Ž .Clinica Chimica Acta 308 2001 99–106www.elsevier.comrlocaterclinchim

A new homogeneous enzyme immunoassay for thyroxine usingglycogen phosphorylase b–thyroxine conjugates

Christina D. Karapitta a,b, Aristotelis Xenakis a,), Athanassios Papadimitriou c,Theodore G. Sotiroudis a

a Industrial Enzymology Unit, Institute of Biological Research and Biotechnology, The National Hellenic Research Foundation,48 Vassileos Constantinou AÕe., 11635, Athens, Greece

b MEDICON S.A., Gerakas, Greecec Department of Nuclear Medicine, NaÕy’s Hospital, Athens, Greece

Received 10 October 2000; received in revised form 5 February 2001; accepted 14 February 2001

Abstract

Ž .Background: Measurement of serum thyroxine T concentration is important for diagnosis of thyroid gland diseases.4

We developed a practical homogeneous enzyme immunoassay for thyroxine analysis in unextracted sera. Methods: AŽ .thyroxine derivative conjugated to a reactive sulfhydryl group of glycogen phosphorylase b GPb . Conjugation caused

inhibition of enzyme activity and the enzyme conjugate was re-activated upon the binding of a polyclonal anti-T antibody.4

Antibody-activation was blocked by the presence of free T . Results: Conjugation affected the allosteric character of the4

enzyme and the K for the allosteric activator AMP was increased 28 times, while anti-T antibody partially reversed thism 4

effect. The optimum concentration ratio of enzyme conjugate to anti-T antibody was determined, and T was measured with4 4

desired sensitivity and accuracy in the range between 10 and 240 mgrl. Furosemide was used to inhibit the interaction ofthyroxine with serum T -binding sites. Human serum T values obtained by this method correlated well with those obtained4 4

Ž .by a radioimmunoassay ys1.9q1.0 x, rs0.97, Ns72 . Conclusions: Chemical modification of glycogen phosphory-lase b with a T derivative led to the development of a simple homogenous enzyme immunoassay for T analysis with the4 4

desired sensitivity and accuracy. q 2001 Elsevier Science B.V. All rights reserved.

Keywords: Homogeneous enzyme immunoassay; Thyroxine; Rabbit muscle glycogen phosphorylase b

AbbreÕiations: GP, glycogen phosphorylase; T , L-thyroxine;4Ž .RIA, radioimmunoassay; SMCC, 4- maleimodomethyl -cyclohe-

xane-1-carboxylic acid N-hydroxysuccinimide ester; MCC-T ,4w Ž . xN- 4- maleimidomethyl -cycloexan-1-yl tetraiodothyronyl carbox-

amide; TEA, triethanolamine; BSA, bovine serum albumin; SDS,sodium dodecyl sulfate

) Corresponding author. Tel.: q301-727-3893; fax: q301-727-3758.

Ž .E-mail addresses: arisx@eie.gr A. Xenakis , tsotir@eie.grŽ .T.G. Sotiroudis .

1. Introduction

Certain enzymes, when covalently conjugated withlow molecular weight haptens, can be inhibited oractivated by anti-hapten antibodies. Among others,

w x w xlysozyme 1 , malate dehydrogenase 2,3 , glucosew x6-phosphate dehydrogenase 4 show this behavior.

Inhibition or activation of enzyme activity by anti-hapten antibodies can be blocked by prior incubation

0009-8981r01r$ - see front matter q 2001 Elsevier Science B.V. All rights reserved.Ž .PII: S0009-8981 01 00469-7

( )C.D. Karapitta et al.rClinica Chimica Acta 308 2001 99–106100

of the antibodies with the free hapten. This providesthe basis for homogeneous enzyme immunoassay fordetermination of various haptens. Much of the workconcerning homogeneous enzyme immunoassay sys-tems has come from commercial organizations. These

Ž . w xsystems include the EMITe Syva technique 4,5Ž .and the CEDIAe Microgenics technique devel-

oped using enzyme fragments prepared by recombi-w xnant DNA technology 6,7 .

Ž .Glycogen phosphorylase GP is an allosteric en-zyme catalyzing the degradative phosphorolysis ofglycogen to glucose 1-phosphate. The rabbit muscleenzyme exists in two intercovertible forms, the de-phosphorylated form b and the phosphorylated forma. GPb is inactive and can be activated either bynoncovalent cooperative binding of AMP or by cova-

w xlent phosphorylation to form GPa 8,9 . GPb consistsof two identical polypeptide chains. GPb from rabbit

w xmuscle contains 842 amino acids per subunit 10and has nine sulfydryl groups important for enzyme

w xactivity 11 . GP exists in human serum of healthysubjects predominantly in the b-form at very low

Ž . w xconcentrations -5 mgrl 12,13 .We describe here the conjugation of thyroxine

Ž .T to GPb from rabbit muscle by the use of 4-4Ž .maleimodomethyl cyclohexane-1-carboxylic acid

Ž .N-hydroxysuccinimide ester SMCC , a reagent witha maleimide group, which reacts specifically with

w x–SH groups 14 , for the development of a newhomogeneous enzyme immunoassay for T analysis4

using this conjugate. A kinetic analysis providinginsight into the mechanism of inhibition and anti-body-induced re-activation of phosphorylase is alsopresented. Our results permit the development of aclinically useful assay for serum T .4

2. Materials and methods

2.1. Materials

ŽAll chemicals, polyclonal anti-T antibody T-4.2652 prepared in rabbits and monoclonal anti-T4

Ž . Žantibody T-3901 were from Sigma St. Louis, MO,. Ž .USA Sephadex G-25 fine was from Amersham

Ž .Pharmacia Biotech Uppsala, Sweden . Thyroxineserum calibrators were from Bayer DiagnosticsŽ .Munich, Germany . 0.25 mm silica gel 60 thin layer

Ž .chromatography plates 5721 were obtained fromŽ .Merck Darmstadt, Germany .

2.2. Purification of rabbit muscle phosphorylase b

Rabbit muscle GPb was prepared according tow xFischer and Krebs 15 , using 2-mercaptoethanol in-

stead of cysteine and recrystallized at least fourtimes before use. GP concentration was determinedspectrophotometrically using the absorbance index

1% w xE s13.2 16 .280 nm

2.3. Synthesis and purification of MCC-T4

w Ž . xThe N- 4- maleimodomethyl -cyclohexan-1-yl te-Ž .traiodothyronyl carboxamide MCC-T was synthe-4

sized with modifications as described previouslyw x Ž17,18 . 1 mmolrl SMCC solution of 3 grl in

.dimethylformamide was added to 1 mmolrl T4Ž .solution of 1 grl in dimethylformamide in a finalreaction volume of 1 ml. The reaction was carriedout for 90 min at 308C. The product MCC-T was4

purified by thin layer chromatography on 0.25 mmŽ .Silica Gel 60 plates Merck with ethyl acetateracetic

Ž .acidrwater 90r10r5, vrv as a mobile phase. Theprominent band at R s0.67 was cut out and thef

product was extracted with methanol. The samplewas filtered and the filtrate was condensed with arotary evaporator until 1 ml. The product was quanti-fied spectrophotometrically. The hormone derivativewas stored as a methanol solution in the dark aty208C.

2.4. Conjugation of MCC-T to phosphorylase b4

The enzyme conjugates were prepared by chemi-cally coupling MCC-T to GPb. In the coupling4

Ž .procedure, 0.4 ml of GPb 10.3 mmolrl in 50mmolrl TEA-HCl buffer pH 6.8 was mixed with0.037 ml of various concentrations of MCC-T in4

methanol. The mixture was incubated at 308C for 15min. A Sephadex G-25 column was used for theseparation of the enzyme conjugates from free MCC-T . The column was equilibrated with 50 mmolrl of4

TEA-HCl buffer pH 6.8 and 1 mmolrl EDTA andeluted with the same solution. The enzyme fractionswere collected and the protein concentration wasdetermined. The enzyme conjugates were stored at

( )C.D. Karapitta et al.rClinica Chimica Acta 308 2001 99–106 101

y208C after the addition of 50% vrv glycerol, 0.03molrl of 2-mercaptoethanol and 1 grl of bovine

Ž .serum albumin BSA and could be used for at least6 months.

2.5. Determination of phosphorylase b actiÕity

Phosphorylase b activity in the direction of phos-phorolysis of glycogen was carried out using theauxiliary assay system, as described by Helmreich

w xand Cori 19 with some modification. The finalreaction mixture was 0.6 ml and contained 4 mgrlof GPb, 2.5 kUrl of glucose-6-phosphate dehydro-genase, 0.6 kUrl of phosphoglucomutase, 1 mmolrlof NADPq, 1 mmolrl of magnesium acetate, 1mmolrl of glucose 1,6-diphosphate, 16 mmolrl ofphosphate pH 7, 1 mmolrl of AMP, 5 grl ofglycogen, 0.15 mmolrl of EDTA and 20 mmolrl ofTris-acetate buffer pH 7.4 containing 0.03 molrl ofNaCl, 0.1 mmolrl of MgCl , 0.1 grl of BSA, 0.52

grl of gelatin, 0.1 grl of NaN and 0.2 mmolrl of3w xN-ethylmaleimide 18 . Enzyme, AMP and glycogen

were preincubated for 15 min at 308C, before thereaction was initiated with this mixture. The reactionwas stopped by the addition of 0.05 ml of 13 grlSDS at time selected. NADPH formed in the reactionwas measured spectrophotometrically at 340 nm.

Kinetic data were analyzed by the use of thew xnon-linear regression program GraFit 20 .

2.6. Determination of the number of T conjugated4

to GPb

The number of molecules of T conjugated per4

subunit of GPb was calculated using the method ofw xSaboori et al. 21 for the determination of iodine in

iodoproteins. The method was based on the catalyticŽ .activity of iodine in the reduction of ceric Ce IV to

Ž . Ž .Ce III by As III . The reduction leads to decol-oration of yellow ceric ion to a colorless cerious ion,a process which can be monitored spectophotometri-cally. The rate of change in absorbance is propor-

w xtional to the concentration of the iodine 21 .

2.7. Assay of thyroxine

For thyroxine analysis, the following reagentswere prepared.

Antibody reagent: This contained polyclonal anti-T antibody, 40 mmolrl of furosemide, 0.8% vrv of4

surfactant Tween 20 and 80 mmolrl of Tris-acetatebuffer pH 7.4 containing 0.12 molrl of NaCl, 0.4mmolrl of MgCl , 0.4 grl of BSA, 2 grl of gelatin,2

0.4 grl of NaN and 0.8 mmolrl N-ethylmaleimide.3

Enzyme reagent: This contained GPb-T conju-4

gate, 6 mmolrl of AMP and 30 grl of glycogen.Blank enzyme reagent: This contained 6 mmolrl

of AMP and 30 grl of glycogen.Substrates reagent: This contained 5 kUrl of glu-

cose 6-phosphate dehydrogenase solution, 1.2 kUrlof phosphoglucomutase solution, 2 mmolrl ofNADPq, 2 mmolrl of magnesium acetate, 2 mmolrlof glucose 1,6-diphosphate and 32 mmolrl of Pisolution pH 7.

Thyroxine serum calibrators: These were providedas lyophilized serum-based preparations containing0, 20, 40, 80, 120 and 240 mgrl of L-thyroxine.

The assay was performed as follows: 0.05 ml ofcalibrator or serum sample and 0.15 ml of antibodyreagent were incubated at 308C for 15 min followedby the addition of 0.1 ml of enzyme reagent andincubation of the mixture for 15 min at 308C. Fi-nally, 0.3 ml of the substrates reagent was added andafter further incubation at 308C for 30 min, thereaction was stopped by the addition of 0.05 ml of13 grl SDS. The absorbance was measured at 340nm. For each sample, the respective blank samplewas measured by the addition of 0.1 ml of blankenzyme reagent instead of the enzyme one. Theabsorbance value for each sample was calculated byabstracting the absorbance blank value from the sam-ple value. The standard curve was constructed byplotting the absorbance at 340 nm vs. calibrator T4

concentration.

2.8. Radioimmunoassay

For comparison, we assayed thyroxine with aw xT4-solid-RIA kit 22 , a product of the Institute of

Radioisotopes and Radiodiagnostic Products,Demokritos, NCSR, Athens, Greece.

2.9. Subjects

Blood samples without anticoagulants were ob-tained from patients of Navy’s Hospital, Athens.

( )C.D. Karapitta et al.rClinica Chimica Acta 308 2001 99–106102

3. Results

Conjugation of amide MCC-T to GPb at increas-4

ing molrmol ratio of amide to GPb caused inhibitionof enzyme activity, which increased with increasing

Ž .MCC-T Table 1 . The kinetics of inhibition indi-4

cated that the reaction of conjugation was instanta-neous for all the concentrations of MCC-T used.4

The activity of enzyme conjugates prepared wasdetermined in the presence and absence of excess

Ž .anti-T antibody polyclonal or monoclonal and the4

number of haptens bound to an enzyme subunit wasmeasured. Polyclonal anti-T antibody induced re-4

activation of GPb-T conjugates, while monoclonal4

anti-T antibody led to either re-activation or in-4

creased inhibition of the enzyme activity dependingŽ .on the extent of conjugation Table 1 . Re-activation

of GPb-T conjugates was blocked by addition of4Ž .free T not shown . This permitted an enzyme im-4

munoassay for the determination of T to be set up.4

The GPb-T conjugate with 0.9 molecules of thyrox-4Žine bound per enzyme subunit coupling ratio GPb:

.MCC-T 1:2 was employed for the homogeneous4

enzyme immunoassay, because this conjugate wasthe most re-activated by excess of polyclonal anti-body.

In an effort to understand the mechanism of inhi-bition of phosphorylase b by MCC-T conjugation, a4

comparative study of the kinetic parameters of thenative and modified enzyme in the presence andabsence of polyclonal anti-T antibody was under-4

taken. As shown in Table 2, the conjugation led toincreases in the K value for both substrates, Pi andm

glycogen. Anti-T antibody addition largely reversed4

these changes. In parallel, the conjugated enzymeŽ .lost its allosteric character Hill coefficient, n-1

and the K value for AMP was increased aboutm

28-fold. Anti-T antibody partially reversed these4

effects reducing the K value for AMP of them

modified enzyme about sixfold.To select the optimum concentration ratio of en-

zyme conjugate to anti-T antibody that could be4

used to assay T concentration with desired sensitiv-4

ity and accuracy, we first followed the time course ofthe re-activation of GPb-T conjugate by polyclonal4

anti-T antibody. Re-activation of GPb-T was com-4 4

pleted after the incubation of the conjugate and theŽ .antibody for 15 min at 308C data not shown .

Furthermore, the optimization procedure was similarw xto that for theophylline analysis in serum 4 . GPb-T4

conjugate was mixed with gradually increasingamounts of polyclonal anti-T antibody and the en-4

zyme re-activation was determined. As shown in Fig.1, enzyme activity was increased with increasingantibody concentration. Since the immunoassay mustbe able to discriminate among various T concentra-4

tions likely to be encountered when assaying patient’sspecimens, the optimum ratio of antibody to enzymeconjugate in the presence of different T concentra-4

tions within the desired assay range was determined.Fig. 2 illustrates an assessment of the system abilityto discriminate among hypothyroidism, normal and

Table 1Activity of GPb-T conjugates in the absence and presence of polyclonal and monoclonal anti-T antibodiesa

4 4

Coupling % GPb-T % GPb-T activity % GPb-T activity Molecules T4 4 4 4breaction ratio activity in the presence of in the presence of bound per GPb

b b cGPb:MCC-T polyclonal anti-T monoclonal anti-T subunit4 4 4

1:1 83.5"1.3 93.9"2.6 99.7"3.9 0.31:2 56.2"8.8 73.9"4.1 68.2"5.7 0.91:3 43.3"3.8 50.6"4.2 41.7"4.8 1.01:4 34.9"7.2 39.7"5.2 29.9"4.8 1.31:5 31.8"1.3 34.5"9.2 25.4"7.0 1.61:6 28.2"1.8 31.2"5.0 22.2"4.7 1.6

aActivity of enzyme conjugates was measured using a final concentration of 4 mgrl of GPb-T conjugate and 10 ml of antibody.4b100% enzyme activity was the activity of purified GPb prior to conjugation. Activities are given as mean values"S.D. for five

experiments.c w xThe number of T molecules bound per enzyme subunit was determined according to Ref. 21 .4

( )C.D. Karapitta et al.rClinica Chimica Acta 308 2001 99–106 103

Table 2Kinetic parameters of native GPb and GPb-T conjugate in the absence and presence of polyclonal anti-T antibodya

4 4

b c dEnzyme Pi Glycogen AMPe f e f e fŽ . Ž . Ž .V K mmolrl n V K grl n V K mmolrl nmax m max m max m

native GPb 22.7"0.2 1.4"0.1 1.4"0.1 24.5"0.4 0.007"0.001 F1 23.4"0.1 28"1 1.6"0.1GPb-T 6.4"0.1 6.7"0.2 1.3"0.1 6.7"0.1 0.026"0.001 F1 7.2"0.2 772"70 F14

GPb-T qanti-T 6.0"0.1 3.4"0.1 1.2"0.1 5.6"0.1 0.016"0.001 F1 6.1"0.1 127"11 F14 4

aActivity of enzymes was measured using a final concentration of 0.67 mgrl of enzyme and 0.83 ml of polyclonal anti-T antibody.4b Measured in the presence of 5 mmolrl AMP, 5 grl of glycogen and Pi concentrations between 0.5 and 50 mmolrl.c Measured in the presence of 5 mmolrl AMP, 22 mmolrl Pi and glycogen concentrations between 0.01 and 5 grl.d Measured in the presence of 5 grl of glycogen, 22 mmolrl Pi and AMP concentrations between 0.01 and 5 mmolrl.eV is mmol NADPH miny1 mgy1.maxf The Hill coefficients n were obtained by replotting the corresponding data of the table in the term of a Hill plot.

Žhyperthyroidism T concentrations 20, 80 and 1804.mgrl of T , respectively . The determined optimum4

antibody title was 0.83 ml for the given amount ofthe GPb-T conjugate for 80 and 180 mgrl of T ,4 4

while at 20 mgrl of T the range of optimum4

antibodyrenzyme conjugate ratio was broad. Fig. 3indicates a representative standard curve for ourhomogeneous enzyme immunoassay for T using six4

thyroxine serum calibrators.The sensitivity of the present immunoassay was 5

mgrl T . This value was determined by calculating4

the concentration of T , which would give a response4Žequal to 2 S.D. above that of T zero calibrator 0.04

.mg T rl .4

Fig. 1. Re-activation of GPb-T activity by various concentrations4

of polyclonal anti-T antibody. Activity of conjugate was mea-4

sured using a final concentration of 0.67 mgrl of GPb-T conju-4

gate and anti-T antibody as indicated. A and Ao: absorbance at4

340 nm in the presence and absence of antibody, respectively.

Analytical recovery studies were carried out byadding T to T -free human sera to give samples with4 4

20, 45, 80, 120 and 160 mgrl. Analytical recoverieswere 98%, 97%, 106%, 101% and 97%, respec-tively.

To evaluate the applicability of the present im-munoassay, we compared the values for 72 serumsamples as measured with our homogeneous enzymeimmunoassay with those values obtained by a ra-dioimmunoassay. The regression equation was ys

Ž .1.9q1.0 x range examined 10–240 mgrl of T4Žand the correlation coefficient was 0.97 data not

Fig. 2. Determination of optimum anti-T antibodyrenzyme con-4

jugate ratio. Activity of enzyme conjugates was measured as inFig. 1. Each curve represents the assay response A–AT , when4

1.3 mgrl GPb-T conjugate interacted with various concentrations4Ž . Ž .of anti-T antibody, in the absence A and presence AT of4 4

Ž .three thyroxine concentrations as indicated: 20 mgrl ` , 80Ž . Ž .mgrl v and 180 mgrl ' . A and AT : absorbance at 340 nm4

in the absence and presence of T , respectively.4

( )C.D. Karapitta et al.rClinica Chimica Acta 308 2001 99–106104

Fig. 3. Calibration curve for thyroxine. Activity of enzyme conju-gate was measured using a final concentration of 0.67 mgrl ofGPb-T conjugate and 0.83 ml polyclonal anti-T antibody. Thy-4 4

roxine assay was performed as described in Materials and meth-ods.

.shown . However, because correlation coefficientsare measures of the association between two meth-

w xods but not of the agreement between them 23 , thedegree of agreement was assessed using the Bland–

w xAltman 23 graphical technique by the calculatedvalue of the bias derived from the mean and S.D. ofthe differences . As shown in Fig. 4, for 72 samplesthe mean difference was y1.15 mgrl, which indi-cated that the homogeneous assay was in concor-dance with the RIA method and did not systemati-

Fig. 4. Bland–Altman plot of differences between RIA and thepresent homogeneous enzyme immunoassay.

Table 3Precision of thyroxine assay a

Mean mgrl CV%bWithin-run

Sample 1 46.7"2.5 5.4Sample 2 61.7"1.7 2.8Sample 3 131.0"9.0 6.9

cBetween-runSample 1 12.0"2.0 16.7Sample 2 53.0"3.0 5.7Sample 3 186.0"9.9 5.3

a T concentration was determined according to the assay of T4 4Ž .Materials and methods .

b Mean"S.D. for eight measurements in one assay.c Mean"S.D. for six measurements in six consecutive days.

cally under- or overestimate the T values. Values4

are fairly well distributed above and below the zeroŽ .line, while three samples 4.2% had values )2

S.D. above or below the mean, i.e. outside the limitsof agreement. However, discrepancies betweenmethods are in many cases )10%.

Within-run precision was estimated by analyzingthree samples eight times on the same day; the

Ž .coefficients of variation CVs ranged from 2.8% to6.9%. Between-run precision was measured in dupli-cate for three samples on six different days; CVs

Ž .ranged from 5.3% to 16.7% Table 3 .

4. Discussion

Modification of –SH groups of GPb with MCC-T4

led to inhibition of the enzyme activity, while bind-ing of excess of polyclonal anti-T antibody to the4

modified enzyme led to re-activation. One –SH groupwas modified per enzyme subunit at the ratio se-lected for immunoassay enzyme conjugate. This con-jugate lost its allosteric character and the K valuem

for AMP was increased 28 times. It is known thatcysteine 318 is near the allosteric binding site and itsaccessibility was calculated as 77.8% at GPb dimer

w xusing the program NACCESS 24 . The –SH group˚of cysteine 318 is 9.03 A from atom N1 of adenosine

˚and 14.7 A from phosphorus of the phosphate groupŽof AMP the distances were calculated using the

Ž . w x.program CONTACT CCP4, 1994 25 . All the

( )C.D. Karapitta et al.rClinica Chimica Acta 308 2001 99–106 105

above, in addition to the fact that the conjugationreaction was instantaneous, lead to the conclusionsthat cysteine 318 was the target of modification byMCC-T and that this modification was the cause of4

enzyme inhibition. The presence of the anti-T anti-4

body triggered conformational changes to the en-zyme conjugate inducing enhancement of the affinityof GPb for AMP and a concomitant increase of theenzyme activity.

The homogeneous enzyme immunoassay for T4

analysis described here is simple and sufficientlysensitive and accurate for determining thyroxine con-centrations in serum between 10 and 240 mgrl.Also, it could be adapted to an automated analyzer.Our results compare favorably with those obtainedby a radioimmunoassay. Although, serum contains a

Ž .small amount of glycogen phosphorylase -5 mgrlw x12,13 , its presence does not influence the determi-nation of T with our enzyme immunoassay.4

Furosemide is successfully used for blocking thew xbinding of T to serum proteins 26 and no pretreat-4

ment of serum samples is needed, in contrast to otherw xhomogeneous immunoassays 27 . The most com-

mon compound used to block the binding of T to4

thyroxine-binding globulin, 8-anilino-1-naph-w xthalenesulfonic acid 28 , was excluded due to ab-

w xsorption measurement interference 29 and GP inhi-w xbition 30 .

The homogeneous enzyme immunoassays are rel-atively insensitive and therefore only suitable foranalytes circulating at relatively high concentrations.However, it is recognized that the great success ofthe similar system developed by the Syva scientistswas its application to therapeutic drug monitoringw x31 . By increasing the sensitivity of our homoge-neous enzyme immunoassay 15–20-fold using fluo-rometric determination of NADPH produced in the

w x Ž .GP assay 32 data not shown , our immunoassaycan be employed for the determination of variouslow molecular weight drugs, hormones or neuro-transmitters with functional amino groups, such astriiodothyronine, amphetamine, mescaline, seroto-nine and norepinephrene. GPb modified by similarprocedures is expected to give enzyme-hapten conju-gates with properties similar to those of GPb-T .4

Furthermore, although there are many other ways tomeasure thyroxine and it is difficult to apply ourcurrent methodology in free T analysis, the private4

sector may commercially develop this new im-munoassay for total thyroxine.

Acknowledgements

This work has been supported by the Greek Gen-eral Secretariat of Research and Technology, GrantFPEP 55, and by MEDICON. We thank Dr. N.G.Oikonomakos for his valuable suggestions.

References

w x1 Rubenstein KE, Schneider RS, Ullman EF. Homogeneousenzyme immunoassay. A new immunochemical technique.Biochem Biophys Res Commun 1972;47:847–51.

w x2 Rowley GL, Rubenstein KE, Huisjen J, Ullman EF. Mecha-nism by which antibodies inhibit hapten–malate dehydroge-nase conjugates. An enzyme immunoassay for morphine. JBiol Chem 1975;250:3759–66.

w x3 Ullman EF, Yoshita RA, Blakemore JI, Maggio E, Leute R.Mechanism of inhibition of malate dehydrogenase by thyrox-ine derivatives and reactivation by antibodies. Homogeneousenzyme immunoassay for thyroxine. Biochim Biophys Acta1979;567:66–74.

w x4 Chang J, Gotcher S, Gushaw JB. Homogeneous enzymeimmunoassay for theophylline in serum and plasma. ClinChem 1982;28r2:361–7.

w x5 Ullman EF, Blakemore J, Leute RK, Eimstad W, Jaklitsch A.Homogeneous enzyme immunoassay for thyroxine. ClinChem 1975;21r7:1011.

w x6 Hederson DR, Friedman SB, Harris JD, Manning WB, Zoc-coli MA. CEDIAe, a new homogeneous immunoassay sys-tem. Clin Chem 1986;32r9:1637–941.

w x7 Khanna PL, Dworschack RT, Manning WB, Harris JD. Anew homogeneous enzyme immunoassay using recombinantenzyme fragments. Clin Chim Acta 1989;185:231–9.

w x8 Graves DJ, Wang JH. a-Glucan phosphorylases—chemicaland physical basis of catalysis and regulation. In: Boyer PD,editor. 3rd edn. The enzymes vol. 7. New York: AcademicPress, 1972:432–82.

w x9 Sotiroudis TG, Cazianis CT, Oikonomakos NG, Evan-gelopoulos AE. Effect of sodium cholate on the catalytic andstructural properties of phosphorylase b. Eur J Biochem1983;131:625–31.

w x10 Titani K, Koide A, Hermann J, Ericsson LH, Kumar S, WadeRD, et al. Complete amino acid sequence of rabbit muscleglycogen phosphorylase. Proc Natl Acad Sci U S A 1977;74:4762–6.

w x11 Zarkadas CG, Smile LB, Madsen NB. Sulphydryl groups ofmuscle phosphorylase. Thiol sequences and subunit structure.J Mol Biol 1968;38:245–7.

( )C.D. Karapitta et al.rClinica Chimica Acta 308 2001 99–106106

w x12 Krause EG, Will H, Bohm M, Wollenberger A. The assay of¨glycogen phosphorylase in human blood serum and its appli-cation to the diagnosis of myocardial infarction. Clin ChimActa 1975;58:145–54.

w x13 Rabitzsch G, Noll F, Hofmann U, Krause EG, ArmbrusterFP. Basal concentration of the isoenzyme BB of the glycogenphosphorylase b in human blood. Clin Chim Acta 1993;214:109–11.

w x14 Yoshitake S, Yamada Y, Ishikawa E, Masseyeff R. Conjuga-tion of glucose oxidase from aspergillus niger and rabbit

Žantibodies using N-hydroxysuccinimide ester of N- 4-.carboxycyclohexylmethyl -maleimide. Eur J Biochem

1979;101:395–9.w x15 Fischer EH, Krebs EG. Muscle phosphorylase b. In: Colow-

ick SR, Kaplan NO, editors. Methods in enzymology vol. 5.San Diego: Academic Press, 1962:369–73.

w x16 Kastenschmidt LL, Kastenschmidt J, Helmreich E. Subunitinteractions and their relationship to the allosteric propertiesof rabbit skeletal muscle phosphorylase b. Biochemistry1968;7:3590–608.

w x17 Kohrle J, Rasmussen UB, Rokos H, Leonard JL, Hesch RD.¨Selective affinity labeling of a 27-kDa integral membraneprotein in rat liver and kidney with N-bromoacetyl deriva-tives of L-thyroxine and 3,5,3X-triiodo-L-thyronine. J BiolChem 1990;265r11:6146–54.

w x18 Yamamoto R, Hattori S, Inukai T, Matsuura A, YamashitaK, Kosaka A, et al. Enzyme immunoassay for Thyroxine andTriiodothyronine in human serum, with use of a covalentchromatographic separation method. Clin Chem 1981;27r10:1721–3.

w x19 Helmreich E, Cori CF. The role of adenylic acid in theactivation of phosphorylase. Proc Natl Acad Sci U S A1964;51:131–8.

w x20 Leatherbarrow RJ. GraFit Version 3.0. Staines. UK: ErithakusSoftware, 1992

w x21 Saboori AM, Rose NR, Butscher WG, Burek CL. Modifica-tion of a nonincinerative method for determination of iodinein iodoproteins. Anal Biochem 1993;214:335–8.

w x22 Petrou PS, Kakabakos SE, Koupparis MA, Christofidis I.Antibody coating approach involving gamma globulins fromnon-immunized animal and second antibody antiserum. J

Ž .Immunoassay 1998;19, 4 :271–93.w x23 Bland JM, Altman DG. Statistical methods for assessing

agreement between two methods of clinical measurement.Lancet 1986;i:307–10.

w x24 Hobbard SJ, Thornton JM. NACCESS, computer program.Department of Biochemistry and Molecular Biology, Univer-sity College London, 1993

w x25 Collaborative Computational Project, Number 4, The CCP4Suite: programs of protein crystallography. Acta Crystallogr1994;D50:760–3.

w x26 Stockigt JR, Lim CF, Barlow JW, Wynne KN, Mohr VS,Topliss DJ, et al. Interaction of furosemide with serumthyroxine-binding sites: in vivo and in vitro studies andcomparison with other inhibitors. J Clin Endocrinol Metab1985;60r5:1025–31.

w x27 Van Lente F, Galen RS. Determination of thyroxine byenzyme immunoassay. In: Maggio ET, editor. Enzyme-Im-munoassay. 5th edn. Boca Raton, FL: CRC Press, 1985:135–53.

w x28 Evans SE, Burr WA, Hogan TC. A reassessment of 8-anilino-1-naphthalene sulphonic acid as a thyroxine bindinginhibitor in the radioimmunoassay of thyroxine. Ann ClinBiochem 1977;14r6:330–4.

w x29 Yu XC, Margolin W. Inhibition of assembly of bacterial cellX Ždivision protein FtsZ by the hydrophobic dye 5,5 -bis- 8-

.anilino-1-naphthalenesulfonate . J Biol Chem 1998;273r17:10216–22.

w x30 Madsen NB, Shechosky S, Fletterick RJ. Site-site interac-tions in glycogen phosphorylase b probed by ligands specificfor each site. Biochemistry 1983;22r19:4460–5.

w x31 Ullman EF, Maggio ET. Principles of homogeneousenzyme-immunoassay. In: Maggio ET, editor. Enzyme-Im-munoassay. 5th edn. Florida: CRC Press, 1985:105–34.

w x32 Chaiken R, Pagano D, Detwiler TC. Regulation of plateletphosphorylase. Biochim Biophys Acta 1975;403:315–25.