Journal of Virological Methods, 4 (1982) 249- 251

Elsevier Biomedical Press

249

DETERMINATION OF IgG- AND Ighi-CLASS ANTIBODIES TO MUMPS VIRUS BY

SOLID-PHASE ENZYME IMMUNOASSAY

OLLI MEURMAN’,3, PENTTI HANNINEN’ , RANGACHAR V. KRISHNA3 and THEDI ZIEGLER’,”

1 Department of Virology, and ’ Department of Infectious Diseases, University of Turku, SF-20520

Turku 52, Finland; and 3 Institute of Medical Microbiology, CH- 9000 St. Gallen, Switzerland

(Accepted 3 February 1982)

An indirect enzyme immunoassay (EIA) for the determination of IgG and IgM antibodies to

mumps virus is described. Viral antigens and control antigens were adsorbed onto polystyrene micro-

titer plates, and antibodies attached to the antigens were detected by subsequent binding of com-

mercial peroxidase-labeled antibodies to the heavy chains of human IgG and IgM immunoglobulins.

A comparison of antibody titers obtained by the EIA and by indirect immunofluorescence test showed

a close concordance between these two tests, with EIA, however, being more sensitive. Occasional cross-

reactions between mumps and parainfluenza antibodies were detected in the IgG antibody test but

not in the IgM antibody test. In sera from 84 patients with mumps infection, all cases were diagnosed

by the EIA IgM antibody assay, 96% from the fist serum specimen. Mumps was diagnosed by comple-

ment fixation (CF) in 71% of these cases: unclear or erroneous results with parainfluenza titer in-

creases in 10% and no diagnosis in 18% of the cases. The EIA IgM antibody assay was thus better than

the CF test for the diagnosis of acute mumps infection.

indirect enzyme immunoassay IgG IgM mumps antibodies

INTRODUCTION

Although the diagnosis of mumps infection in typical cases of parotitis can usually

be made without laboratory tests, a rapid and reliable laboratory diagnosis is important

for the differential diagnosis of other manifestations of mumps infections, such as men-

ingoencephalitis, pancreatitis, or orchitis which often occur without any signs of paro-

titis.

The demonstration of specific IgM antibodies has been shown to be a practical method

for rapid diagnosis in several acute viral infections, including mumps (Nicolai-Scholten

et al., 1980; Ukkonen et al., 1980). In the present report we describe a simple indirect

microtiter plate enzyme immunoassay (EIA) for the detection of IgG and IgM antibodies

to mumps virus. The EIA test was applied to the serological diagnosis of mumps virus

infections and compared with the complement fixation (CF) test with respect to sen-

sitivity and specificity.

0166-0934/82/0000-0000/$02.75 @ 1982 Elsevier Biomedical Press

250

MATERIALSANDMETHODS

Patients and sera The study material comprised the following groups: 1) 111 serum specimens from

healthy medical students; 2) paired serum specimens from 84 patients with clinically

typical mumps infection; 3) serial serum specimens from three patients with mumps

infection; 4) paired serum specimens from 10 patients with parainfluenza virus infection

verified by viral antigen detection by immunofluorescence (Gardner and McQuillin,

1980); 5) 22 serum specimens collected from patients with rheumatoid arthritis and

containing rheumatoid factor (RF).

Antigens Mumps virus (a wild strain isolated in this laboratory) was grown in African green

monkey kidney (Vero) cells. Eagle’s minimum essential medium (EMEM) without serum

was used as maintenance medium. When a cytopathic effect (CPE) was seen in 75% of

the cells, the cells were washed with cold phosphate-buffered saline (PBS), pH 7.4, and

scraped into PBS. The cells were disrupted with ultrasonic treatment, crude cell debris

removed by low-speed centrifugation and the supernatant centrifuged for 2 h at 25,000

r.p.m. in a Beckman SW 27.1 rotor. The pellet was resuspended in PBS and used as anti-

gen in EIA. The optimal dilution of each antigen batch was determined by box titra-

tion with a known positive and a known negative serum. Control antigen was prepared

in a similar way from uninfected Vero cells and diluted to the same protein content as

the virus antigen. Antigen and control antigen were prepared in the same way from para-

influenza type 1 (Sendai) virus grown in Madin-Darby canine kidney (MDCK) cells

and from parainfluenza type 2 and 3 viruses grown in Vero cells.

EIA procedure The antigens and corresponding control antigens were diluted in PBS and 75 ~1 ah-

quots of the antigen suspension incubated in the wells of flat-bottom polystyrene micro-

titer plates overnight at room temperature. After incubation, the antigen suspensions

were aspirated off and the wells allowed to dry in air. The plates were stored at t4”C

and washed with PBS before use.

Seventy-five ~1 aliquots of test sera (four-fold dilutions starting at 1 : 40) were in-

cubated in the antigen and control antigen-coated wells for 2 h at 37°C. A positive and

a negative control serum were included on each plate. After washing with PBS contain-

ing 0.1% Tween 20, 75 /..d aliquots of a 1 : 500 dilution of peroxidase-conjugated swine

anti-human-IgG or -1gM (heavy chain-specific, Orion Diagnostica, Finland) were pipetted

into the wells and incubated for 2 h at 37°C. PBS containing 5% normal porcine serum

and 0.5% Tween 20 was used as diluent for both test sera and conjugated anti-human

immunoglobulins. After washing as before, 75 ~1 aliquots of freshly prepared substrate

(1,2_phenylenediamine, 1 mg/ml, with 0.03% hydrogen peroxide in citrate-phosphate

buffer, pH 5.5) were added. The microtiter plates were incubated in the dark for 30

251

min at room temperature and 150 ~1 aliquots of 4 M sulphuric acid were added to each

well to stop the reaction. The absorbances of the solutions in each well were measured

directly on the plate with a vertically measuring photometer (Titertek Multiskan, Eflab,

Finland) at 492 nm. The end-point titer was regarded as the highest serum dilution where

the absorbance in the virus antigen-coated well was 2.1 times that in the control antigen-

coated well with the proviso that the absorbance value in the virus antigen-coated well

should be at least 0.1.

Indirect immunofluorescence test Mumps virus was grown in Vero cells as mentioned above. At 75% CPE, the cells

were trypsinized into EMEM containing 10% calf serum, washed three times with cold

EMEM and resuspended in EMEM to obtain the optimal cell density for the indirect

immunofluorescence test. Ten ~1 of cell suspension were pipetted into each well of the

polystyrene immunofluorescence plates, the cells were allowed to dry in air and were

then fixed in absolute methanol for 10 min at 4°C. For the IgG antibody assay whole

sera were used while for the IgM antibody assay the sera were fractionated by column

chromatography on agarose and the IgM fractions collected (Pyndiah et al., 1977).

Serial dilutions of the test sera or of the IgM fractions were pipetted into the wells and

incubated for 30 min (IgG antibody assay) or for 3 h (IgM antibody assay) at 37°C.

After washing, the wells were incubated with FITC-conjugated anti-human-IgG or

-1gM (Dako, Denmark) for 30 min at 37°C. The highest dilution of serum or IgM frac-

tion showing clear intracytoplasmic fluorescence was taken as the end point-titer.

Measurement and removal of rheumatoid factor RF levels of the serum specimens were determined by EIA according to Ziola and

Tuokko (1980) and the results expressed as I.U./ml using an international reference

preparation.

Serum specimens were absorbed by incubation with latex particles coated with aggre-

gated human IgG (Vejtorp, 1980).

RESULTS

In both IgG and IgM antibody assays the negative and positive sera produced suffi-

ciently low absorbance values with the control antigen, especially at dilutions of 160

and higher (Fig. 1). Almost similar values were noted when negative sera were incubated

with mumps antigen, although it was a rather constant finding that the negative sera

also gave somewhat higher absorbance values with the mumps antigen than with the

control antigen. A serum which was mumps antibody-positive commonly gave 5-30

times higher absorbance values with the mumps antigen than with the control antigen.

The distribution of mumps IgG and IgM antibodies in sera of 111 healthy medical

students is shown in Table 1. None of the medical students had IgM antibodies to mumps.

IgG antibodies were found in 83% of the sera, whereas 17% of the young adults studied

had no antibodies at the serum dilution 1 : 40.

252

5 1. Y

1.1

0..

t

IgG 1.

.

\ l.l

.

2 4 6 8

.

\, .

‘CtM

Fig. 1. Representative results of mumps IgG (left panel) and IgM (right panel) antibody test obtained

when a positive and a negative serum were tested with mumps antigen and with control antigen-

coated plates. Positive serum, mumps antigen (0). Positive serum, control antigen (0). Negative serum,

mumps antigen (w). Negative serum, control antigen (0).

TABLE 1

Distribution of mumps IgG and IgM antibody titers in sera from 111 healthy medical students mea-

sured by EIA

Class of antibody Number of sera having a mumps EIA titer of

<40 40 80 160 320 640 1280 2560 5120

IgC 19 4 9 10 22 25 14 6 2

IgM 100 0 0 0 0 0 0 0 0

The sensitivity of the EIA test was compared to that of the indirect immunofluor-

escence test by parallel examination of 36 serum specimens from patients with acute or

remote mumps infection. The EIA titers obtained were about lo-30 times higher than

the immunofluorescence titers and some additional positives were obtained with EIA.

On the whole, a good agreement between these two tests was observed (Fig. 2).

Rheumatoid factor interference in the EIA IgM antibody assay was studied using 22

sera with variable amounts of RF. None of the mumps IgG antibody-negative sera gave

a false-positive IgM result, irrespective of the amount of RF present. When sufficient

amounts of both RF and mumps IgG antibody were present, false-positive IgM results

were observed. However, in most of the cases these IgM titers were very low and in only

two sera false-positive IgM titers comparable to those detected in acute mumps infections

253

. . . .

. . . . . .

. . . . . .

. . . ::

. . . . . .

:: . . .

~2 2 3 4 5 6 7 8 9 10

EIA IgG titer (log,xlO)

. . . . . ::

. . . . . . .

. . . .

. . . . .

.‘f. . . . . . .

~2 2 3 4 5 6 7 8 9 10

EIA IgM titer (log,xlO)

Fig. 2. Comparison of mumps IgG antibody titers (upper panel) and IgM antibody titers (lower panel)

obtained by enzyme immunoassay (ISA) and by indirect immunofluorescence (IF) in 36 serum speci-

mens. For IgG antibody titers the linear regression T’ = 0.62, and for IgM antibody titers rZ = 0.75.

were obtained. The false-positive results could be avoided by absorption of RF with

aggregated human gamma-globulin (Table 2).

Among 10 patients with parainfluenza virus infection, one (with a type 2 infection)

TABLE 2

Representative results of rheumatoid factor (RF) interference in EIA for mumps IgM antibodies

Patient Mumps

Igc; titer

RF units (I.U./ml) Mumps IgM titer

Before After Before After

absorption absorption absorption absorption

1 < 40 184 NT < 40 NT

2 160 32 1.3 80 < 40

3 2560 30 0.5 320 < 40

4 640 59 2.2 160 < 40

5 2560 126 0.6 320 < 40

6 640 41 0.5 1280 < 40

1 640 553 5.5 1280 < 40

254

showed a significant IgG titer increase to mumps virus. None of the patients had IgM

antibodies to mumps virus.

The appearance and persistence of mumps antibodies were studied by testing serial

serum specimens from three patients with a natural mumps infection. IgM antibodies

reached a maximum level in about one week and started to decline within one month.

All patients still had low levels of IgM antibodies (titers 160-640) when they were

lost from the follow-up about 4 months after the onset of the disease. IgG antibodies

reached maximum levels in l--2 weeks and thereafter remained constant during the

period of study.

The present EIA test was compared with the CF test as a routine diagnostic tool by

testing paired serum specimens (taken 7-21 days apart) from 84 consecutive clinically

typical patients with mumps seen at the Department of Infectious Diseases, University

of Turku, between 1975 and 1977. A significant (four-fold or higher) increase in mumps

CF antibody titer was detected in 67 patients (80%) and, using the mumps EIA IgG anti-

body titer, in 50 patients (60%). If the acute-phase serum specimen was taken during days

O--6 after the onset of illness, the CF test detected increases in 90% and EIA IgG test

in 75% of the patients. Of the acute-phase specimens 32 were negative in the CF test

against only seven in the EIA IgG test. Mumps IgM antibodies were detected by EIA

in all 84 patients and in 81 cases in the first serum specimen examined. The three acute-

phase specimens which were IgM negative had been taken 2,2 and 7 days after the onset

of the disease, respectively.

In the CF test, seven patients with a signi~cant rise in titer to mumps virus there

were simultaneous CF titer increases to one or more parainfluenza viruses. Further-

more, two patients had CF titer increases to parainfluenza 2 virus but not to mumps

virus. All these nine patients had IgM antibodies to mumps. When eight of these patients

were tested for IgG and IgM antibodies to parainfluenza viruses by EIA, IgG antibody

titer increases to parainfluenza types 1, 2, and 3 were detected in 2, 6, and 3 cases, re-

spectively, whereas IgM antibodies to parainfluenza viruses were not detected (Table 3).

DISCUSSION

The serological diagnosis of mumps infections is based mostly on the demonstration

of a signi~cant titer increase between acute and convalescent phase serum specimens

by the CF test. The known disadvantages of this method are the requirement of paired

serum specimens which makes the diagnosis slow, the missing of titer increases by de-

lay in collecting the acute-phase specimen, and the cross-reactions between mumps

and parainfluenza viruses (Lennette et al., 1963). In our 84 patients, the CF test gave

a mumps diagnosis in 60 patients (71%) an unclear result with titer increases to both

mumps and parainfluenza viruses in seven patients (8%), an erroneous parainfluenza

diagnosis in two patients (2%) and no diagnosis at all in 15 patients (18%) the latter

being mostly cases where the acute-phase serum specimen was obtained during the

second week after the onset of illness or later.

255

TABLE 3

Mumps and parainfluenza virus CF and EIA IgG and IgMa antibody titers in paired serum specimens

from mumps patients with significant CT: titer increases to parainflucnza viruses

Patient Days Antibody titer

and

age tyr)

after Mumps Para 1 Para 2 Para 3

onset CF IgG LgM CF IgG CF IgG CF fgG

V.T. (8) 9 64 1280 5120

19 256 5120 20480

E.P. (11) 5 <4 1280 5120

15 64 5120 10240

S.T. (15) 9 16 5120 5120

21 64 5120 10240

A.T. (16) 2 32 640 2560

22 64 1280 2560

P.H. (17) 9 64 10240 20480

23 64 10240 20480

VS.-L. (27) 8 32 2560 2560

16 128 2560 5120

H.U. (40) 9 < 4 320 1280

21 32 1280 1280

E.V. (41) 5 < 4 1280 5120

15 64 5120 10240

<4 320 16 2560 16 1280

4 320 128 10240 16 2560

<4 160 4 80 16 10240

32 640 8 320 32 10240

4 320 16 5120 8 2560

16 1280 16 20480 16 5 120

<4 320 <4 2560 < 4 1280

<4 640 16 10240 < 4 2560

16 640 8 10240 16 5120

32 1280 32 20480 32 5120

4 1280 64 10240 64 5120

32 2560 256 40960 512 20480

16 320 <4 1280 16 2560

256 320 16 2560 256 10240

4 160 32 2560 < 4 5120

8 160 128 20480 4 20480

a IgM antibodies to parainfluenza viruses were negative and are not presented in the table.

The EIA IgG test was not superior to the CF test, since the more rapid appearance

of EIA IgG antibodies made it even more difficult to detect significant titer elevations

between acute and convalescent phase serum specimens. In addition, the EIA IgG anti-

body test was hampered by similar cross-reactions between mumps and parainfluenza

viruses as the CF test.

On the other hand, the EIA IgM antibody test gave a clear diagnosis of mumps in

all 84 patients, and in 81 patients (96%) from the first available serum specimen. Cross-

reactions between mumps and parainfluenza viruses seem not to interfere in the IgM

antibody assays, since neither mumps IgM antibodies in parainfluenza infections nor

parainfluenza IgM antibodies in mumps infections were detected. This is in agreement

with the results of Nicolai-Scholten et al. (1980) and of Ukkonen et al. (1980) who did

not observe false-positive mumps IgM antibody reactions in sera obtained from 23 and

12 patients with parainfluenza infections, respectively. Also, cases where the collection

of specimens has been delayed can be easily diagnosed since mumps IgM antibodies

persist for at least 3-4 months. This rather long persistence of IgM antibodies can, how-

ever, cause difficulties in the timing of the infection by means of the IgM antibody

titer, and careful attention must be paid to the clinical history of the patient, as pointed

out by Nicolai-Scholten et al. (1980).

IgM rheumatoid factor can cause false-positive IgM antibody results in indirect im-

munoassays, e.g. in rubella serology (Meurman and Ziola, 1978, Vejtorp, 1980). For the

diagnosis of mumps, however, this is usually not a serious problem since the patients

are mostly children and young adults without systemic disease and without detectable

RF activity. When false-positive results occur, the titers are mainly low and can be con-

trolled by re-testing after removal of the RF by absorption procedures.

The EIA IgG antibody assay, although not practical for the diagnosis of acute mumps

infections, has been shown to be a sensitive and reliable method for the determination

of mumps immunity (Leinikki et al., 1979). Although in some parainfluenza infections

antibodies which react in the mumps IgG assay are produced, these are mainly of low

titer and possibly transient in nature. The complete agreement between mumps EIA

IgG antibody assay and the neutralization test, reported by Leinikki et al. (1979), as

well as our finding of a lack of immunity to mumps in 17% of adults, although as a

result of childhood infections practically all adults have antibodies to at least one of the

parainfluenza viruses (La Placa and Moscovici, 1962) clearly speak against a serious

interference by parainfluenza antibodies in the mumps EIA IgG antibody assay.

ACKNOWLEDGEMENTS

The excellent technical assistance of MS Kaija Johansson is gratefully acknowledged.

This study was supported by a grant from the Emil Aaltonen Foundation, and from

the Sigrid Juselius Foundation, Finland.

REFERENCES

Gardner, P.S. and J. McQuilLin, 1980, Rapid Virus Diagnosis. Application of Immunofluoresence,

2nd ed. (Butterworths, London).

La Placa, M. and C. Moscovici, 1962, J. Immunol. 88, 72.

Leinikki, P., I. Shekarchi, N. Tzan, D.L. Madden and J.L. Sever, 1979, Proc. Sot. Exp. Biol. Med.

160, 363.

Lennette, E.H., F.W. Jensen, R.W. Guenther and R.L. Magoffin, 1963, J. Lab. Clin. Med. 61, 780.

Meurman, O.H. and B.R. Ziola, 1978, J. Clin. Pathol. 31,483.

Nicolai-Scholten, M.E., R. Ziegelmaier, F. Behrens and W. Hopken, 1980, Med. Microbial. Immunol.

168,81.

Pyndiah, N., P. Price, M. Jung, J. Wilhelm and U. Krech, 1977, Experientia 33, 1678.

Ukkonen, P., 0. Vaisanen and K. Penttinen, 1980, J. Clin. Microbial. 11, 319.

Vejtorp, M., 1980, J. Virol. Methods 1, 1.

Ziola, B. and H. Tuokko, 1980, Acta Pathol. Microbial. Stand. 88C, 127.