean

394.4 256.9

98.6-100.4

98.9-99.8

Mean

100.75 99.00 101 05

96.75 99.50 99.35

95.3-104.3 Anti-Lp(a) capture and detection antibodyb

9960

Serum

97.55

0.0

93.8-101.3

114.7 100.1-106.7 103.40 Serum 208.7 97.5-1 02.9 100.24 Serum 101 35 364.8 98.0-i 04.7 Serum 384.6 99.8-106.0 102.90 Serum F isoform 97.55 205.2 94.4-100.7 93.8-106.7 Mean 100.66 aF isoform of Lp(a) added at 57, 114, 228, and 456 mg/L;n = 6. 5F Isoform of Lp(a) added at 55.5, 105, 210, and 420 mg/L; n = 6. 858

Interassay#{176}

Intraassayb

CLINICALCHEMISTRY,Vol.38, No. 6, 1992

recommend using the anti-Lp(a) antiserum as the capture antibody and the anti-apo B as the detection antibody. This combination agreed best with the values obtained by rocket immunoelectrophoresis in the diagnostically important range of 150-350 mgfL. In addition, the potential cross-reactivity of anti-Lp(a) antiserum with plasminogen can be eliminated as previously recommended (27). We dedicate this article to Professor Dr. Anton Holasek on the occasion of his 70th birthday and in recognition of his merits and continuous efforts to stimulate research in clinical chemistry in Austria. The technical assistance of Andreas Meimtzer, Gerhard Ledinski, and Joachim Greilberger is appreciated. This work was supported by the Jubil#{225}umsfonds der Osterreichischen National-

bank project no. 3696. References 1. Dahlen Gil, Guyton JR, Attar M, Farmer JA, Kautz JA, Gotto AM. Association of levels of lipoprotein Lp(a), plasma lipids, and other lipoproteins with coronary artery disease documented by angiography. Circulation 1986;74:758-65. 2. Armstrong VW, Cremer P, Eberle E, et al. The association between serum Lp(a) concentrations and angiographically assessed coronary atherosclerosis-dependence on serum LDL levels.

Atherosclerosis 1986;62:249-57. 3. Sandkaxnp M, Funke H, Schulte H, Kohier E, Asamann G. Lipoprotein(a) is an independent risk factor for myocardial infarction at a young age. Clin Chem 1990;36:20-3. 4. Koltringer P, Jurgens G. A dominant role of lipoprotein(a) in the investigation and evaluation of parameters indicating the development of cervical atherosclerosis. Atherosclerosis 1985;58: 187-98. 5. Murai A, Miyahara T, Fujimoto KN, Matsuda M, Kaxneyama M. Lp(a) lipoprotein as a risk factor for coronary heart disease and cerebral infarction. Atherosclerosis 1986;59:199-204. 6. Zenker G, KOltringer Jurgens G. Lipoprotein(a)

P, Bone G, Niederkorn

K, Pfeiffer

K,

as a strong indicator for cerebrovascular disease. Stroke 1986;17:942-5. 7. Jurgens G, KOltringer P. Lipoprotein(a) in ischemic cerebrovascular disease: a new approach to the assessment of risk for stroke. Neurology 8. HoffHF,1987;37:513-5. Beck (]J, Skibinski CIII, et at. Serum Lp(a) level as a predictor of vein graft stenosis after coronary artery bypass surgery in patients. Circulation 1988;77:1238-44. 9. Cushing GL, Gaubatz JW, Nava ML, et al. Quantitation and localization of apolipoproteins(a) and B in coronary artery bypass vein grafts resected after re-operation. Arteriosclerosis 1989;9: 593-603. 10. Jurgens G, Ashy A, Zenker G. Raised serum lipoprotein during treatment with lovastatin. Lancet 1989i:911-2. 11. Koatner GM,Gavish 0, Leopold B, Boizano K, Weintraub MS, Breslow JL. HMG CoA reductase inhibitors lower LDL cholesterol without reducing Lp(a) levels. Circulation 1989;80:1313-9. 12. Gaubatz JW, Heideman C, Gotto AM Jr, Morrisett JD, Dahien GH. Human plasma lipoprotein(a). Structural properties. J Biol

Chem 1983;258:4582-9.

13. Jurgens G, Kostner GM. Studies on the structure

lipoprotein.

Isolation and partial characterization specific antigen. Immunogenetica 1975;1:560-9.

of the Lp(a) of the Lp(a)

14. Utermann G, Kraft HG, Menzel HJ, Hopfwieser T, Seitz C. Genetics of the quantitative Lp(a) lipoprotein trait. I. Relationship of Lp(a) glycoprotein phenotypes to Lp(a) lipoprotein concentrations in plasma. Hum Genet 1988;78:41-6. 15. Gaubatz JW, Ghanem KI, Guevara J Jr, Nava ML, Patsch W, Morrisett JD. Polymorphic forms of human apolipoprotein(a): inheritance and relationship of their molecular weights to plasma levels of lipoprotein(a). J Lipid Res 1990;31:603-13. 16. Eaton DL, Fleas GM, Kohr WJ, et at. Partial amino acid sequence of apoprotein(a) shows that it is homologous to plasminogen. Proc Nati Aced Sci USA 1987;84:3224-8. 17. McLean JW, Tomlinson JE, Kuang WJ, et at. cDNA sequence of human apolipoprotein(a) is homologous to plasminogen. Nature 1987;330:132-7.

18. Jurgens G, Marth E, KoetnerGM,Holasek A. Investigation of the Lp(a) lipoprotein aggregating and enzymatic activity associated with the Lp(a) polypeptide. Artery 1977;3:13-26. 19. Salonen E-M, Jauhiainen M, Zardi L, Vaheri A, Ehnholm C. Lipoprotein(a) binds to fibronectin and has serum proteinase activity capable of cleaving it. EMBO J 1989;8:4035-40. 20. Scanu AM. Lp(a) a potential bridge between the fields of atherosclerosis and thrombosis. Arch Pathol Lab Med 1988;112: 1045-7. 21. MEewu AD, Durrington PN. Lipoprotein(a): structure, properties and possible involvement in thrombogenesis and atherogenesis. Atherosclerosis 1990;85:1-14. 22. Loscalzo J, Weinfeld M, Fleas GM, Scanu AM. Lipoprotein(a), fibrin binding, 10:240-5.

and plasminogen

activation.

Arteriosclerosis

1990;

23. Edelberg JM, Gonzales-Gronow M, Pizzo SV. Lipoprotein(a) inhibits streptokinase-mediated activation of human plasminogen. Biochemistry 1989;28:2370-4. 24. Edelberg JM, Gonzales-Gronow M, Pizzo SV. Lipoprotein(a) inhibition of plasminogen activation by tissue-type plasminogen activator. Thromb Res 1990;57:155-62. 25. Albers JJ, Adolphson JL, Hazzard WR. Radioimmunoasaay of human plasma Lp(a) lipoprotein. J Lipid Res 1977;18:331-8. 26. Abe A, Maeda 5, Makino K. Enzyme linked immunoadsorbent

assay of lipoprotein(a) in serum and cord blood. Cliii Chim Acta 1988;177:31-40. 27. Fleas GM,Snyder ML, Scanu AM. Enzyme-linked immunoassay for Lp(a). J Lipid Res 1989;30:651-62. 28. Cazzolato G, Prakash G, GreenS, Kostner GM. The determination of lipoprotein Lp(a) by rate and end point nephelometry. Clin Clam Acta 1983;135:203-8. 29. MSrz W, Groil W. Quantification of human serum lipoprotein Lp(a): zone immunoelectrophoresis assay, a new sensitive method as compared to electroimmuno assay. Clin Chim Acta 1983;134: 265-79.

30. Bruckert E, Truffert J, De Gennes JL, et al. Does electrophoresis reliably screen for high serum Lp(a)? Cliii Chim Acta 1990; 188:71-8. 31. Labeur C, Shepherd J, Roaaeneu M. Immunological assays of apolipoproteina in plasma: methods and instrumentation. Clin Chem 1990;36:591-7. 32. HemmilA I. Lanthanides as probes for time-resolved fluorometric immunoasaays. Scand J Cliii Lab Invest 1988;48:389-400. 33. Mukkala V-M, Mikola H, Hemmilfi I. The synthesis and use of activated N-benzyl derivatives of diethylenetriazninetetraacetic acids: alternative reagents for labeling of antibodies with metal

ions. Anal Biochem 1989;176:319-25. 34. Evangelista RA, Pollak A, Allore B, Templeton

EF, Morton

RC, Diamandis EP. A new europium chelate for protein labelling and time-resolved fluorometric applications. Clin Biochem 1988; 21:173-8. 35. Diamandia EP, Bhayana V, Conway K, Reichstein E, Papanastasiou-Diamandis. Time-resolved fluoroimmunoassay of cortisol in serum with a europium chelate as label. Clin Biochem 1988;21:291-6. 36. Utermann G. The mysteries of lipoprotein(a). Science 1989; 246:904-10. 37. Albers JJ, Marcovina SM, Lodge MS. The unique lipoprotein(a): properties and immunochemical measurement [Review].

Clin Chem 1990;36:2019-26. 38. Wong WLT, Eaton DL, Berloui A, Fendly B, Hass PE. A monoclonal-antibody-based enzyme-linked ixnmunosorbent assay of lipoprotein(a). Clin Chem 1990;36:192-7.

CLINICAL CHEMISTRY, Vol. 38, No. 6, 1992

859