--~~-------------------------------~v

An ELISA for the quantification of chickenimmunoglobulin (IgY) in various liquid media"

Matthias Fischer und Andreas HlinakInstitute of Virology, Department of Veterinary Medicine, Free University of Berlin

SummaryThe development of an assay for the quantification ofchicken immunoglobulin (Ig Y) and data evaluation isbeing demonstrated. An enzyme immunoassay wasestablished to perform the quality control of Ig Y purifi-cationsfrom the egg yolk and to control IgY producingcells in vitro. Each step of the test procedure and possi-ble alternatives were described. The assay was charac-terized by special test parameters (sensitivity, detectabil-ity, accuracy, precision, specifity). Moreover, this simpleandfast enzyme immunoassay is a special supplement tothe alternative antibody production from egg yolk.

Key words: enzyme immunoassay, quantification,chicken immunoglobulin, IgY

1 Introduction

The aim of our study was the estab-lishment of an assay for the quantifi-cation of chicken IgG or the so-called IgY. Special requirements forthis test were a high reproducibility,a high detectability and accuracy ofthe assay.IgY is the major traction of anti-

bodies (ab) in chicken serum and itoccurs in egg yolk as well (Warr etal., 1995). The function of this im-munoglobulin class is comparable tothat of mammalian IgG. On the otherhand the structure of the IgY wasidentified as IgA-like (Hadge andAmbrosius, 1983, 1984, 1988 ;Hadge,1985).IgY has many special properties.

Egg yolk antibodies have no interac-tions with rheumatoid factors in se-rum of mammalians, and the risk ofgetting false-positive results in im-munoassays is therefore small. IgY

* This work was supported by grants fromthe Bundesministerium iur Bi/dung, Wissen-scren, Forschung und Tecbnoioqie.

ALTEX 13,3/96

Zusammenfassung: ELISA zur Quantifirierung vonHiihner-IgY in verschiedenen flussigen MedienDie Entwicklung eines Enzymimmunoassays zur Quanti-fizierung von Ig Y in Antikiirperpriiparationen aus demHiihnerei sowie zur Kontrolle von. Antikorper-produrie-renden Zellen in vitro wird beschrieben. Die Etablierungder einzelnen Schritte des Testablaufes sowie insbeson-dere die Charakterisierung des ELISA durch spezifischeTestparameter wird gezeigt. Im Ergebnis stellt derschnell und einfach. durchrufiihrende Enzymimmunoas-say einen spezifischen Beitrag zur alternativen Antikor-pergewinnung uber das Huhnerei dar.

does not bind to protein A and G , tomammalian Fe receptors and mam-malian complement. Anti-mammali-an TgG antibodies which cross-reactto 19G from various mammalianspecies show no or very little crossreaction to IgY. Because of theseproperties and the large evolutionarydistance from chicken to mammals,the egg yolk antibodies offer anumber of applications which are notfeasible using mammalian antibodies(Brandt et al., 1981; Larsson andSjonquist, 1990).Furthermore, the stability of IgY

against physical and chemical influ-ences is much higher than that ofmammalian IgG (Losch et al., 1986).Due to these specifities the IgY

finds increased interest in laboratorypractice. There is a growing numberof publications on the use of polyclo-nal antibodies extracted from eggyolk (LOsch et al., 1986 ; Schade ctal., 1991 ; Gassmann et al., 1990).Another benefit is the efficiency of

the production of egg yolk antibod-ies. The quantity of immunoglobulinseparated from egg yolk of immu-

nized chickens is in most caseshigher than that obtained from theserum of rabbits. Losch ct aI. (1986)described that one egg yolk canprovide 40 - 500 mg IgY. Thisamount of IgY contains 2 - 10 %specific antibodies (Schwarzkopf,1994). Kowlaczyk et al.(l985)showed an amount of IgY in the eggyolk of J 57 mg/ml with a rangefrom 5.3 mg/ml to 43.3 mg/ml, Thatmeans, in lheory an amount of 120mg Ig can be obtained from one egg.Normally, laboratory mammals

such as rabbits, goats, or sheep areused for antibody production. Thismethod includes two procedurescausing distress to the animals: Thefirst is the immunisation itsel f,whereas the second one consists inthe bleeding of the animal as aprerequisite for antibody prepara-tion. In using chickens for antibodyproduction the painful step of bloodcollecting can be replaced by eggcollecting and IgY extraction fromegg yolk. Therefore, the so-calledIgY-technology is a refined methodfor animal welfare.

179

FISCHER UNDHUNAK w~-----------------------------------~~--.,.""

But a central problem of this alter-native antibody production is thepreparation of the antibodies fromthe egg yolk and the quantification ofIgY (Fischer et al., 1996).The quantification of IgY is also

useful for the monitoring of immu-noglobulin-producing cells in vitroor the documentation of the immuno-logical status in chickens.We developed a simple assay to

measure IgY. The results of theevaluation procedure with specialtest parameters were satisfactory.All elements of the assay are com-

mercially avialable. The applicationof the test in other laboratories iseasy to perform. It will be possible toestablish this assay in a diagnostic orimmunological lab without specialequipment.

2 Material and Methods

2.1 The IgY samplesProbes containing the chicken anti-bodies were sera of specific patho-gen free chickens and conventionalhoused chickens, cell culture super-natants of IgG-producing Bvlympho-cytes and several IgY preparationspurified from egg yolk.

2.2 ToolsThe ELISA was performed withmicrotiterplates (Nunc-Irnmuno Pla-te F 96, PolySorp). The rabbit anti-chicken IgG antibody, the rabbitPOD-labeled anti-chicken antibody

(conjugate), and the IgY standardpreparation were obtained from SIG-MA I Immunochernicals (Germany).The results of the purity analysis

of the IgY standard preparation weredemonstrated in a previous study(Fischer et al. ,1996). We establisheda dilution curve of the IgY standard.The dilution steps (expected concen-trations) of this standard curve areshown in Tab.l.

2.3 DevicesA SLT-reader system (SLTLabinstruments,Germany) with inte-grated software (EASY -FIT) wasused to measure the extinction andthe data evaluation.

2.4 Assay procedureFirst, microtiterplates were coatedwith the catch antibody. A rabbitanti-Ig Y -antibody was diluted1:10.000 in coating-buffer (1.59gNa2CO, plus 2.94g NaHC03; ad1000 ml Aqua dest., pH 9.6) and50J.lI per well were added. The plateswere incubated for 4 hours at roomtemperature and after that overnightat 4°C.The plates were washed three

times with washing buffer (8,0 gNaCl; 0,2 g KH2P04; 2,9 g Na2HP04

x 12 H20; 0,2 g KCL; 0,5 ml Tween20; ad 1000,0 ml Aqua dest., pH7,4). All further washing steps werecarried out with this buffer. Samplescontaining the IgY were diluted inseveral different proportions in

Table: The theoretical IgY concentration according to weight and proteindetermination compared with the results obtained from the assay.

expected concentration (ng/ml) found concentration (ng/ml)

200100502512.56.253.1251.5630.7810.3810.1950.097

1981025025.212.86.543.2071.6430.80.4050.1860.117

180

washing butler plus 10 % Gelafusal(Serumwerk Bernburg; Germany) ina ratio 1:10. Following 90 min incu-bation at 37°C, the plates werewashed three times, and 50 III perwell POD-labeled anti-chicken IgGwere added (1: 10.000 in washingbuffer I 10 % Gelafusal). Plates wereincubated for 90 min at 37°C. Afterthis time the plates were washed and50 J.lI of substrate chromogen solu-tion (5 mmol11 o-phenyldiarnin;5mmoll1 H202 in citrate buffer; pH5,0) were pipetted in each well. Thereaction was stopped with 1M H2S04

(50 ul/well) after 15 min. The ab-sorbtion was measured at 492 nmusing a SLT-reader system.

2.5 Assay parametersThe standardization of the assay wasperformed with the following para-meters: sensitivity, detectability, ac-curacy, precision and specifity (Tijs-sen, 1987; Olechnowitz et al., 1990).Sensitivity is the dose response

curve corresponding to changes inthe amount of the reactant . A highsensitivity signifies a stronger re-sponse to slight changes in the con-centration.Detectability defines the detection

limit or the ability to detect smallantibody or immunoglobulin levels.Accuracy is the conformity of re-

sults to an accepted standard value(reference accuracy) or true value.Precision is the calculated result

with standard deviation or thenumber of digits (decimals) in whicha result is expressed.Specifity refers to the degree of

discrimination of the assay betweennegative and positive samples.

3 Results

3.1 Test procedureWe developed an enzyme immun-oassay to detect IgY in variousfluids, especially in different antibo-dy preparations from the egg yolkand in the supernatant of antibody-producing cells, in particularB-lymphocytes. To isolate the IgY

from the egg yolk fluid we used

ALTEX 13,3/96

d";1 FISCHER UND HUNAK--~~~-------------------------------•••<

different separation techniques, com-binations and/or variants of thesemethods (Fischer et aI., 1996). Theresults of these antibody extractionmethods were used for the establish-ment of our assay.The catch-antibody in the test sys-

tem was a rabbit anti-IgY antibodycoated on the microtiterplate. Theother, POD-labeled, antibody wasused to detect to bound IgY . Both ofthese mammalian antibodies werepolyclonal.We were able to use the catch-

antibody coated microtiterplates af-ter an storage at -20 DC over sixmonths.In pre-experiments, we examined

different test conditions, e.g. theconditions of protein coating, theantibody concentrations, the neces-sity of a blocking step and the timeregime of the separate test reactions(Hlinak et al., 1992; Minning et al.,1993). Following these experiments,we could define the described testprocedure.We established no blocking step in

our test procedure. There was noimprovement in minimizing the nonspecific reactions, neither by varyingthe blocking procedure in time andprotein concentration, nor by usingother proteins, like horse serum ormilk protein/here not shown).Different concentrations of the

anti-Ig Y conjugate and the substratesolution were tested. The only effectfound was a slight improvement ofthe assay sensitivity correlated withan increasing substrate concentra-tion.

3.2 Test parametersThe assay had a good sensitivity toIgY concentration between 200 ngper ml and to 0.78 ng/rnl. This meetsexactly the linear part of the sigmoi-dal curve illustrated in fig. 1.Our assay showed its detection

limit at a concentration of 0.78 ng/ml. It has to be stressed that this meansthe limit for quanti fication only.The table compares measured and

proceeded data with the theoreticalamount of IgY according to theweight and protein concentrations.

ALTEX 13,1/%

2.000

1.600

1.200coUc~ O.BOOOJ

0.400

0.0000.010 0.100 1.000 10.00 100.0

IgY concentration in ng/ml1000

Figure 1: Standard curve simulated on the SLT-device (EASY-FIT SLT-systemsoftware) with the data of seven assays. The curve fitting has been performed by afour point interpolation in the lin-Iog modus.The plot could be explained by: OD=A+(D-A):(1 +eB(C,'n(KO)))OD = optical density measuredKO = concentration proceeded from the optical densityA and D = optical density of the asymptotic points to which the curve is convergatingagainst on each side.Band C = are two parameter proceeded by the software by linear regression.

The illustrated parameter is calledaccuracy (T~issen, 1987).The standard deviations as shown

in fig. 1 and 2 show the precision ofthe assay.Figure 3 illustrates the spccifity of

the assay.

According to Rammensee (1984)and Rose et a1.( 1974), IgM and IgAhave never been found in the eggyolk of not incubated eggs. Never-theless, we tested the cross reactivityto IgA and TgM in our assay. TheIgA was preparated from chicken

100

detection limit

90 interassayvariation

coefficient (n=7)for calculate dresults in %

so

70

60

50

40

30

20

10

200 100 50 25 6,25 3,12~ 1,51:53 a,HI' 0,30'1 0,195 0,09712,5

expected IgY concentration in nq/rnl

Figure 2: The between day quotient of variation (of seven assays) referring to theconcentrations proceeded from the measured optical density (fourfold determination).

181

FISCHER UNDHUNAK ~~-------------------------------~~--¥~

25 12.5 6.25 3.125 1,563 0,781 0,391 0,195 0,098

expected IgY concentration in ng/ml

Figure 3: The specifity of the assay shown by fourfold determinations of IgY indifferent concentrations (visible in the diagram) and compared with a negative control(horse serum twofold determinatcd). The white sector of the columns shows theextinction determinated for the horse serum. The arrow marked with d.1.points to thedetection limit or the lowest concentration where a positive-negative discrimination isstili possible.

0,6

0,4

0,2

200 100 50

bile (Erhard et al., 1992) and the IgMwas a generous gift of S. Khilkow(Institute of Agricultural Biotech-nology, Moscow, Russia). We couldnot find any cross reactivities in thisexamination.

4 Discussion

The described assay was developedwith the following requirements:• inexpensive• easy handling and establishment• high detectability and accuracy.We established a simple and fast testprocedure to measure IgY in variousliquid media. We used polyclonalantibodies as catch-antibody and asdetection POD-labeled antibody inan enzyme immunoassay.The choice of a polyclonal anti-

body indicated several advantagesand disadvantages, which were eval-uated. In contrast to polyclonal anti-bodies, the monoclonal antibody(rnab) reacts with "UIle:" specificepitope. This results in an assay,which is 100% specific and repro-ducible. On the other hand, due tothe high specifity of the mab to adefined cpitope, a lower amount of

detection limit

j'-----'--'-'--',I~'M~

immune complexes is formed, whichmeans a weaker amplification of thechromogenic reaction or a lowersensitivity. For the same reason thedetectability of the assay increaseswith the use of polyclonal antibodies(Kiessig, 1991). Furthermore, thecomplicated way of production andcharacterization of monoclonal anti-bodies did not meet the requirementsfor our assay.We found a detection limit of 0.78

ng/ml. It exceeds the sensitivity ob-tained by Erhard et al. (1992). Theydemonstrated a detection limit of 20ng TgY per ml in an assay based onmonoclonal antibodies.Polyclonal antibodies can not be

characterized in the same way asmab because they are a heterogenousfraction of antibodies to severalepitopes. Therefore the assay has tobe equilibrated to every new anti-body charge (Kiessig, 1991). How-ever, we did not find any differenceswhen changing the IgY charge orprcpurntion. The polyclonal antibod-ies proved to be a powerful andspecific antibody as well as an inex -pensive one.The statistical work has been re-

duced to the determination of the

standard deviation and the variationcoefficient. The interpretation of thevariation-coefficient (CV) representssimple statistical work. Nevertheless,it provides a lot of information.In the statistical parameters, the

assay meets the requirements ofPorstmann and Kiessig (1992) whorecommend a CV <10% for theintraassay variances and a CV <15%for a day to day imprecision.According to McLaren et

al.(l98l), for measurements on dif-ferent days a variation coefficient(interassay variation) below 10% issaid to be in a good range. Thatmeans the figure is very tight to thisupper level in our assay. This facthas to be considered in the interpre-tation of the results. Around thedetection limit (0,78 ng/ml) the vari-ation coefficient has increased morethan 50%. Therefore we fixed thedetection limit at this concentration.Nevertheless, the assay is able todetect IgY below this limit, as wellas in a YES-or-NO-answer.The parameters of the test system

seem to be sufficient for a test withthe use of polyclonal antibodies. Thesensitivity is suitable to the effort ofquantitation of IgY in laboratorypractice. As mentioned before, wedeveloped a more sensitive assay byusing an indirect sandwich tech-nique. Porstmann and Kiessig (1992)recommended the use of an antibodyproduced in the same species as thespecific catch-antibody for the anti-species antibody. Nevertheless, thedetectability dropped so much that apositive-negative discrimination wasno longer possible,In conclusion, we demonstrated a

simple, fast and inexpensive test-procedure for the quantification ofIgY, based on polyc1onal antibodiesin an enzyme immunoassay. Thisassay is characterized by very goodtest parameters and its establishmentis possible without special laboratoryequipment. Therefore this enzyrl.1eimmunoassay is a special supple-ment to the alternati ve way ofantibody production from egg yolk.

ALTEX 13, 3/9n

z~ FTSCHER UND HUNAK

--h~~---------------------------¥G~References

Brandt, C. D., Kim, H. W., Rodriguez,W. J., Thomas, L., Yolken, R. H.,Arrobio, 1. 0., Kapikian, A. Z.,Parrott, R. H. and Channot, R. M.(1981). Comparison of direct elec-tron microscopic, Immun electronmicroscopic, and rota virus enzyme-linked immunosorbent assay for de-tection of gastroenteritis viruses inChicken. J. Clin. Micrbiol. 13 , 976-981.

Erhard, M. H., Quistorp, I. von,Schranner, I., Jiingling, A., Kaspers,8., Schmidt, P. and KUhlmann, R.(1992). Development of specific en-zyme-linked immunosorbent antibo-dy assay systems for the detection ofchicken lmmunglobulins G, M andA using monoclonal antibodies.Poult. Sci. 71, 302-310.

Fischer, M., Hlinak, A., Montag, T.,Claros,M., Schade, R. and Ebner, D.(1996). Vergleich von Standardme-thoden zur Praparation von Dotter-antikorpern. Tierarzti.Praxis 24,411-418.

Gassmann, M., Weiser, T., Thomrnes,P. and Hiibscher, U. (1990). DasHiihnerei als Lieferant polyklonalerAntikorper. Schweiz; Arch. Tier-he ilk. 132,289-294.

Hadge, D. and Ambrosius, H. (1983).Evolution of low molecular weightimmunglobulins. IlL The immunglo-bulin of chicken bile - not an TgA.Mol.Immunol. 20, 597-606.

midge, D. and Ambrosius, H. (1984).Evolution of low molecular weightirnmunglobulins. IV. IgY-like im-munglobulins of birds, reptils am-phibians - precursors of mammalianIgA. Mol. Immunol. 21,699-707.

midge, D. (1985). Zur Evolution derImmunglobuline. All. Immunol. ]1,597-605.

Hadge, D. and Ambrosius H. (1988).Cornperative studies on the structureof bilary immunglobulins of someavian species, II. Antigenic proper-ties of the biliary immunoglobulin ofchicken, turkey, duck and goose.Do Compo lnununol. 12.319-329.

Hlinak, A.. Dahms. H. and Kraus, G.(1992). Optimierung eines Enzy-mimmunoassay zum Nachweis vanAntikorpern gegen aviare Reoviren- unter besondcrcr Berucksichtigung

ALTEX 13,3/96

der Ergebnisberechnung und Testin-terpretation, J. Vet. Med. ]9, 428-442.

Kiessig, S. (1991). Enzymimmunoas-says. BioTec 4, 30-34.

'Kowalczyk, K., Daiss, J., Halpern, J.and Roth, T. F. (1985). Quantitationof maternal-fetallgG transport in thechicken.lmmunol. 54,755-762.

Larsson, A. and Sjoquist, J. (1990).Chicken IgY: Utilizing the evolutio-nary difference. Camp. Immun.Microbial. Infect. Dis. 13, 199-201.

Losch, U., Schranner, L, Wanke, R.and JUrgens, L (1986). The chickenegg an antibody source. J. Vet. Med.B 33,609-619.

McLaren, M. L, Lillywhite, J. E. andAu, A. C. S. (1981). Indirect enzymelinked immunosorbent assay (ELI-SA): practical aspects of standardi-zation and quality control. Med. Lab.s-: 38,245-249.

Minning, P., Hlinak, A. and Klaczin-ski, K. (1993): Enlwicklung cinesEnzyrn-Immuno-Assay zum Nach-weis von Antikorpern des Huhnesund der Maus gegen das Virus derNewcastle -Disease. Bed. Munch.Tierarril. Wschr. 106,5,149-153.

Olechnowitz, A. F., Miko, A., Koeper-nik, H., Starick, E. and Frobe, L(1990): Die Validierung eines Enzy-mimmunoassays zum Nachweis vonAntikorpern gegen das Rinderleuko-sevirus. Arch. expel'. Vet. med. 44, 2,279-288.

Porstmann, T. and Kiessig, S., (1992).Enzyme immunoassay techniques,An overview. 1. lmmunol. Meth.114, I-Hi.

Rammensee, M. (1984). Irnrnunglobu-linkonzentration in Serum, Sekrctcn,Ei und Embryo des Haushuhns, Uni-versitat Mtinchen: Vet. rned. Disser-tation.

Rose, M. E., Orlans, E. and Buttress,N. (l974L Immunglobulin classes inthe hens egg: their segration in yolkand white. Eur. J. lmmunol. 4, 521"':'523.

Schade, R., Pfister, C, Halatsch, R.and Henklein, P. (1991). PolyclonalIgY Antibodies from Chicken EggYolk-an Alternative to the Produc-tion of Mammalian IgG Type Anti-bodies in Rabbits. ATLA 19, 403-419.

Schwarzkopf, C. (1994). Gewinnung

und immunologische Charakterisie-rung speziessperifischer If!.Y sowiederen Einsarz zur Bestimmung derWirrsrierart aus dem Abdominalbluthtimatogener lnselaen. Freie Univer-sitar Berlin: Vet. med. Dissertation.

Tijssen, P. (1987). Practice and theoryof enzyme imrnunoassays, In R. H.Burdon and P. H. V. Knippenberg(ed.), Laboratory techniques in bio-chemistry and molecular biology.Vo1.15 (pp. 385-421). Amsterdam:Elsevier Science Pub.

WalT, G. W., Magor, K. E. and Hig-gins, D. A. (1995). IgY: clues to theorigins of modern antibodies. Immu-nol. Today 16,8,392-398.

Correspondence addressDr. Andreas HlinakInstitut fur VirologieFB Veterinarmedizin (Mi lle)Luisenstr. 56D-101l7 Berlin

183