CONTENTS

20 B I O C H E M I C A ■ N O . 4 [ 1 9 9 6 ]

F E A T U R E S

Development of a Digoxigenin-labeledPeptide: Application to a

Chemiluminoenzyme Immunoassay ofBradykinin in Inflamed Tissues

IntroductionImmunological techniques have partially satisfied the increasing need for specific,

sensitive methods for detecting and quantifying biologically active peptides in bodyfluids and tissues. For example, measurement of peptide levels in blood has beengreatly simplified by radioimmunoassays (RIA) and enzyme immunoassays (EIA) forpeptides. While the sensitivity level of these peptide immunoassays has been some-what increased by using avidin-biotin technology for two-site competitive EIA (2–6),we propose an alternative for competitive EIA of peptides – one that employs digoxi-genin-labeled peptides as a tracer. Although digoxigenin (DIG) has been broadly usedto label proteins and cDNA probes (7), here we apply it to the labeling of a tracer pep-tide for the first time.

Bradykinin (BK) is one of a group of powerful proinflammatory autacoids and isbelieved to play a role in acute and chronic inflammatory processes. Previous methodsof quantifying bradykinin in biological samples have been hindered by the enzymaticlability and the low concentration of BK in tissues and body fluids (8). The digoxi-genin-labeled bradykinin tracer has been purified, characterized, and used for thedevelopment of a heterogeneous competitive EIA using polyclonal anti-BK antibodiesfor the immunoconcentration step and alkaline phosphatase-conjugated anti-DIG Fabfragments as an intermediate reaction. Immune complexes are then detected usingdioxetane derivatives as substrates for alkaline phosphatase. Previously, these chemilu-minescent substrates had been almost exclusively applied to molecular biology (9) andsandwich EIA methods (10). After analytical validation, this method has been appliedto extracts of normal and carrageenan-inflamed tissues.

ANICK DÉCARIE1, GUY DRAPEAU2, JEAN CLOSSET3, RÉJEAN COUTURE1, AND ALBERT ADAM4

1 Département de Physiologie, Faculté de Médecine, 4Faculté de Pharmacie, Université de Montréal,Montréal, Québec, Canada, H3C 3J7

2 Centre de Recherche, Hôtel-Dieu de Québec, Québec, Canada G1R 2J63 Laboratoire d’endocrinologie, CHU, Université de Liège, 4000 Liège, BelgiqueReprints requests should be sent to:Dr. Albert Adam, Faculté de Pharmacie, Université de Montréal, C.P. 6128, Succursale A, Montréal,Québec, Canada, H3C 3J7. Email: adama@ere.umontreal.ca

Editor’s note: This is a modified version of an article originally published in Peptides. Reprinted by permission of the publisher from “Development of digoxigenin-labeled peptide: Application to chemilu-minescence enzyme immunoassay of bradykinin in inflamed tissue.” Anick Décarie, Guy Drapeau,Jean Closset, Réjean Couture, and Albert Adam; Peptides 15(3):511– 518.. Copyright 1994 by ElsevierScience Inc. This Biochemica article focuses primarily on the generation and application of digoxigenin-labeled peptides. For more-detailed descriptions of the biological effect of bradykinin in inflamed tissueand the influence of kininase inhibitors, please refer to the original Peptides article (1).

We have developed an ultrasensitivechemiluminoenzyme immunoassay(CLEIA) using digoxigenin-labeled bradykinin as a tracer for the quanti-fication of kinins in tissue samples.Rabbit polyclonal IgGs directed againstthe C-terminal end of bradykinin wereused for the immunoconcentration stepalong with a dioxetane derivative alkaline phosphatase substrate for therevelation step. This sensitive assaycould detect bradykinin levels as low as 0.1 fmol/ml with ED50 of 0.78 pmol/mlin extracts of normal and carrageenan-inflamed tissues.

CONTENTS

the BK standard or 50 µl diluted biologi-cal sample and 50 µl DIG-BK. The plateswere incubated for 16 h at 4°C underagitation to allow the immunologicalreaction to occur. After removal ofunbound material by a careful washcycle, the bound DIG-BK was reactedwith alkaline phosphatase-conjugatedanti-DIG Fab fragments for 2 h at 37°C.Finally, 100 µl of a chemiluminescentalkaline phosphatase substrate was in-cubated in the plate wells for 30 min at37°C to measure alkaline phosphataseactivity and reveal the presence ofimmune complexes. The intensity oflight emission was measured at 540 nmon a luminometer, and the results wereexpressed in Relative Light Units (RLU).

Application of DIG-BK in an animal model

The preparation of an animal modelof acute inflammation, tissue samplingand processing, immunoreactivity pro-files of the inflamed tissue extract, andstatistical data analysis are beyond thescope of this Biochemica article and wereperformed as described in the originalPeptides paper (1).

ResultsCharacterization of the DIG-BK

Mass spectrometry of DIG-BK corrob-orated the calculated molecular weight(1603.4 g). The DIG-labeled BK corre-sponds to the structure represented inFigure 2. The immunoreactivity of theDIG-BK derivative was tested by radio-immunoassay using successive dilutions(from 100 to 0.004 pmol/ml) of thisderivative (Figure 3). The calibrationcurve for BK and the displacement curveobtained with DIG-BK display ED50s of1.60 and 1.67 pmol/ml, respectively.Moreover, the slopes of these curves werenot statistically different from each other(F[1,28] = 2.95, p>0.05) with values of1.01 and 0.93, respectively.

21B I O C H E M I C A ■ N O . 4 [ 1 9 9 6 ]

F E A T U R E S

Materials and MethodsPreparation and purification of antibodies

Antibodies against BK were producedin albinos rabbits (Charles River, St.Constant, Québec, Canada) immunizedwith BK that had been covalently linkedto bovine serum albumin by the glutar-aldehyde method (11). After four boost-ings, we obtained antiserum exhibiting atiter of 1/150,000 when assayed in RIA(30% binding). For the EIA method, IgGswere purified by affinity chromatographyfor protein G. The concentration of thepurified material was measured by itsabsorption at 278 nm. The purity of thismaterial was verified using the A278/252nm ratio (12).

Preparation of the DIG-labeledbradykinin (DIG-BK)

Five milligrams of Digoxigenin-3-0-methylcarbonyl-ε-aminocaproic acid-N-hydroxysuccinimide ester (DIG-NHSester) was added to 16 mg bradykinindissolved in 5 ml of DMF (pH adjusted to9.0 with N-methylmorpholine). Afteradding 1.5 mg 1-hydroxybenzotriazoleto the mixture, the reaction was moni-tored by analytical HPLC as describedbelow. After 5 h, the reaction mixturewas loaded on a preparative Michel-Miller column (22 X 130 mm) packed

with a Vydac™ C18 15–20 µm resin pre-equilibrated with 0.05% trifluoroaceticacid (TFA)/water. The incubation mix-ture was separated using a gradient of5–70% acetonitrile/water containing0.05% TFA; a total volume of 600 ml elu-ent was pumped through the column at arate of 6.5 ml/min and collected in6–8 ml fractions. Fractions containingthe pure product (DIG-BK) were pooledand lyophilized. Purity of the final prod-uct was evaluated by analytical HPLC,and its identity was confirmed by FABmass spectrometry.

Separations were achieved with aVydac 10 µm (3.9 X 300 mm) reversephase C18 column using a linear gradientof 5–65% acetonitrile/0.05% TFA/waterat 2 ml/min over a period of 20 min. Theabsorbance was measured at 214 nm.

Immunoreactivity of DIG-BKImmunoreactivity of DIG-labeled BK

was tested by radioimmunoassays usingsuccessive dilutions of this synthesizedmaterial. The incubation medium (0.5 ml)contained standard BK (500 pmol–5 fmol /ml) or DIG-BK dilutions (100 µl),plus 100 µl [3H]-BK (10,000 dpm), 100 µl antiserum (final dilution,1/150,000), and 200 µl assay buffer (200 mM Tris-HCl, pH 7.4, containing 2 g/l lysozyme). After an overnight in-cubation at 4°C, the separation of boundfrom free fraction was achieved by ad-ding 500 µl of a charcoal mixture (1 g/lcharcoal, 0.1 g/l dextran T70 in water).The tubes were then vortexed, allowed tostand for 15 min at 10°C, and centri-fuged at 3,000 rpm for 15 min at 4°C.The radioactivity of the supernatant wasthen counted in a beta counter.

CLEIA of BKFigure 1 summarizes the different

steps of this enzyme immunoassay.Unless otherwise indicated, all washing,dilution, and incubation steps were per-formed in the same buffer: 50 mM Tris-HCl (pH 7.4) containing 100 mM NaCland 0.5 ml/l Tween®20.

Microtiter plates (96-well) were coat-ed with 100 µl polyclonal anti-BK IgG,diluted in 100 µM carbonate coatingbuffer (pH 9.5), for 24 h at 4°C. Theplates were washed and saturated withincubation buffer containing 5 g/l gelatinfor 2 h at 37°C. After another washingstep, the plates were incubated with 100 µl of a mixture containing 50 µl of

Schematic representation of the dif-ferent steps of chemiluminescent enzymeimmunoassay: 1-Addition of BK ( ) and BK-DIG towells coated with polyclonal anti-BK IgG; 2-Competition between BK and BK-DIG for polyclon-al anti-BK IgG (–⟨ ); 3-Bound DIG-BK is let to reactwith anti-DIG-labeled Fab fragments coupled withalkaline phosphatase (ALP); 4-Immune complexesare revealed by measuring the ALP activity with achemiluminescent alkaline phosphatase substrate.Intensity of light emission is measured at 540 nm.

Figure 1

Schematic structure of DIG-labeled BK.Figure 2

chemiluminescent alkalinephosphatase substrate

CONTENTS

22 B I O C H E M I C A ■ N O . 4 [ 1 9 9 6 ]

F E A T U R E S

Analytical characteristics of CLEIA for BK

Figure 4 shows a typical calibrationcurve obtained in the conditions opti-mized by coating IgG at a concentrationof 2.5 µg/ml and incubating a fixedamount (0.61 pmol/ml) of DIG-labeledBK with increasing concentrations of BKranging from 0.023 to 23.8 pmol/ml. Thecalibration curve is characterized by anED50 of 0.78 pmol/ml. The minimal

detectable concentration of BK, calcu-lated as the light emission of the zeromean concentration (597 RLU) minus 2standard deviations (0.48 RLU), was0.1 fmol/ml (15).

Intra- and inter-assay precision wasassessed on blank (non-inflamed) tissueextract spiked with BK at concentrationsof 6.0, 1.5, and 0.35 pmol/ml in tripli-cate. The samples were analyzed on fivedifferent plates. The intra- and inter-assaycoefficients of variation are provided inTable 1.

Tissue content and biological effect ofBK and influence of kininase inhibitors

As this Biochemica article focuses solelyon the development of the sensitivechemiluminoenzyme immunoassay usingDIG-labeled bradykinin, more in-depthinvestigations into the role of BK and kininase inhibitors can be found in the

original Peptides article (1). To brieflysummarize the results detailed there, thissensitive chemiluminoenzyme immuno-assay was capable of detecting a 7-foldincrease in immunoreactive kinins in carrageenan-inflamed tissues. It also per-mitted the detection of significantlyenhanced kinin tissue content when a mix-ture of inhibitors of kininase I (mergepta)and kininase II (captopril) was coinjectedwith carrageenan. Thus, this assay providesbiochemical evidence that kinins may act aspro-inflammatory mediators, and it high-lights a compensatory increase of kininase Iand II activities in inflamed tissues.

DiscussionThis article describes the use of DIG-

labeled peptides as tracers for com-petitive immunoassays. Here we usedDIG-labeled bradykinin to apply thisnew analytical approach to the measure-ment of bradykinin in tissue. Becausedigoxigenin occurs exclusively in Digi-talis plants, this bioanalytical indicatorsystem prevents non-specific reactionswith endogenous substances found inanimal materials (7). This property isparticularly advantageous when com-pared with biotin, which is present invarious mammalian tissues (16, 17).

Different digoxigenin derivatives arecommercially available for the labeling ofproteins and cDNA. Depending on thenature of the probe to be identified orquantified, either hydrocarbon or amino-terminal groups of proteins can be usedas coupling residues. In our example, theDIG-NHS ester was bound specifically tothe amino-terminal arginine. After purifi-cation of the tracer, mass spectrometryconfirmed the specificity of this cou-pling. The specificity of this DIG-BKbinding was also apparent from its effec-tiveness in RIA, in which the DIG-labeledBK exhibited the same immunoreactiveprofile as the native BK. Once character-ized, the tracer can be stored at –20°C foras long as two years without losing itsproperties. This high stability, coupledwith the DIG-labeled tracer’s well-defined activity, provides obvious advan-tages over radioactive and other non-radioactive tracers.

Various immunological methods, in-cluding RIA, have been used to quantifyBK in biological samples (18–20). BK hasalso been detected by EIA (13, 21, 22),and one of these methods used biotin-labeled BK as a tracer in a non-competi-

Calibration curve of BK (•

) and curve displacement obtained with immunoreactive DIG-BK (•) measured by radioimmunoassay.Figure 3

7

6

5

4

3

2

1

0

CP

M (x

100

0)

10–3 10–2 10–1 1.0 101 102 103

LOG concentration (pmol/mL)

Calibration curve of BK (•

) and competition curve with serial dilutions (1:1 to 1:4096) of aninflamed paw extract (•) measured by chemiluminoenzyme immunoassay. The light emission of the zeromean concentration (B0) is represented by (▲).

Figure 4

700

600

500

400

300

200

100

0

Ligh

t em

issi

on (R

LU)

10–4 10–3 10–2 10–1 1.0 101 102

LOG concentration (pmol/mL)

Bradykinin Intra-assay Inter-assayconcentration coefficient of coefficient of

variation (%) variation (%)

6.0 pmol/ml 2.2 4.1

1.5 pmol/ml 3.3 5.0

0.35 pmol/ml 4.0 7.4

Table 1. Intra- and inter-assay precision of CLEIAwith DIG-labeled bradykinin.

CONTENTS

23B I O C H E M I C A ■ N O . 4 [ 1 9 9 6 ]

F E A T U R E S

tive approach (13). Compared to theseassays, the new CLEIA demonstrates ahigher sensitivity. In fact, our assayallows us to detect amounts as low as0.1 fmol/ml of immunoreactive BK.Earlier studies show considerable varia-tion in sensitivity, ranging from58 fmol/ml (23) to 2 fmol/ml (24).Shimamoto and Iimura (25) reported adetection limit of 0.25 fmol/tube withoutany further specification. The analyticalperformance of the new assay is somewhata result of the affinity of the polyclonal IgGused for the immunoconcentration stepbut is mainly an outcome of the high specific activity (one molecule of DIG forone molecule of peptide) without loss ofimunoreactivity, the chemiluminescencedetection method, and the low back-ground values. Like others (26, 27), wehave shown that using such dioxetanederivatives as chemiluminescent alkalinephosphatase substrates leads to 5–10-foldincreases in the sensitivity level of sand-wich immunoassays using a visible lightabsorption (10).

AcknowledgementsThis work was supported by a grant

from the Fonds de la Recherche en Santédu Québec (FRSQ) to A.A. and R.C. R.C.and G.D. are scholars of the FRSQ; A.D.holds a studentship from the Formationde Chercheurs et l’Aide à la recherche.The authors are grateful for the secret-arial assistance of Christiane Laurier andthe graphic work of Claude Gauthier.

References1. Décarie, A., Drapeau, G., Closset, J., Couture, R. and

Adam, A. (1994) Development of digoxigenin-labeled peptide: Application to chemiluminoenzymeimmunoassay of bradykinin in inflamed tissues.Peptides 15(3):511–518.

2. Diamandis, E. P. and Christopoulos, T. K. (1991) TheBiotin-(strept)avidin system: principles and appli-cations in biotechnology. Clin. Chem. 37:625–636.

3. Strasburger, C. J. and Kohen, F. (1990) Two-site andcompetitive chemiluminescent immunoassays.Methods in Enzymology 184:481–491.

4. Ternynck, T. and Avrameas, S. (1990) Avidin-biotinsystem in enzyme immunoassays. Methods inEnzymology 184:469–481.

5. Wilchek, M. and Bayer, E. A. (1990) Introduction toavidin-biotin technology. Methods in Enzymology184:5–13.

6. Wilchek, M. and Bayer, E. A. (1990) Avidin-biotinmediated immunoassays: overview. Methods inImmunology 184:469–481.

7. Kessler, C. (1991) The digoxigenin:anti-digoxigenin(DIG) technology – a survey on the concept and real-ization of a novel bioanalytical indicator system.Molecular and Cellular Probes 5:161–205.

8. Bhoola, K. D., Figueroa, C. D. and Worthy, K. (1992)Bioregulation of kinins: kallikreins, kininogens, andkininases. Pharmacol. Rev. 44:1–80.

9. Schaap, A. P., Akhavan, H. and Romano, L. J. (1989)Chemiluminescent substrates for alkaline phos-phatase: applications to ultrasensitive enzyme-linked immunoassays and DNA probes. Clin. Chem.35:1863–1864.

10. Legris, F., Martel-Pelletier, J., Pelletier, J. P., Colman,R. and Adam, A. (1994) An ultrasensitive chemi-luminoenzyme immunoassay for the quantificationof human tissue kininogens: application to synovialmembrane and cartilage. J. Immunol. Methods168:111–121.

11. Avrameas, S. (1969) Coupling of enzymes to pro-teins with glutaraldehyde, use of conjugates for thedetection of antigens and antibody. Immuno-chemistry 6:43.

12. Peterson, G. L. (1983) Determination of total proteinMethods in Enzymology 91:95–119.

13. Anumula, K. R., Schulz, R. and Back, N. (1990)Immunologic methods for quantitative estimation ofsmall peptides and their application to bradykinin.J. Immunol. Meth. 135:199–208.

14. Regoli, D. and Barabé, J. (1980) Pharmacology ofbradykinin and related kinins. Pharmacol. Rev.32:1–46.

15. Chard, T. (1987) “Characteristics of binding assays-sensitivity” in Laboratory techniques in biochemistryand molecular biology, volume 6, part 2, An intro-duction to radioimmunoassay and related tech-niques, (Burdon, R. H. and van Knippenberg,P. H., eds.). pp. 161–174, Elsevier Press, Amsterdam.

16. LeVine, S. M. and Macklin, W. B. (1988) Biotinenrichment in oligodendrocytes in the rat brain.Brain Res. 444:199–203.

17. Naritoku, W. Y. and Taylor, C. R. (1982) A competitivestudy of the use of monoclonal antibodies usingthree different immunohistochemical methods: anevaluation of monoclonal and polyclonal antibodiesagainst human protatic and acid phosphatase.J. Histochem. Cytochem. 30:253–260.

18. Goodfriend, T. L. and Ball, D. L. (1969) Radio-immunoassay of bradykinin: chemical modificationto enable use of radioactive iodine. J. Lab. Clin. Med.73:501–511.

19. Talamo, R. C., Haber, E. and Austen, K. F. (1968)Antibody to BK: effect of carrier and method of coupling on specificity and affinity. J. Immunol.101:333–341.

20. van Leeuwen, B. H., Millar, J. A., Hammat, M. T. andJohnston, C. I. (1983) Radioimmunoassay of bloodbradykinin: purification of blood extracts to preventcross-reaction with endogenous kininogen. Clin.Chim. Acta 127:343–351.

21. Geiger, R. and Miska, W. (1986) Determination ofbradykinin by enzyme immunoassay. Adv. Exp. Med.Biol. 198:531–536.

22. Ueno, A., Ohishi, S., Kitagawa, T. and Katori, M.(1979) Enzyme immunoassay of bradykinin. Adv.Exp. Med. Biol. 120A:195–202.

23. Barabé, J., Huberdeau, D. and Bernoussi, A. (1988)Influence of sodium balance on urinary excretion ofimmunoreactive kinins. Am. J. Physiol. 254:F484–F491.

24. Scicli, A. G., Mindroiu, T., Scicli, G. and Carretero, O.A. (1982) Blood kinins, their concentration in normalsubjects and in patients with congenital deficiencyin plasma prekallikrein and kininogen. J. Lab. Clin.Med. 100:81–93.

25. Shimamoto, K. and Iimura, O. (1987) “Measurementof circulating kinins, their changes by inhibition ofkininase II and their possible blood pressure lower-ing effect” in Vasodepressor Hormones, p.297–307, Birkäuser Verlag, Basel.

26. Bronstein, I., Voyta, J. C., Thorfe, G. H., Kricka, L. J.and Armstrong, G. (1989) Chemiluminescent assayof alkaline phosphatase applied in an ultrasensitiveenzyme immunoassay of thyrotropin. Clin. Chem.35:1441–1446.

27. Bronstein, I., Edwards, B. and Voyta, J. C. (1989) 1,2-dioxetanes: novel chemiluminescent enzyme sub-strates. Applications to immunoassays. J. Biolum.Chemilum. 4:99–111.

Digoxigenin-3-0- 1 333 054 5 mgmethylcarbonyl-ε-aminocaproic acid-N-hydroxy-succinimide ester(DIG-NHS ester)

Anti-Digoxigenin-AP, 1 093 274 150 UFab fragments (200 µl)

Lysozyme 107 255 1 g1 243 004 5 g

837 059 10 g1 585 657 25 g

Product Cat. No. Pack Size

CSPD® chemi- 1655 884 1 mlluminescent substrate

Also Available Cat. No. Pack Size