Journal of Viroiogicai Methods, 16 (1987) 65-74 Elsevier

JVM 00581

Measurement of polypeptide- and antigenic site- specific antibodies to measles virus using a

competitive enzyme immunoassay

Mika J. MikehI, Erling Norrby* and Aimo Salmi”

‘Viral Pathogenesis Research Unit, Department of Medical Microbiology and Infectious Diseases, University of Alberta, Edmonton, Alberta, Canada; ‘Department of Virology, Karolinska Institute,

Stockholm, Sweden

(Accepted 8 January 1987)

Summary

A competitive enzyme immunoassay (EIA) for measuring polypeptide- and an- tigenic site-specific antibodies against measles virus (MV) was developed. Three monoclonal antibodies (MAbs) against the fusion (F) and one MAb against the hemagglutinin (H) protein of MV were labelled with biotin and used in bio- tin-streptavidin EIA system. The inhibition of the binding of the labelled MAbs by the bound serum samples was measured in EIA and the titers were defined by reading at the level of 25% inhibition. The titers increased from acute to early convalescent measles sera in all the tests and antibody levels to the F protein in- creased earlier than antibodies to the H protein. The ratio of antibodies against 2 different antigenic sites on the F protein was signi~cantly higher in subacute scle- rosing panencephalitis patients than in normal blood donors. Similarly, SJL mice immunized with MV antigen had higher ratios of the corresponding antibodies than Balb/c mice. The results indicate that antibody response to different polypeptides and antigenic sites on the same polypeptide can be studied with the developed method and that the responses against different antigenic sites are independent.

Measles; Competitive EIA; Monoclonal antibody; Site-specific response

Correspondence address: A. Salmi, Department of Medical Microbiology and Infectious Diseases. University of Alberta, Edmonton, Alberta T6G 2H7, Canada.

~166-093~/87~$03.50 @ 1987 Elsevier Science Publishers B.V. (Biomedical Division)

66

introduction

Measles virus (MV), a member of Paramyxoviridae, usually causes a febrile dis- ease associated with exanthem followed by recovery and life-long immunity. It is also known to cause chronic infections such as subacute sclerosing panencephalitis (SSPE), a rare and fatal central nervous system disease occurring years after the acute disease. However, the mechanisms leading to such persistency are poorly known and no generally accepted mechanism has been described.

Antibodies have been shown to play a significant role in the establishment of MV persistency both in vivo (Rammohan et al., 1981) and in vitro (Fujinami and Oldstone, 1980). It has been recently shown that monoclonal antibodies (MAbs) against certain epitopes on the hemagglutinin (H) protein of measles can change the expression of other structural polypeptides in a cell culture (Fujinami et al., 1984). In many other viral diseases, the importance of antibodies has also been emphasized both in pathogenesis and in protective immunity (Dietzschold et al., 1981; Talbot et al., 1984). Experiments with MAbs have shown that neutralizing antibodies are elicited only against some epitopes on viral surface proteins. This concept is also important in development of component vaccines. Therefore, a de- velopment of assays for measuring antibodies against restricted viral determinants is warranted.

Hemagglutination inhibition and hemolysis inhibition tests have previously been used to measure antibodies against MV H and fusion (F) proteins, respectively. However, these methods are neither sensitive nor component-specific since some hemagglutination inhibiting antibodies also inhibit hemolysis (Norrby and Goll- mar, 1975). Other methods like Western blotting, radioimmunoprecipitation (Graves et al., 1984), and radioimmunoassay (Trudgett et al., 1980) have some in- herent disadvantages reflecting the degree of denaturation of the proteins or the poor quantitation of the antibody amounts,

In the present paper, we describe a method for measuring antibodies to poly- peptides and antigenic sites in sera using a competitive EIA. The specificity of an- tibody response to the H and F proteins of MV was studied in individuals with a typical infection, those with a past history of uncomplicated infection and in SSPE patients using competition immunoassays in which previously characterized mouse MAbs (Sheshberadaran and Norrby, 1986; Sheshberadaran et al., 1983) compete with antibodies in sera for binding to virus antigen on solid phase, The results show that antibody response both to different polypeptides and antigenic sites on the same polypeptide are independent phenomena. This conclusion was further con- firmed by studies of immunization of 2 inbred mouse strains with purified MV an- tigen .

Materials and Methods

Viral antigens

Both purified MV and cell lysate prepared from infected cells were used as viral antigens in EIA. A laboratory MV strain isolated from an acute measles patient

was purified by the method described earlier (Vainionp~~ et al,, 1978). Briefly, Vero cells grown in roller bottles were infected with the wild type MV at the mul- tiplicity of infection of 0.01. When CPE reached 75 to 100%, the medium was har- vested and cell debris removed by low speed centrifugation. The medium was con- centrated with an Amicon hollow fiber apparatus (HlPlOO Cartridges, Amicon Ltd., Surrey, UK) and centrifuged for 2 h at 25000 rpm (Spinco SW 27 rotor) in a 18/36% (w/w) potassium tartrate step gradient. The material at the interface was collected and centrifuged in a linear 18 to 36% (w/w) potassium tartrate gradient for 16 h at 25000 rpm (Spinco SW 27 rotor). The virus material was diluted in GNTE buffer (0.2 M glycine, 0.2 M sodium chloride. 0.02 M Tris-HCl, f1.002 M

EDTA, pH 7.8), pelleted, and resuspended in phosphate-buffered saline (PBS). Cell lysate antigen was prepared as earlier described by Tuokko and Salmi (1983).

MV infected cells were harvested when CPE had just extended through the cell monolayer. The cells were washed and resuspended in PBS and then homogenized in a Sorwall R Omnimixer (Norwalk, CT). The material was centrifuged at 2000 rpm and the supernatant centrifuged for 30 min at 22500 rpm (Spinco SW 27 ro- tor). The pellet was suspended in PBS. The antigens were stored at -70°C and sonicated before use as antigen in EIAs.

MAbs to MV H and F proteins

The production and characterization of the MAbs used in this study have been

described previously (Sheshberadaran and Norrby, 1986; Sheshberadaran et al., 1983). After preliminary experiments, one MAb against H (144) and 3 MAbs against F protein (19 GD6, 9DB10, and 19HB4) were selected to study the spec- ificity of the antibody response to measles H and F proteins and antigenic site-spe- cific response to the F protein.

Purification and biotinylation of MAbs

Immunoglobulin fractions were isolated from ascites fluids by precipitation at u”C (ice bath) with an equal volume of saturated ammonium sulfate (pH 7). Salt was removed from the precipitates by Sephadex 25 column chromatography. The spe- cific antibody fraction was isolated in a Mono-Q anion exchange column in fast performance liquid chromatography system (Pharmacia Ab, Sweden). After over- night dialysis of the antibody fraction against PBS (pH 7.4). immunoglobulin con- centrations were estimated by spectrophotometry and adjusted to 1 mg/ml. N-hy- droxysuccinimidobiotin (Sigma, St. Louis, MO.) was dissolved in 25% N’, N’- dimethylformamide and added to immunoglobulin solutions at a molar ratio of

68

100: 1 (biotin:immunoglobuIin). The mixture was incubated at room temperature for 3 h and dialyzed overnight against PBS. The aliquots were stored at -70°C until used.

Sera included in this study were from 10 adult normal blood donors (NBDs) with a past uneventful measles infection in childhood, acute and early convalescent sera from 10 patients with uncomplicated measles infection (collected at the time of rash onset and 7 to 10 d after the onset of rash), and from 10 SSPE patients. As a neg- ative serum, a pool of sera taken from 1-3-yr-old unvaccinated children without a history of measles was used. This pool did not bind in direct EIA to measles an- tigen.

Inbred strains of Balb ciA and SJL mice were used in this study. Ten mice of each strain were injected S.C. with 50 kg of purified wild type strain of MV in Freund’s complete adjuvant. Blood samples were obtained by bleeding from the retro-orbital plexus under ether anesthesia 7 d before and at Days 7, 14, 21, and 28 after virus injection. Plasma samples were diluted ZO-fold in EIA diluent (see below) and stored at -70°C until tested.

The solid-phase immunoassay principle employing microtiter plates was used throughout. Preliminary studies showed that purified virus was needed for optimal results with anti-F MAbs whereas cell lysate antigen was equal to purified virions and was used when antibodies to the H polypeptide were tested. No significant variation between different antigen batches was found. Microtiter plates were coated with either 300 ngiwell (0.1 ml/well) of purified virions or 700 nglwell of cell lysate antigen. After overnight incubation at room temperature, free binding sites on the plastic plates were blocked with 0.2 ml/well of EIA diluent (PBS sup- plemented with 0.5% bovine serum albumin, 0.5% Tween 20, and 0.1 mM mer- thiolate). After an incubation period of 60 to 90 min at 37“C, the plates were used in the assays.

For competition binding assays, serial 4-fold dilutions (starting from I:200 di- lution) of the sera were incubated in antigen-coated plates overnight at room tem- perature. The plates were washed 3 times with a washing solution (PBS supple- mented with 0.1% Tween 20) and previously determined optimal amount of biotinylated MAbs giving 0Dz8a values of about 1.0 were added. After a 60 min incubation at 37°C the plates were washed as above and horseradish peroxidase (HRPO)-labelled streptavidin (Amersham, UK) was added. After a further 6Ct-min incubation at 37”C, the plates were washed 4 times and 0.1 ml/well of substrate solution (30 mg ~-phenylenediamine plus7 p.1 of 30 % hydrogen peroxide in 10 ml

69

s 100

g 80

3 60

x 2 40

8 20

S

0 0. I I .o IO 100 1000 0. I I .o IO 100 1000

COMPETING ANTIBODY cog / ml ) COMPETING ANTIBODY (fig/ml )

Fig. 1. Homologous and heterologous inhibition of the binding of MAbs to MV antigen. Unlabelled

MAb against MV H (MAb 144)) was incubated on the plates followed by incubation with biotin-la-

belled MAb I44 (0) or 3 MAbs against F protein, 19GD6 (Cl), 9DBltl (O), or 19HB4 (W).

Fig. 2. Inhibition of binding of 2 different MAbs to MV antigen. Unlabelled MAb 19GD6 inhibits the

binding of the labelled homologous antibody (0) but does not significantly inhibit the binding of an-

other MAb (19DBlO) to the same F polypeptide (0).

of 0.1 M citrate buffer, pH 5.5) was added. After 15 to 30-min incubation in dark, the reaction was terminated by adding 0.1 ml/well of 1 N HCl and OD,,, values were read with a Titertek Multiskan photometer (Eflab, Helsinki, Finland).

In competition binding assays, binding of negative serum to viral antigen was taken to represent background and binding of each labelled MAb to antigen as maximum binding. This difference was taken to represent 100% binding and rel- ative inhibition by the competing sera was calculated on this scale. The level for defining titers of the antibodies was set to 75% of maximal binding as not even high concentrations of nonspecific antibodies caused more than 20% inhibition.

Results

Polypeptide and antigenic site specificity of MAbs to H and F proteins in competi- tion EIAs

The polypeptide and antigenic site specificity of the MAbs was characterized us- ing competition EIA as shown in Fig. 1. When MAb 144 to H was used as a com- peting unlabelled antibody, it blocked the binding of the labelled MAb 144 but did not block the binding of the F protein-specific MAbs 19GD6, 19HB4, or 9DBlO. A similar experiment was also done using 3 MAbs against the F protein as unla- belled competing antibodies. The binding of MAb 144 to H was not inhibited by the binding of these antibodies. These results show that antibodies to MV H and F proteins could be measured separately under these experimental conditions.

When 3 MAbs against the F protein were used in the competition binding assay,

P 100

E $ 80

g 60

x 2 40

B 20

3

0

70

2 6 32 126 512 2046

I /SERUM DILUTION ( x IO2 )

. 8

0 . 56

0

. . i

NBD SSPE

Fig. 3. A typical inhibitory assay for measurement of antibodies in human scra. Different human sera

were incubated on antigen plates and the binding inhibition of labelled MAb I44 was measured. One

measles-negative serum (0). one serum taken from the acute (0) and convalescent (W) phases of a

measles patient, one serum from a NBD with past measles history (A). and one SSPE serum (0) were

used in this test.

Fig. 4. Ratio of anti-F antibody titers (MAb 19HB4IMAb 9DBlO) in human sera. Serum specimens

from 10 NBDs and 10 SSPE patients were tested.

some overlapping in binding of antibodies to different epitopes was found. How- ever, 2 major antigenic sites could be distinguished on the F protein. One site was defined by the MAb 19GD6 or 19HB4, and another site was recognized by MAb 9DBlO when 19GD6 was used as the unlabelled antibody (Fig. 2). Although low- level cross-inhibition overlapping was seen between these sites, there was almost a lOO-fold difference between the MAbs in antibody concentrations causing 25% inhibition. This was also confirmed using MAb 19HB4 as a competing antibody, which did not block considerably the binding of MAb 9DBlO (data not shown).

TABLE 1

Mean titers and standard deviations (log,) of antibodies to different antigenic sites on H and F pro-

teins. Ten patients each from NBDs, patients with acute measles, or convalescing from it. and SSPE

patients were tested.

Antigenic site de- NBD

fined by MAb Acute Convalescent SSPE

I44 8.9k1.4 c7.10 9.9kO.7 13.4% I .6 19GD6 8.0tl.0 8.751.4 12.1~0.6 13.5* 1.4 9DBlO 8.220.8 7.8kO.5 10.1-+1.2 ll.h+l.l

19HB4 8.3kO.7 9.021.4 13.0?0.6 13.22 I .4

71

The results show that antibody responses against these 2 distinct antigenic sites on the F polypeptide can be measured with this technique.

Antibodies specific to polypeptides and antigenic sites in human sera

Antibody titers against the H polypeptide and against 2 different antigenic sites on the F protein were measured using competitive inhibition assays. Results of a typical assay are presented in Fig. 3. The antibody amounts were read as titers at the level corresponding to 25% inhibition of the maximal binding. Antibodies to the antigenic sites recognized by the 4 mouse MAbs were present in each serum

group. All the antibody titers measured increased significantly from acute infection to

early convalescent sera from measles patients. The lowest mean titer in acute sera was against the site on the H protein; only one of the patients had a higher log, titer than 7.6 (threshold level). The highest mean titer in both acute and early con- valescent sera was against the antigenic site of the F protein defined by MAb 19HB4; also, the titer increase was largest for these antibodies (see Table 1).

All antibody titers to F protein were lower in NBDs than in early convalescent sera (Mann-Whitney U test, P values ranged from 0.05 to 0.01). Although the ti- ter against the epitope on the H protein was low in relation to titers against the F protein in early convalescent sera, this was the highest mean titer in the NBD group. This shows that the antibody response to this particular site is developed late during the infection. In SSPE patients, all titers were considerably higher which is in accord with the known hyperimmune state of these patients (Salmi et al., 1972). Only one of the mean titers (see Table l), the one defined by MAb 9DB10, was clearly lower than the others (P values ranged from 0.05 to 0.02).

12. .

IO- .

F” 8- d - .

6- El - b ” .

.

2- 0 i; ’

8AL8 C SJL DAYS AFTER IMMUNIZATION

Fig. 5. Kinetics of antibody synthesis against the H epitope MAb 144 (A) and the F protein MAb 19HB4 (B) in two inbred mouse strains, Balb c/A (0) and SJL (0).

Fig. 6. Ratio of anti-F antibody titers (MAb 19HB4/MAb 9DBlO) in Balb c/A and SJL mice at 28 d after immunization.

72

The relative amounts of antibodies against the antigenic sites was assessed by calculating antibody titer ratios. A significant difference between the NBDs and SSPE patients (PCO.02, Mann-Whitney U test) was seen in MAb 19HB4/9DBlO

antibody titer (anti-F) ratios which were higher in SSPE patients (see Fig. 4). All anti-F/anti-H ratios were found to be close to each other in both SSPE and NBD groups whereas early convalescent sera had significantly (P~0.01) higher ratios.

Antibody response in two mice strains after immunization

All antibodies defined by the four MAbs were present in sera of both Balb c/A and SJL mice after immunization with purified measles virions. All antibody titers increased until Day 28 in both strains (Fig. 5). The antibody titers to F protein were significantly higher on Day 28 p.i. in Balb c/A than in SJL mice which reflects the fact that also total antibody titers were significantly higher in Balb c/A mice (data not shown). However, the anti-H antibody titers did not differ significantly between the 2 strains. Anti-F/anti-H antibody titer ratios were thus higher in Balb c/A mice than in SJL mice. A difference between these mouse strains was also found in the antibody titer ratios of the F protein epitopes 19HB419DBlO. SJL mice had on Days 7 to 28 higher ratios (see Fig. 6) than Balb c/A strain.

Discussion

In this study, we have used competition binding EIA to measure polypeptide- and antigenic site-specific antibodies against F and H proteins of MV. The assay proved to be useful in detection of these antibodies in serum specimens. We were able to detect specific antibodies against at least 2 different antigenic sites on the F protein and one on the H protein.

The specificity of the assay to measles H and F proteins was demonstrated by allowing the previously characterized MAbs against these polypeptides to compete with each other. Practically no cross-reactions were found between the antibodies to H and F proteins. The assay was also found to be specific for measuring anti- bodies against 2 antigenic sites on the F and one epitope on the H proteins as only slight overlapping was detected.

The developed assay offers obvious advantages over the previously used meth- ods for measuring antibodies to single viral proteins and antigenic sites. First, the method does not require purification of the polypeptides which is time-consuming and often leads to at least partial denaturation and disappearance of some epi- topes. In methods such as Western blotting, the antigens are denaturated which definitely changes the antigenicity of the polypeptides. In our method, either in- fected cell lysate or purified virions can be used as antigen on the solid phase both of which should be closer to the native form of proteins. Secondly, it is possible to quantify the specific antibodies and express the amount in titers. By using the standard curves with known amounts of antibodies, it is possible to quantify the immunoglobulins in sera as pg/ml. Recently, a similar assay was used for meas-

73

uring antibodies in human sera against antige~ic sites on herpes simplex glycopro- teins {Ross et al., 1985). However, in that assay ratios of inhibition were calcu- lated with one serum dilution and thus the method is poorly quantitative,

To compare the relative amount of different antibodies in different serum groups, we calculated antibody titer ratios. In anti-F/anti-H antibody titer ratios, the con- valescent sera had significantly higher values than NBDs or SSPE patients which shows that the levels of antibodies to antigenic sites on different polypeptides do not increase at corresponding rates after infection. We found a difference between NBDs and SSPE patients in one of the ratios defined by 2 MAbs to F protein. There was also individual variation of this ratio within the groups. In immunized mice, a difference between the strains in the same ratio of titers against F protein was also noticed. Another finding was that, although Balb c/A mice had signifi- cantly higher titers to all sites on the F protein, the titers against the antigenic site on the H in SJL mice were at about the same levels. These results suggest that antibody response to different proteins and even to single antigenic sites on the same polypeptides are independent phenomena.

In measles as in many other viral diseases, antibodies against certain polypep- tides may have an essential role in the outcome of the disease (Rammohan et al., 1982). As well, only some epitopes on viral proteins might be important in pro- tective immunity and in determining the pathogenicity of the virus, as has been shown with some animal models (Dietzschold et al., 1983; Talbot et al., 1985; Love et al., 198s). Until now, the only method for measuring neutralizing antibodies in polyclonal sera has been the laborious neutralization test in tissue culture. How- ever, it does not give any information on the level of antibodies against single an- tigenic sites which might be important when new vaccines are being developed, If neutralizing MAbs are available, it is easy by our assay to estimate and to some extent quantify the antibodies with similar neutralizing capacity in polyclonal sera. Such tests and other applications of the method described here may turn out to be important in studies of virus immunopathogenesis.

This study was supported by Alberta Heritage Foundation for Medical Re- search.

References

Dietzschold, B., Wunner, W.H,. Wiktor, T.J., Lopes, AD., Lafon, M., Smith, C.L. and Koprowski, H. (1983) Characterization of an antigenic determmant of the glycoprotein that correlates with pathogenicity of rabies virus, Proc. Natl. Acad, Sci. U.S.A. 80, 70-74.

Fujinami, R.S., Norrby, E. and Oldstone, M.B.A. (1984) Antigenic modulation induced by mono- clonal antibodies: antibodies to measles virus hemagglutinin alters expression of other viral poly- peptides in infected cells. J. Immunol. 132, 2618-2621.

74

Graves, M., Griffin, D.E., Johnson, R.T., Hirsch. R.L., De Soriano, I.L.. Roedenbeck, S. and Vais-

berg, A. (1984) Development of antibody to measles virus polypeptides during complicated and un-

complicated measles virus infections. .I. Viral. 49, 409-412. Love, A., Rydbeck, R., Kristensson, K., Grvell, C. and Norrby, E. (1985) Hemagglutinin-neuramin-

idase glycoprotein as a determinant of pathogenicity in mumps virus hamster encephalitis: analysis

of mutants selected with monoclonal antibodies. J. Viral. 53, 67-74.

Norrby, E. and Gollmar, Y. (1975) Identification of measles virus-specific hemolysis-inhibiting anti-

bodies. Infect. Immun. 11, 231-239.

Oldstone, M.B.A. (1983) Immunopathology of persistent viral infection. In: The Biology of Immu-

nologic Disease (Dixon, F.J. and Fisher, D.W.. eds.), pp. 283-294. Sinauer Associates Inc., Sun-

derland, MA.

Rammohan, K.W., McFarland, H.F. and McFarlin, D.E. (1981) Induction of subacute murine measles

encephalitis by monoclonal antibody to virus hemagglutinin. Nature (London) 290, 588589.

Rammohan, K.W., McFarland, H.F. and McFarlin, D.E. (1982) Subacute sclerosing panencephalitis after passive immunization and natural measles infection: role of antibody in persistence of measles

virus. Neurology 32, 390-394.

Ross, C., Glorioso, .I., Sacks, S., Lavery, C. and Rawls, W. (1985) Competitive inhibition by human

sera of mouse monoclonal antibody binding to glycoproteins C and D of herpes simplex virus types

1 and 2. J. Viral. 54, 851-855.

Salmi, A.A., Norrby, E. and Panelius, M. (1972) Identification of different measles virus-specific an- tibodies in the serum and cerebrospinal fluid from patients with subacute sclerosing panencephalitis

and multiple sclerosis. Infect. Immun. 6, 248-254.

Sheshberadaran, H., Chen, S.N. and Norrby, E. (1983) Monoclonal antibodies against five structural

components of measles virus. I. Characterization of antigenic determinants on nine strains of mea-

sles virus. Virology 128, 341-353. Sheshberadaran, H. and Norrby, E. (1986) Characterization of epitopes on the measles virus hemag-

glutinin. Virology 152, 58-65.

Talbot, P.J., Salmi, A.A., Knobler, R.L. and Buchmeier, M.J. (1984) Topographical mapping of ep-

itopes on the glycoproteins of murine hepatitis virus-4 (strain JHM): correlation with biological ac-

tivities. Virology 132, 25(t260.

Talbot, P.J., Salmi, A.A., Knobler, R.L. and Buchmeier, M.J. (1985) Epitope-specilic antibody re-

sponse to murine hepatitis virus-4 (strain JHM). J. Immunol. 134. 1217-1224.

Trudgett, A., Bellini, W.J., Mingioli, E.S. and McFarlin, D.E. (1980) Antibodies to the structural polypeptides of measles virus following acute infection and in SSPE. Clin. Exp. Immunol. 39,

652-656.

Tuokko, H. and Salmi, A. (1983) Detection of IgM antibodies to measles virus by enzyme-immu- noassay. Med. Microbial. Immunol. 171, 187-198.

Vainionpaa, R., Ziola, B. and Salmi, A. (1978) Measles virus polypeptides in purified virions and in

infected cells. Acta Pathol. Microbial. Stand. Sect. B 86. 379-385.