antibody class capture assay (acca) for rubella-specific igm antibody

10
Journal of Medical Virology 1055-64 (1982) Antibody Class Capture Assay (ACCA) Rubella -Specific Ig M Antibody M. Isaac and R.A. Payne Public Health Laboratory, Cockett Road, Swansea, Wales Enzyme-linked immunosorbent assays for IgM antirubella were carried out on 1,546 sera, using an IgM capture method with a F (ab’)t conjugate (ACCA). Under the conditions described, sera containing IgM antirubella bound up to 15 times as much enzyme activity as negative specimens. Paired serum specimens from 27 patients, serial serum specimens from 6 pa- tients, and single serum specimens from 15 patients who had had recent rubella were examined by the haemagglutination inhibition test (HAI) in the presence and absence of 2-mercaptoethanol following sucrose density gradient centrifugation (SDGC). ACCA confirmed all the results found with HA1 following SDGC. Spe- cimens were examined from ten patients with congenital rubella; ACCA con- firmed the results found with both immunofluorescence following SDGC and ra- dioimmunoassay. Pre- and post-vaccination specimens from 123 patients who had been vaccinated against rubella were examined. An IgM response could only be demonstrated in the 57 cases when IgG was absent in the first specimen. The specificity of the assay was confirmed by testing 31 serum specimens from rubella immune patients that also contained rheumatoid factor, 163 serum speci- mens from patients with acute infections other than rubella, and 12 serum speci- mens from infants with miscellaneous neonatal abnormalities other than congenital rubella. The ACCA proved a simple, sensitive, and specific test for IgM antiru- bella and the results compared favourably with those obtained by the SDGC technique. Key words: ACCA, IgM antirubella, F (ab’), peroxidase, RF INTRODUCTION The detection of IgM antirubella is the most useful serological method of diagnosing recent infection. It is usually demonstrated after the physical separation of Accepted for publication March 29, 1982. Address reprint requests to M. Isaac, Public Health Laboratory, Cockett Road, Swansea, Wales. 0146-6615/82/1001-0055 $03.00 0 1982 Alan R. Liss, Inc.

Upload: m-isaac

Post on 15-Jun-2016

214 views

Category:

Documents


2 download

TRANSCRIPT

Page 1: Antibody class capture assay (ACCA) for rubella-specific IgM antibody

Journal of Medical Virology 1055-64 (1982)

Antibody Class Capture Assay (ACCA) Rubella -Specific Ig M Anti body M. Isaac and R.A. Payne

Public Health Laboratory, Cockett Road, Swansea, Wales

Enzyme-linked immunosorbent assays for IgM antirubella were carried out on 1,546 sera, using an IgM capture method with a F (ab’)t conjugate (ACCA). Under the conditions described, sera containing IgM antirubella bound up to 15 times as much enzyme activity as negative specimens.

Paired serum specimens from 27 patients, serial serum specimens from 6 pa- tients, and single serum specimens from 15 patients who had had recent rubella were examined by the haemagglutination inhibition test (HAI) in the presence and absence of 2-mercaptoethanol following sucrose density gradient centrifugation (SDGC). ACCA confirmed all the results found with HA1 following SDGC. Spe- cimens were examined from ten patients with congenital rubella; ACCA con- firmed the results found with both immunofluorescence following SDGC and ra- dioimmunoassay. Pre- and post-vaccination specimens from 123 patients who had been vaccinated against rubella were examined. An IgM response could only be demonstrated in the 57 cases when IgG was absent in the first specimen.

The specificity of the assay was confirmed by testing 31 serum specimens from rubella immune patients that also contained rheumatoid factor, 163 serum speci- mens from patients with acute infections other than rubella, and 12 serum speci- mens from infants with miscellaneous neonatal abnormalities other than congenital rubella. The ACCA proved a simple, sensitive, and specific test for IgM antiru- bella and the results compared favourably with those obtained by the SDGC technique.

Key words: ACCA, IgM antirubella, F (ab’), peroxidase, RF

INTRODUCTION

The detection of IgM antirubella is the most useful serological method of diagnosing recent infection. It is usually demonstrated after the physical separation of

Accepted for publication March 29, 1982.

Address reprint requests to M. Isaac, Public Health Laboratory, Cockett Road, Swansea, Wales.

0146-6615/82/1001-0055 $03.00 0 1982 Alan R. Liss, Inc.

Page 2: Antibody class capture assay (ACCA) for rubella-specific IgM antibody

56 Isaac and Payne IgM from IgG by either SDGC [Vesikari and Vaheri, 19681 or exclusion chromatogra- phy [Gupta et al, 19711, followed by comparative HA1 using overnight incubation in the presence and absence of 2-mercaptoethanol(2-ME) [Caul et al, 19741. Alternative- ly, the relative concentration of IgG to IgM may be reduced following treatment with staphylococcal protein A [Ankerst et al, 19741. The first two techniques are time con- suming or require expensive apparatus and are applicable only to small numbers of spe- cimens. Treatment with protein A removes only IgGl, IgG2, and IgG4 and also results in some loss of IgM [Chantler et al, 19761.

Alternatively IgM can be separated more easily from other immunoglobulins by selective binding to plastic material coated with anti-human p chain immunoglobulin (ACCA) [Duermeyer and Van der Veen, 19781. The antiviral activity of the immobi- lised IgM antibody can then be assayed by adding a corresponding viral antigen, which in turn can be detected by antibody labeled with enzyme or by radioisotope as has been described for hepatitis A virus [Duermeyer, Wielaard, and Van der Veen, 1979; Fleh- mig et al, 19791 and rubellavirus [Mortimer et al, 1981a; Diment and Chantler, 1981; Vejtorp, 19811.

The Gilford PRSO processor/reader provided an automatic means of performing the ACCA and was used for the automatic addition to, aspiration from, and spectro- photometric reading of cuvettes. The development and use of an automated ACCA for IgM antirubella is described.

MATERIALS AND METHODS Materials

Rabbit anti-human IgM, code number 10-091, obtained from Dako, Copenha- gen, Denmark; horse radish peroxidase (HRP) (RZ of 3.0) obtained from EAB, Ltd,, Tafarnaubach Estate, Tredegar, Gwent, United Kingdom; goat antirubella immuno- globulin peroxidase conjugate, code number VRU 100-E, normal rabbit serum (NRS) and normal goat serum (NGS) obtained from North East Biomedical Laboratories, PO Box 16, Uxbridge, Middlesex, England; rubella haemagglutinin (a Tween and ether- treated, alkaline extract of rubella infected BHK 21 cells) and control antigen (unin- fected BHK 21 cells treated in the same manner) prepared by the Division of Microbio- logical Reagents and Quality Control of the Public Health Laboratory Service; phosphate buffered saline, 0.01 M, pH 7.2 (PBS); PBS containing 0.05% (v/v) Tween 20 (PBST); enzyme substrate, 2,2/ azino-di (3 ethyl benzthiazolin) 6/-sulfonic acid (ABTS) obtained from Sigma Chemical Co., Poole, Dorset, England (20 mg was dis- solved in 100 ml of 0.05 M citrate phosphate buffer pH 4.0 containing 200 pl of 30% (W/V) hydrogen peroxide); EIA processor/reader (PR50) and Cuvettes, code number 1414x9, both obtained from Gilford Instruments, Corning Ltd., Halstead, Essex, England; bovine serum albumin (BSA) fraction V obtained from Miles Research Laboratories Ltd., Slough, England; normal human sera (NHS), rubella antibody-free when tested by HAI, radial haemolysis (RH) [Kurtz et al, 19801 and IgG ELISA [Voller and Bidwell, 19761; protein A-Sepharose (SPA) CL4B and Sephacryl S300 obtained from Pharmacia Fine Chemicals, Hounslow, Middlesex, United Kingdom; pepsin ob- tained from Sigma Chemical Co., Poole, Dorset, England; Millex-HA filter (0.45 pm) obtained from Millipore (UK), Ltd., Abbey Road, London, United Kingdom.

Specimens and Controls Sera from 57 patients with clinical rubella were examined; 47 had acute rubella

and 10 had congenital rubella. Pre- and post-vaccination specimens from 123 vaccinees were also examined.

Page 3: Antibody class capture assay (ACCA) for rubella-specific IgM antibody

Detection of IgM Antirubella 57 To assess the specificity of the ACCA, sera were tested from patients with infec-

tious mononucleosis, hepatitis B (HBs Ag positive), psittacosis, toxoplasmosis, sys- temic lupus erythematosis, neonatal abnormalities other than congenital rubella, mea- sles, and parainfluenza infections. Other control sera were from patients with high levels of rheumatoid factor (RF) and from patients with autoimmune liver disease. Also included were sera from patients containing more than 50 iu/ml of IgG antiru- bella but in which specific IgM antirubella was not detectable by SDGC.

Preparation of Immunoglobulin Fraction Human sera containing more than 500 iu/ml of IgG antirubella were pooled and

passed through a column of SPA. The IgG subgroups 1, 2, and 4 were bound to the SPA. After washing with PBS, the bound IgG was eluted with 0.1 M acetic acid in 0.15 M NaCl [Coding, 19761. The protein content of the eluted IgG subgroups was mea- sured [Lowry et al, 19511.

Preparation of F (ab92 Fragments Pepsin was added to the eluted IgG at a ratio of 150 (W/W) and the pH adjusted

to 4.5. The Fc region of the IgG was digested for 18 hr at 37°C and the reaction was ter- minated by raising the pH to 7. This pepsin digest was passed through the SPA column that bound undigested IgG. The PBS eluate was concentrated to 1 ml by ultrafiltration and passed through a 45 x 1.5 cm column of Sephacryl S300. Fractions of 2.5 ml were collected and those showing HA1 activity were pooled and concentrated to contain 6 mg/ml. This constituted the F (ab/)z solution.

Preparation of F (ab92 Conjugate Conjugation of the F (ab/)2 and HRP was achieved using the method of Wilson

and Nakane [1978]. HRP (40 mg) was dissolved in 10 ml of distilled water; 2 ml of freshly prepared 0.1 M sodium periodate (NalO,) was added and the solution stirred for 20 min at room temperature. This solution was dialysed against 5 litres of 1 mM sodium acetate buffer, pH 4.4, overnight at 4°C.

The F (ab/)2 solution (60 mg in 10 ml) and the HRP solution were separately ad- justed to pH 9.5 with 0.2 M Na,CO, buffer. The two solutions were mixed immediately and stirred gently for 2 hr at room temperature. The Schiff base formed by the reaction of the HRP-aldehyde and F (abq2 was stabilised by the addition of 4 mg of freshly pre- pared sodium borohydride (Na BH,) in 1 ml water and the mixture was left for 2 hr at 4°C.

The solution was concentrated to 2 ml, dialysed against PBS, passed through a Sephacryl S-300 column (45 x 1.5 cm) and fractions of 2.5 ml volume collected. Absor- bance at 405 nm demonstrated two peaks, the first was peroxidase linked to F (ab/)2 and the second was unconjugated peroxidase. Rubella HA1 tests were performed on the fractions and those showing both antibody and peroxidase were pooled and con- centrated. BSA and merthiolate were added to make a final concentration of 1 'Yo and 0.02%, respectively. The conjugate was sterilised by filtration through a Millex filter, frozen rapidly, and stored at -70°C.

Method Mortimer et a1 [1981a] found that the effect of adding homologous serum to ra-

dioisotope label diluent was to substantially suppress anti-species-mediated uptake of radioactivity to the solid phase. It was therefore decided to investigate the effects of

Page 4: Antibody class capture assay (ACCA) for rubella-specific IgM antibody

58 Isaac and Payne

various concentrations of sera in the diluents used for the ACCA to determine whether a similar reduction in anti-species-mediated uptake of enzyme conjugate (nonspecific binding) could be attained. The ACCA procedure used is described later.

Table I showed that incorporation of NHS in the conjugate diluent reduced the nonspecific binding found when PBST alone was used. A similar decrease in nonspecif- ic binding was shown when NRS was used in the serum diluent. The greatest difference in absorbance between the IgM + ve and IgM - ve sera was obtained with a serum dilu- ent of 2% NRS and a conjugate diluent of 5% NHS.

The optimal dilutions of reagents were determined by chess board titrations using a weakly positive IgM antirubella serum and a negative serum. The greatest difference between IgM + ve and IgM - ve sera was found when the anti-human IgM was used at a dilution of 1/1,000, the test serum at 1/40, the rubella haernagglutinin at 1/20, and the F (ab’)2 conjugate at 1/3,000. All incubation procedures were carried out at ambient temperature of 20°C in a sealed, humid box.

ACCA Procedure Anti-human IgM was diluted in PBS and 200 pl volumes added to cuvettes that

were stored in a sealed, humid box at 4°C for at least 24 hr before use. The required number of cuvettes were washed by the addition and aspiration of PBST, nine times, before adding 200 11 of test or control serum diluted in 2% NRS in PBST. After incu- bation for 3 hr, they were washed as before and 200 pl of rubella haemagglutinin in PBST added. Following overnight incubation, the cuvettes were washed and 200 pl of the F (ab’), antirubella conjugate, diluted in 5% NHS in PBST added, and incubated for 2 hr. The cuvettes were then washed and 250 pl of freshly prepared enzyme sub- strate added. The absorbance value of a control sample of the substrate was adjusted to zero at a wavelength of 415 nm. After incubation for 45 min the absorbances were read and recorded. This procedure was repeated with the same test sera using the antirubella IgG conjugate diluted 1/1,200 in 5% NGS in PBST.

The results were expressed as Test:Negative (T:N) ratios where T was the absor- bance of the test serum and N was the mean absorbance of six negative IgM antirubella control sera (three with and three without IgG antirubella). A ratio of 2 or more was considered to be a positive result.

TABLE I. Effect of Various Serum and Conjugate Diluents on Absorbance Values (A’415)

Percent concentrations in PBST

( N W (NHS) antirubella

Absorbance Values with Serum diluent Conjugate diluent Pooled IgM + ve Pooled IgM - ve

antirubella _____

0 0 0.636 0.212 2 20 0.413 0.286 2 0 0.729 0.301 0 5 0.536 0.170 1 5 0.538 0.164 2 1 0.717 0.210 2 10 0.450 0.119 2 2.5 0.634 0.144 5 5 0.454 0.075 2 5 0.629 0.081

Page 5: Antibody class capture assay (ACCA) for rubella-specific IgM antibody

Detection of IgM Antirubella 59

RESULTS Paired sera from 27 patients with clinical rubella were examined by ACCA and

SDGC. The specimens were collected between 1 and 34 days after onset of the rash. The results obtained are shown in Table 11.

Table I1 shows that IgM antirubella could not be demonstrated by either SDGC or ACCA in all 21 specimens collected within 2 days of onset of the rash. However, IgM antirubella could be demonstrated by both methods in the three specimens col- lected 5 days after the onset of the rash. It also shows that there was complete agree- ment between the results obtained with SDGC and ACCA.

Serial serum specimens from six patients with clinical rubella were examined by ACCA and SDGC. The specimens were collected between 1 and 98 days after onset of the rash. The results obtained are shown in Table 111.

Table 111 shows that IgM antirubella could be detected in one specimen collected as early as 3 days after onset of rash. In three specimens IgM antirubella was still present at 62,63, and 87 days but could not be detected in two specimens at 70 and 98 days after onset of rash.

Serum specimens from 10 patients with congenital rubella and 12 patients with miscellaneous neonatal abnormalities, other than congenital rubella, were examined.

TABLE II. Results of Paired Sera From Patients With Clinical Rubella

Patient ___

1

2

3

4

5

6

7

8

9

10

11

12

13

14

Days after rash

1 14 14 28 8

19 1 7 2

13 1

17 1

14 2

17 1

16 10 34

1 7 1

11 1

15 2

17

_____

ACCA T:N ratio SDGC Patient

1 .O 4.0 5.0 9.0 5.8 4.0 1.2 9.0 1.4 5.8 1 .O 2.3 1 .O 3.8 1.4 5.2 1 .o 4.1 6.3 4.5 1.3 5.0 1.2 2.5 1.2 9.0 1.5 9.4

15 + + 16 + + 17 +

18

19 +

20 +

21 +

22 +

23 + + 24 +

25 +

26 +

27 + +

-

- + -

-

-

-

-

-

-

-

-

ACCA T:N

Days after rash ratio SDGC

2 1.2 - 10 11.6 + 1 1 .o 5 8.4 + 1 1.1

15 6.7 + 1 1 .o

10 4.3 + 1 1.2 -

12 3.7 + 5 3.6 +

15 7.7 + 1 1.3 -

12 2.9 + 1 1.1 -

14 2.8 + 8 4.2 +

16 9.0 + 5 2.8 +

18 5.1 + I 1.1 -

12 4.4 + 2 1 .o -

11 2.7 + 1 I .O

12 5.3 +

-

-

-

-

Page 6: Antibody class capture assay (ACCA) for rubella-specific IgM antibody

60 Isaac and Payne The results were compared with those obtained by Dr. Cradock-Watson using immu- nofluorescence (IF) following SDGC [Cradock-Watson et al, 19791 and by Dr. Mortimer using antibody capture radioimmunoassay (MACRIA) [Mortimer et al, 1981al and are shown in Table IV.

All the results obtained with the three tests were in complete agreement. Of the 12 specimens from patients with neonatal abnormalities other than those of congenital ru- bella, all were negative with ACCA and with IF or MACRIA.

Pre- and post-vaccination specimens from 123 patients who had been vaccinated against rubella were examined. The specimens were tested for IgG antirubella by RH and ELISA and for IgM antirubella by ACCA and the results are shown in Table V. The table shows clearly that there was an IgM response only in those 57 patients with- out detectable IgG antirubella.

The ACCA T:N ratios of the sera from the 57 patients were tabulated according to the time after vaccination and the results are shown in Table VI. These results sug- gest that the peak of the IgM antirubella response following vaccination was between 6 and 9 wk.

To assess the specificity of ACCA, sera containing both RF and IgG antirubella, sera from patients with autoimmune liver disease containing antinuclear factor (ANF), together with sera from patients with various diseases were tested using both the F (abOz and the IgG conjugate and the results are shown in Table VII. Table VII shows that there were false positive results with 11 of the sera containing RF when the IgG conjugate was used but there were no false positive results when the F (abOz conjugate was used. There were no false positive results with sera in any of the other categories.

Further investigations of the specificity of ACCA were carried out. Four sera with IgM antirubella, three sera with IgG but no IgM antirubella, and three sera with

TABLE III. Results of Serial Serum Specimens From Patients With Clinical Rubella

ACCA T:N

Patient Days after rash ratio SDGC

28 3 2.1 + 5 2.8 +

10 5.1 + 21 2.9 +

29 1 I .2 - 10 4.0 + 62 11.4 +

30 1 1.2 - 10 3.3 + 29 6.9 +

31 1 1 .0 - 4 2.7 +

10 4.7 + 87 3.1 +

32 5 8.9 + 21 6.3 + I0 1.8 -

33 1 1 .O - 5 9.5 +

63 8.4 + 98 1.8 -

Page 7: Antibody class capture assay (ACCA) for rubella-specific IgM antibody

Detection of IgM Antirubella 61 TABLE IV. Comparison of Results Obtained by IF, MACRIA and ACCA From Patients With Coneenital Rubella

Patient

R A

R B M B J C M F A F W H S M V H J H

Serum specimen ____________

cord 3/12

7/12 newborn cord cord newborn I Yr 33 days newborn 7/12

1 yr

IF

+ + 0 + + + + + + + + +

___- MACRIA ACCA T:N ratio

+ 15.4 + 15.3 0 1.2 + 7.5 + 11.7 + 8.9 + 5.3

NT 6.8 + 9.5 + 2.5 + 8.4 + 8.3

TABLE V. Prevaccination Immunity Status and IgM Response

Prevaccination _ _ _ _ ~ immunitv status Number tested Prevaccination Postvaccination

Presence of IgM

IgG + ve IgG + ve IeG - ve

63 3

57

0 3 0

0 3

57

RF and IgG antirubella were subjected to SDGC. Six fractions of equal volume were collected and tested, in the presence and absence of 2-ME, by ACCA, using the F (ab/)2 conjugate. The results of those sera containing IgM antirubella are shown in Table VIII. The ratios of the 3 RF + ve sera and the 3 IgG + ve antirubella sera were < 1 .O in all six fractions. Table VIII shows that the test detected specific IgM antirubella that was significantly reduced after treatment with 2-ME. The presence of RF or IgG antiru- bella did not cause false positive results.

DISCUSSION Goat antirubella serum (North East Biomedical) was initially used for the prepa-

ration of F (ab/)2 conjugates but did not prove entirely satisfactory because some batches reacted with RF. Precipitation of IgG with ammonium sulphate sometimes caused aggregates of IgG resulting in a conjugate that reacted with RF [Mortimer et al, 1981al. Separation of IgG from goat serum with the SPA column was unsatisfactory [Duhamel et al, 19801. High titre human IgG antirubella serum was readily available and the IgG could be very simply purified using the SPA column. Following pepsin di-

TABLE VI. ACCA T:N Ratios Obtained at Various Times After Vaccination

Weeks after vaccination Number tested Mean T:N ratio Range T:N ratio

3-5 29 4.1 2.2- 10.3 2.3-14.1

10-12 16 3.7 3.1- 7 . 4 6-9 12 5.1

Page 8: Antibody class capture assay (ACCA) for rubella-specific IgM antibody

62 Isaac and Payne

TABLE VII. ACCA Results With F (ab), and IgC Conjugates

T:N T:N ACCA, ratio range

Category Number tested F (ab’)2 conjugate ACCA, ratio range

IgG conjugate _ ~ _ _ _ _ _ _ Infectious

mononucleosis 74 Hepatitis B 1s Psittacosis 7 Antitoxoplasma IgM + ve 64

1.0 - 1.3 1.0 - 1.2 1.1 - 1.2 1.1 - 1.2

1.0 - 1.4 1.1 - 1.3 1.0 - 1.2 1.1 - 1.3

Systemic lupus erythematosis 1 1 .o 1.1

Measles 2 1.0 - 1.3 1 .o Parainfluenza 3 1 1.1 1.1 Antirubellafve (RF+ ve) 31 1.0 - 1.3 (20) 1.0 - 1.3

(11) 2.1 - 19.8

Antinuclear factor 5 1.0 - 1.1 1.0- 1.1 Antirubella+ ve (IgM - ve) 986 1.0 - 1.3 1.1 - 1.4

TABLE VIII. ACCA T:N Ratios After SDGC of Sera Containing IgM Antirubella

Fraction number ~~

Serum 1 2 3 4 5 6

A 3.2 3.8 4.8 3.1 1 .o < 1.0 + 2ME <1.0 < 1.0 <1.0 < 1.0 <1.0 < 1.0

B 3.0 3.0 2.9 2.0 1 .o < 1.0 +2ME <1.0 <1.0 < 1.0 < 1.0 <1.0 < 1.0

C 2.5 2.6 2.5 1 .o < 1.0 < 1.0 +2ME <1.0 < 1.0 < 1.0 < 1.0 <1.0 < 1.0

D 2.5 2.1 1.5 1 .o < 1.0 < 1.0 + 2ME < 1.0 < l . O < 1.0 < 1.0 < 1.0 < 1.0

_ _ ~

gestion of the purified IgG, the SPA column gave excellent separation of undigested IgG from F (ab‘),, which resulted in conjugates that did not react with RF.

During the development of the ACCA, it was found that when PBST alone was used for the dilution of reagents there were high absorbance values with both positive and negative sera. The addition of BSA to any of the diluents failed to reduce the high absorbance levels with negative sera. The inclusion of NRS in the test serum diluent and NHS in the conjugate diluent greatly reduced nonspecific absorbance. Other authors have reported on the efficacy of adding homologous serum to diluents [Hunter and Budd, 19801. The addition of NRS to the test serum diluent probably neutralised anti- species-mediated antibody present in the test serum.

All of the ACCA results on sera from patients with clinical rubella reported in Table I1 were confirmed by SDGC. There were no false positive or false negative re- sults. Although serial specimens were obtained from some patients (Table 111), they were not taken at a time interval suitable for accurately assessing the duration of the IgM antirubella response after onset of rash. However, it was possible to demonstrate IgM antirubella as early as 3 days and as late as 87 days after onset of rash. Vejtorp [1981], using an ELISA antibody capture technique, found that the medium duration of the IgM antirubella reponse was 60 days with a range of 24-140 days.

Page 9: Antibody class capture assay (ACCA) for rubella-specific IgM antibody

Detection of IgM Antirubella 63

It has been reported that indirect RIA on whole serum was a more sensitive means of detecting IgM antirubella in patients with congenital rubella than IF follow- ing SDGC. However, the indirect RIA gave low titre false positive results with some sera containing RF [Cradock-Watson et al, 19791. The problem of false positive results in RIA owing to RF was overcome by MACRIA [Mortimer et al, 1981al but the hand- ling of radioactive compounds required special precautions in the laboratory. As can be seen in Table IV, ACCA was as sensitive and specific as both IF following SDGC and MACRIA for detecting the presence of IgM antirubella in patients with congenital ru- bella. The problem of false positive results with sera containing RF has been overcome in ACCA by the use of F (ab’)z conjugate. Table VII showed that of 3 1 sera containing RF, all were negative when tested with the F (ab/)2 conjugate, whereas 11 gave false positive results when tested with the IgG conjugate. All five sera containing ANF and the sera from patients with various diseases, including one with systemic lupus erythe- matosis, gave negative results with both conjugates. Table VIII clearly shows that the ACCA was detecting specific IgM antirubella.

The levels of IgM antirubella present in postvaccination sera were reported as be- ing considerably lower than those found following naturally acquired infection [Di- ment and Chantler, 19811. Other authors have reported comparable titres with both postvaccination sera and sera from cases of naturally acquired infection [Goldwater and Banatvala, 19811. The method used was a solid phase immunosorbent haemad- sorption, which gave somewhat inconsistent results. The results obtained with ACCA suggested that the levels of IgM antirubella following vaccination were similar to those following naturally acquired infection. The longest period that IgM antirubella could be detected in postvaccination specimens by ACCA was 119 days.

An IgM antirubella response could be demonstrated in the postvaccination speci- men only when IgG was absent in the prevaccination specimen. Sera from three vacci- nees gave positive ACCA results with both pre- and post-vaccination specimens and had high levels of IgG antirubella. Sufficient serum for SDGC was available from only one vaccinee and IgM antirubella was demonstrated by this method in both specimens, Attempts to obtain further clinical details were unsuccessful and the probability of in- fection with rubella virus before vaccination could not be confirmed. Mortimer et a1 [1981b] reported that a vaccine induced IgM antirubella response could only be de- tected in the postvaccination serum if the prevaccination serum was free of IgG antiru- bella when tested by RH, HAI, and ELISA. The results reported here confirm his findings.

There is a need for a reliable, specific test that can detect IgM antirubella. ACCA, using F (ab/)* conjugates, fulfills this need. Treatment of sera to remove nonspecific in- hibitors is unnecessary and only 5 ~1 volumes are required. ACCA is of particular value in the routine testing of sera from patients with suspected rubella infections and their contacts. The PR50 processor/reader, using the ACCA procedure, makes possible the mass screening of antenatal patients for rubella infection and has been adapted for the detection of IgM antitoxoplasma [Payne, Isaac, and Francis, 19821. The development of an ACCA for IgM anticytomegalovirus will enable the mass screening of antenatal patients for the main causes of congenital infections.

ACKNOWLEDGMENTS

We are grateful to Dr. J. Cradock-Watson for supplying the specimens from con- genital cases, to Dr. P. Mortimer for the MACRIA results, and to the Public Health

Page 10: Antibody class capture assay (ACCA) for rubella-specific IgM antibody

64 Isaac and Payne

Laboratories at Coventry, Preston, Norwich, Southampton, and Tooting for addition- al sera. Thanks are extended to Dr. W. Kwantes for constructive criticism in the prepa- ration of the manuscript and to Louise Davies for cheerfully typing it.

REFERENCES

Ankerst J , Christensen P, Kjelltn A, Kronvall G (1974): A routine diagnostic test for IgA and IgM anti- bodies to rubellavirus: Absorption of IgG with Staphylococcus aureus. Journal of Infectious Dis- eases 130:268-273.

Caul EO, Smyth GW, Clarke SKR (1974): A simplified method for the detection of rubella specific IgM employing sucrose density fractionation and 2-mercaptoethanol. Journal of Hygiene (Cambridge)

Chantler SM, Devries E, Allen PR, Hurn BAL (1976): A rapid immunofluorescent procedure for the de- tection of specific IgG and IgM antibody in sera using Staphylococcus aureus and latex-IgG as ab- sorbents. Journal of Immunological Methods 13:367-380.

Cradock-Watson JE, Ridehalgh MKS, Pattison JR, Andersen MJ, Kangro HO (1979): Comparison of im- munofluorescence and radioimmunoassay for detecting IgM antibody in infants with the congenital rubella syndrome. Journal of Hygiene (Cambridge) 83:413-423.

Diment JA, Chantler SM (1981): Enzyme immunoassay for detection of rubella specific IgM antibody. Lancet 1 :394-395.

Duhamel RC, Meezan E, Brendel K (1980): The pH dependent binding of goat IgG and IgG2 to protein A- Sepharose. Molecular Immunology 17:29-36.

Duermeyer W, Van der Veen J (1978): Specific detection of IgM-antibodies by ELISA, as applied in hepa- titis-A. Lancet 2:684.

Duermeyer W, Wielaard F, Van der Veen J (1979): A new principle for the detection of specific IgM anti- bodies applied in an ELISA for hepatitis A. Journal of Medical Virology 425-32.

Flehmig B, Ranke M, Berthold H, Gerth H-J (1979): A solid phase radioimmunoassay for detection of IgM antibodies to hepatitis A virus. Journal of Infectious Diseases 140: 169-175.

Goding JW (1976): Conjugation of antibodies with fluorochromes: Modifications to the standard meth- ods. Journal of Immunological Methods 13:215-226.

Goldwater PN, Banatvala JE (1981): Solid-phase immunosorbent haemadsorption (SPIHAD): A new, rapid, and simple test for rubella-specific IgM. Journal of Virological Methods 2:341-347.

Gupta JD, Peterson V, Stout M, Murphy AM (1971): Single-sample diagnosis of recent rubella by frac- tionation of antibody on Sephadex G-200 column. Journal of Clinical Pathology 24:547-550.

Hunter WM, Budd PS (1980): Circulating antibodies to ovine and bovine immunoglobulin in healthy s u b jects: A hazard for immunoassays. Lancet 2: 1136.

Kurtz JB, Mortimer PP, Mortimer PR, Morgan-Capner P, Shafi MS, White GBB (1980): Rubella anti- body measured by radial haemolysis: Characteristics and performance of a simple screening method for use in diagnostic laboratories. Journal of Hygiene (Cambridge) 84:213-222.

Lowry OH, Rosebrough NJ, Farr AL, Randall RJ (1951): Protein measurement with the Folin phenol re- agent. Journal of Biological Chemistry 193:265-275.

Mortimer PP, Tedder RS, Hambling MH, Shafi MS, Burkhardt F, Schilt U (1981a): Antibody capture ra- dioimmunoassay for antirubella IgM. Journal of Hygiene (Cambridge) 86: 139-153.

Mortimer PP, Edwards JMB, Porter AD, Tedder RS, Mace JE, Hutchinson A (1981 b): Are many women immunised against rubella unnecessarily? Journal of Hygiene (Cambridge) 87: 131-138.

Payne RA, Isaac M, Francis JM (1982): An enzyme-linked immunosorbent assay (ELISA) using antibody class capture for the detection of antitoxoplasma IgM. Journal of Clinical Pathology (in press).

Vejtorp M (1981): Solid phase anti-IgM ELISA for detection of rubella specific IgM antibodies. Acta Pathology Microbiology Scandinavia Section B 89: 123-128.

Vesikari T, Vaheri A (1968): Rubella: A method for rapid diagnosis of recent infection by demonstration of the IgM antibodies. British Medical Journal 1:221-223.

Voller A, Bidwell DE (1976): Enzyme-immunoassays for antibodies in measles, cytomegalovirus infec- tions, and after rubella vaccination. British Journal of Experimental Pathology 57:243-247.

Wilson MB, Nakane PK (1978): Recent developments in the periodate method of conjugating horse radish peroxidase to antibodies. In Knapp W (ed): “Immunofluorescence and related techniques.” Amster- dam: Elsevier, pp 215-224.

73:329-340.