Solid-Phase Radioimmunoassay of Human Immunoglobulin

Transcription

1 INFECTION AND IMMUNITY, Mar. 1977, p Copyright X 1977 American Society for Microbiology Vol. 15, No. 3 Printed in U.S.A. Solid-Phase Radioimmunoassay of Human Immunoglobulin M and Immunoglobulin G Antibodies Against Herpes Simplex Virus Type 1 Capsid, Envelope, and Excreted Antigens KIRSTI 0. K. KALIMO,* REIJO J. MARTTILA, KAISA GRANFORS, AND MATTI K. VILJANEN Departments of Virology, Dermatology and Medical Microbiology, University of Turku, SF Turku 52, Finland Received for publication 26 August 1976 A solid-phase radioimmunoassay developed in our laboratory for detection of human viral immunoglobulin M (IgM) and IgG antibodies was applied to demonstrate human class-specific antibody response against capsid, envelope, and excreted antigens of herpes simplex virus type 1. In primary infections, a clear IgM and IgG antibody response was found predominantly against the envelope components, whereas the IgM and IgG antibodies to the capsid antigen appeared more slowly. Increasing IgG antibody titers to the excreted antigen were also found in primary infections, though appearing more slowly than antibodies to the other subunit antigens. The antibody response against capsid and envelope antigens was not type specific, whereas in primary infections IgG class antibodies against the excreted antigen showed distinct type specificity. In recurrent infections, no significant level of IgM class antibodies was demonstrated, but in the patients with a severe secondary herpes simplex virus infection a definite IgM class antibody response was found against the envelope antigen. In addition, during severe secondary infections the antibody response against the excreted antigen was enhanced. The host IgG antibody response in recurrent infections was directed against the envelope and excreted antigens, whereas the level of the capsid antibodies was relatively stable. Antibody response to subunit antigens of herpes simplex virus (HSV) types 1 and 2, i.e., the envelope, capsid, and soluble antigens, has been studied by the complement fixation technique (6) and by an indirect hemagglutination test (la). The findings hitherto can be summarized as follows: (i) in primary herpetic infections, antibody response against the envelope antigen is predominant compared to that against the capsid antigen; (ii) in patients with recurrent herpetic infections, antibodies against the capsid antigen seem to increase; (iii) recurrent HSV infections seem to broaden heterotypic reactivity between type 1 and type 2 immune sera with the capsid and envelope antigens; (iv) antibodies against the soluble antigen can be readily detected in animal immune sera, but only rarely in human serum specimens; and (v) none of these antigens shows significant type specificity, though in antibody response against proteins excreted from HSV-infected cells, on the other hand, considerable type specificity was observed in animal sera (5). Recently, a sensitive radioimmunoassay (RIA) technique was developed in our laboratory to detect antibodies to rubella virus (4). The same methodology was further applied to the detection of HSV antibodies (Kalimo et al., J. Immunol. Methods, in press) and measles virus antibodies (1) in human serum specimens. The advantages of the technique developed are high sensitivity and the possibility to demonstrate both immunoglobulin G (IgG) and IgM class antibodies. In the present study class-specific antibody responses detected by RIA in human sera against the crude and subunit antigens of HSV type 1 are described, and a possible diagnostic value of the antibody changes during primary and secondary HSV infections is discussed. MATERIALS AND METHODS Virus. For preparing all antigens the VR strains of HSV type 1 grown in Vero cells in 950-ml Roux bottles was used. Culture media. To prepare the crude, capsid, or envelope antigens, Eagle basal medium was supplemented with 0.2% bovine albumin, fraction V. Eagle basal medium without any supplements was used to produce the excreted antigen. 883

2 884 KALIMO ET AL. Crude antigen. Harvesting and purification of the crude antigen was done as previously described (Kalimo et al., in press). In brief, infected cells were harvested when a cytopathic effect (CPE) was evident in 75 to 100% of the cells and separated by lowspeed centrifugation. The cells were made 2% (vol/ vol) in Dulbecco phosphate-buffered saline and disrupted by freezing and thawing, followed by sonication. After low-speed centrifugation, the supernatant was collected and further centrifuged at 100,000 x g for 2 h. The pellet was run through a 10 to 60% (wt/vol) linear sucrose gradient at 100,000 x g for 2 h. The resulting band contained the viral material; the band was collected and sucrose was removed by an additional centrifugation. Capsid antigen. The preparation of the capsid antigen was carried out as described by Martin et al. (6). HSV-infected cells were harvested at 75% CPE into 2.5% Nonidet P-40 solution and sonicated. The suspension was filtered through and 0.22-,um membrane filters (Millex, Millipore S.A , Molsheim-France). The filtrate was layered onto a 30% (wt/vol) sucrose cushion and centrifuged at 100,000 x g for 1 h. The pellet contained the capsid antigen. Envelope antigen. The method of Martin et al. (6) was used. Infected cells were scraped into the maintenance medium at 75 to 100% CPE. After disruption by sonication, the suspension was centrifuged at 100,000 x g for 1 h. The pellet was treated with 50% diethyl ether for 30 min and then dialyzed against 0.85% NaCl buffered at ph 10.4 with 0.05 M glycine NaOH. The dialysate was layered onto a 35% (wt/vol) sucrose cushion and centrifuged for 1 h at 100,000 x g. The supernatant fluid containing the envelope antigen was collected and dialyzed against phosphate-buffered saline. Excreted antigen. The preparation of this antigen was carried out according to Kaplan et al. (5). The culture medium was decanted when the infected cells showed 75% CPE, usually 50 to 55 h postinfection. After low-speed centrifugation to remove the cellular debris, the supernatant was further centrifuged at 100,000 x g for 1 h to sediment the viral particles. The supernatant fluid was extensively dialyzed and considered as the excreted antigen. Electron microscopy. Specimens from the crude, capsid, and envelope antigens were prepared on carbon-coated Formvar grids and negatively stained with 2% phosphotungstic acid, ph 6.5, and examined with a JEM 100C electron microscope (Fig. lac). RIA. The RIA used to detect class-specific HSV antibodies has been described previously in detail (Kalimo et al., in press). Briefly, fourfold serial serum dilutions were incubated in plastic tubes with the antigen-coated polystyrene balls. After incubation for 1 h at 37 C, the serum was aspirated off and the balls were washed twice with 5 ml of tap water. An aliquot of '25I-labeled anti-human IgG or IgM containing 30,000 cpm was added to each tube. The anti-human immunoglobulins were prepared in sheep. For immunization, pure IgG and IgM fractions were used. They were tested to be monospecific by immunodiffusion and immunoelectrophoresis INFECT. IMMUN. (12). After incubation for 1 h at 370C for IgG determination and for 16 h at room temperature for IgM determination, the radioactive solution was aspirated off and the balls were washed as above. The balls were removed into clean plastic tubes and the bound radioactivity was counted in a LKB Wallac 1280 gamma counter. A positive and negative control serum and a buffer blank, where serum was omitted, were included in each test. End-point titer values were obtained from the cpm (log,,,) versus dilution (log2) curve of each specimen. The end-point titer was taken to mean that serum dilution where the cpm values obtained were three times those of the negative serum in the IgG assay and two times the negative in the IgM assay. Titers were expressed as log2 values of reciprocals of the serum dilutions. Serum specimens. The serum specimens consisted of 64 specimens from 39 patients obtained from the Department of Dermatology and Department of Infectious Diseases at Turku University Hospital and from the routine material in our diagnostic laboratory. The specimens comprised the following. (i) Twenty-three specimens were from 10 patients with a primary HSV infection: 6 patients with clinically established diagnosis of HSV type 1 infection, and 4 patients with type 2 infection. (ii) Twenty specimens were from 15 patients with a history of several recurrent herpetic infections during the past year: 11 patients with a history of orofacial lesions, and 4 patients with genital lesions. (iii) Fifteen specimens were from 10 patients with a past history of herpetic infections, with the proviso that the last infection have been more than 1 year previously. In this group no patient gave a definite history of genital lesions. (iv) Six specimens were from 4 patients with a previous history of HSV infections and current severe secondary HSV infection: 1 patient with a serious recurrent stomatitis, 1 with meningitis, and 2 with a generalized eczema herpeticum. The pooled, positive control serum for the crude, capsid, and envelope antigen assays consisted of four convalescent phase specimens, and for the assay of the excreted antigen a large serum pool of one patient with several recurrent HSV infections was used. The negative control serum was pooled from one donor who did not have a history of HSV infections and did not show antibodies to any of the antigens tested in the HSV RIA. The specimens were stored at -200C until used. RESULTS RIA reactivity of different antigens. To establish the reactivity of the subunit antigens in the RIA, tests with positive and negative control sera were performed. The ratio of the radioactivity bound by the positive and by the negative serum, i.e., the binding ratio, in the IgG assay with all antigens was high, sometimes even as high as 30. There was, however, a distinct difference between different antigens regarding the binding ratios. When the balls

3 Pr '.. Al.--,' -C.it Downloaded from on July 24, 2018 by guest FIG. 1. Electron micrographs ofhsv antigens used in the assays stained with phosphotungstic acid. Bar = 100 nm. (a) Crude antigen, (b) capsid antigen, (c) envelope antigen.

4 886 KALIMO ET AL. coated with the capsid antigen were used, the cpm values of the negative sample in the dilutions from 1:8 to 1:4,096 were three to four times higher than the cpm values obtained with the envelope-antigen-coated balls. At higher dilutions the difference reduced to twofold. With the excreted antigen the cpm values for the negative serum were the same as or two times higher than those obtained with the envelope antigen. In the IgM assay the binding ratios between positive and negative samples were INFECT. IMMUN. lower than in the IgG assay, reaching ratios from 4 to 8. As in the IgG assay, the capsid antigen gave a higher nonspecific background than the envelope antigen. IgM class antibodies against the excreted antigen were detected neither in the positive control serum nor in any of the sera tested. Antibody response in patients with a primary HSV infection. The development of IgM and IgG class antibodies in the serum specimens from the patients with a primary HSV infection against the crude and subunit antigens of HSV type 1 is presented in Table 1. In the specimens obtained within a week after the onset of symptoms, IgM RIA antibodies against the crude and envelope antigens were detected in four patients (no. 3, 4, 5, 7), whereas IgM RIA antibodies against the capsid antigen were detected in none of the acute stage specimens. In the convalescent-phase specimens, IgM RIA antibodies against the crude, capsid, and envelope antigens were demonstrated in each ofthe patients. In three specimens (patients no. 2, 5, 10), obtained more than 4 weeks after the onset of symptoms, IgM RIA antibodies against the capsid antigen had disappeared, and in all of these specimens a tendency for the IgM RIA envelope antibodies to decline was also demonstrated. The IgM class antibody response to the differ- TABLE 1. HSV type 1 IgM RIA and IgG RIA antibody titers against crude, capsid, envelope, and excreted antigens in a series of serum specimens taken from 10 patients with a primary herpes simplex infection Titers (log2 values) Clinical diagnosis of patient (AYge) gm IgG onset Crude Capsid Envelope Crude Capsid Envelope Excreted antigen antigen antigen antigen antigen antigen antigen 1. Gingivostomatitis 1 2 <4 <3 <3 <4 <3 <3 < Stomatitis 2 4 <4 <3 <3 <4 <3 <3 < < Gingivostomatitis <3 7.0 <4 <3 <3 < Gingivostomatitis < < Stomatitis < < < < Keratitis 15 3 <4 <3 <3 <4 <3 <3 < <1 7. Herpes genitalis < <3 6.8 < <1 8. Herpes genitalis 21 5 <4 <3 <3 8 < < <1 9. Herpes genitalis < < <1 10. Herpes genitalis 27 7 <4 <3 <3 <4 <3 <3 < < < <1

5 VOL. 15, 1977 ent HSV 1 antigens appeared to be similar in all patients, although six patients had a clinically established type 1 infection and four patients had type 2 infection. IgG RIA antibodies against the crude and envelope antigens were detected in four patients (no. 4, 5, 7, 8) in the specimens obtained within a week after the onset of symptoms. None of those specimens showed IgG RIA antibodies against the capsid antigen. In only one case (no. 4) were IgG RIA antibodies against the excreted antigen detected in the acutephase specimens. In the convalescent-phase specimens, IgG RIA antibodies to the crude, envelope, and capsid antigens could be demonstrated in each patient. In antibody response to the crude, capsid, or envelope antigens used, no difference was found between the patients with type 1 or type 2 infection. On the other hand, IgG RIA antibodies against the excreted antigen could be observed in rather low titers in the convalescent sera from five patients (no. 1, 2, 3, 4, and 5), all of them with a clinically established type 1 infection. None ofthe four patients with a clinical diagnosis of genital herpes infection had antibodies against the excreted antigen. Also, one patient with a clinical diagnosis of keratitis did not show antibodies to the excreted antigen. Antibodies in patients with a previous history of herpetic infection. IgG RIA antibodies to each HSV subunit antigens were detected in every patient with a past history of herpetic infections or with a history of several recurrent infections (Table 2). The mean antibody titer to the capsid antigen was rather similar as between the patients with a history of several recurrences and the RIA ANTIBODIES TO HSV SUBUNIT ANTIGENS 887 patients with a past exposure. However, the mean envelope antibody titer was slightly higher in the former group than in the patients with past infections, though the difference was not statistically significant. A statistically significantly (P < 0.01) higher IgG RIA antibody titer against the excreted antigen was found in the patients with several recurrences compared with the patients with past HSV infections. When the mean IgG RIA antibody titers against the excreted antigen were studied in the patients with several recurrent infections, a statistically significantly (P < 0.05) higher mean titer was observed in the patients with orofacial lesions compared with the patients with genital herpes infections. The material, however, is rather limited, and no definite conclusions are possible. No other differences in the IgG RIA antibody levels to the other HSV antigens tested could be observed between these clinically established type 1 and type 2 recurrent HSV infections. No significant IgM RIA antibodies were demonstrated to any of the antigens in the patients with past HSV infections or with several recurrent infections. However, IgM antibody activity less than two times the cpm values of the control negative serum was found in some of the specimens taken during recurrences. When comparing the mean IgG RIA capsid antibody titers in the patients with recurrent infections or with a past history of herpetic infections with the patients convalescing from an acute HSV infection, no differences were demonstrated (Table 2). However, in the antibodies against the envelope antigen, a slightly lower mean titer was found in the patients with recurrent HSV type 1 infections than in the TABLE 2. Comparison between mean IgG RIA antibody titers against HSV capsid, envelope, and excreted antigens in 10 patients with a primary herpetic infection, in 15 patients with a history ofrecurrent infections, and in 10 patients with past herpetic infections Clinical diagnosis Titer (log2 value) No. of Capsid antigen Envelope antigen Excreted antigen specimens %PainsPa- % Pa- Mean + SDa % Patients Mean + SD tients pos- Mean + SD tients positive itive Primary infection 10 NDb None c 10 Acute phase Primary infection ± ± Convalescent phase Recurrent infections Orofacial lesions Recurrent infections ± ± ± Genital lesions Past infections ± ± asd, Standard deviation. bnd, Not detectable. Only one specimen positive.

6 888 KALIMO ET AL. patients in the convalescent phase; but, on the other hand, the patients with several recurrences showed a somewhat higher mean titer than the patients with past infections. A statistically significantly (P < 0.05) lower mean antibody titer against the envelope antigen was observed in the patients with past infections when compared with the patients convalescing from acute infection. Fifty percent of the convalescent sera showed antibodies with a low mean titer to the excreted antigen, whereas all sera from the patients with several recurrences or with past infections contained antibodies to that antigen in a significantly (P < 0.001) higher mean titer, thus demonstrating the slow appearance and increase of these antibodies. In four patients with a severe secondary HSV infection (Table 3), IgM class antibodies to the envelope antigen were demonstrated, but none of those patients showed IgM antibodies against the capsid antigen. In two patients from whom paired samples were available, a fourfold IgG antibody rise against the excreted antigen was observed, whereas one of the patients also showed a similar antibody rise against the envelope antigen and almost fourfold antibody rise against the capsid antigen. DISCUSSION In the present study, human antibody response to subunit antigens of HSV type 1 was studied by solid-phase RIA during primary and secondary herpetic infections. This highly sensitive method, which allows both IgM and IgG class-specific antibody detection, was applied to demonstrate antibodies to the HSV crude, capsid, envelope, and excreted antigens. In the primary infections, an IgM class antibody response was clearly detected in all patients. In the acute phase of the infection, the IgM response was detected against the crude and envelope antigens, but not against the capsid antigen. However, during the convalescent INFECT. IMMUN. phase, IgM antibodies were also demonstrated against the capsid antigen in all patients. In three patients from whom the specimens were obtained more than 4 weeks after onset, the IgM RIA antibodies against the capsid antigen had disappeared; simultaneously the IgM envelope antibody titers also decreased. An IgG class antibody response against the envelope and crude antigens was mainly demonstrated concurrently with the appearance of the IgM class antibodies, whereas the IgG capsid antibodies were detectable somewhat later. Thus, it is evident that during primary HSV infection the host IgM and IgG antibody response is primarily directed towards the envelope antigen of HSV and that antibodies against the internal components appear more slowly. Similar findings have been also previously obtained by less sensitive techniques, namely, the complement fixation test (6) and the indirect hemagglutination test (la). The predominant host response against the envelope components of HSV is conceivably connected with the early appearance of virusinduced antigens in the external membranes of HSV-infected cells. In tissue culture conditions after a few hours of infection, the cell surface acquires virus-specific antigens (3, 8, 10), which both antigenically (8) and biochemically (9) correspond to the proteins in the viral envelope. Thus, already in an early phase of infection the antigenically altered cell surfaces are available to the host immune mechanisms. In addition, the antibodies against the envelope components seem to be of primary importance to the host since they contain the neutralizing activity (7, 11) Ṫhe results of this study also indicate that HSV type 1 and type 2 share common antigens both in the envelope and in the capsid; thus a direct typing of antibody response using these subunit antigens does not seem to be possible. However, a distinct type specificity in the antibody response was observed against the excreted antigen, since all patients with a genital TABLE 3. HSV type 1 IgM RIA and IgG RIA antibody titers against capsid, envelope, and excreted antigens in fbur patients with a severe secondary herpetic infection Clinical diagnosis Age (yr) Titer (log2 value) Days 1gM IgG after on- _ set Capsid an- Envelope Capsid anti- Envelope Excreted tigen antigen gen antigen antigen Eczema herpeticum 20 4 <3 < < Eczema herpeticum 25 5 < < Stomatitis 46 8 < Meningitis <

7 VOL. 15, 1977 primary HSV infection were negative with regard to those antibodies, whereas most of the patients with a clinical diagnosis of HSV type 1 infection showed increasing antibody titers to the excreted antigen. On the other hand, four patients with recurrent genital lesions displayed antibodies to the excreted antigen of HSV type 1, but the possibility of the previous type 1 infection could not be excluded. Thus, the excreted antigen seems to have promising type specificity also in human HSV infections, as has previously been observed in animal models (5). Further studies, however, are required to establish the final usefulness of this antigen in serological typing of HSV infections. In earlier studies the antibody levels to the capsid antigen have shown a tendency to increase during recurrent HSV infections (la, 6). Our present results indicate that these antibodies remain relatively stable on the level reached in primary infection even in the patients with several recurrences. On the other hand, the level of antibodies to the envelope antigen seems to increase in the patients with several recurrent HSV infections, suggesting that the antigenically altered cell surfaces are the main antigenic stimulus also during recurrences. However, an increased antibody level to the excreted antigen was also found in these patients and an enhanced antibody response was found against the excreted antigen in the patients with a severe secondary HSV infection, a finding which could be used as a criterion in differentiating primary and secondary infections. The difference between the slow primary response to the excreted antigen and the enhanced reactivity during the secondary infections compared with the antibody response against the capsid and envelope antigens remains obscure at the moment. However, on the basis of our findings it seems reasonable to suppose that the excreted antigen differs antigenically from the capsid and envelope components of HSV. Consequently, the excreted proteins might be nonstructural ones as suggested (5) and as is the case with the proteins excreted from pseudorabies virus-infected cells (2, 5). Although IgM class antibodies could not be detected against any of the subunit or crude HSV antigens in ordinary recurrent infections, in the patients with a severe secondary infection, an IgM class antibody response against the envelope antigen was clearly demonstrated in all patients. In those patients a pronounced IgG class antibody response against the excreted antigen was also observed and, in one patient, IgG class antibody response against the envelope and capsid antigens. Our findings indicate that an IgM class antibody response is RIA ANTIBODIES TO HSV SUBUNIT ANTIGENS 889 connected with severe, secondary HSV infections and suggest that also during recurrent infections of a less severe nature there might be a host IgM antibody response to HSV, though to a minor degree and beyond the sensitivity of the RIA technique used. ACKNOWLEDGMENTS This study was supported by grants from The Academy of Finland, Medical Research Council. The excellent technical assistance of Marita Maaronen and Soile Niittoaho is gratefully acknowledged. We are grateful to P. Terho for kindly providing some of the serum specimens, and to H. Kalimo for examining the electron microscopic specimens. LITERATURE CITED 1. Arstila, P., T. Vuorimag, K. Kalimo, P. Halonen, M. Viljanen, K. Grenfors, and P. Tolvanen Solid-phase radioimmunoassay for IgG and IgM antibodies against measles virus. J. Gen. Virol. 34: la. Back, A. F., and N. J. Schmidt Reactivity of envelope, capsid, and soluble antigens of herpesvirus hominis types 1 and 2 in the indirect hemagglutination test. Infect. Immun. 10: Erickson, J. S., and A. S. Kaplan Synthesis of proteins in cells infected with herpesvirus. IX. Sulfated proteins. Virology 55: Ito, M., and A. L. Barron Surface antigen produced by herpes simplex virus (HSV). J. Immunol. 108: Kalimo, K. 0. K., 0. H. Meurman, P. E. Halonen, B. R. Ziola, M. K. Vilbanen, K. Granfors, and P. Toivanen Solid-phase radioimmunoassay of rubella virus immunoglobulin G and immunoglobulin M antibodies. J. Clin. Microbiol. 4: Kaplan, A. S., J. S. Erickson, and T. Ben-Porat synthesis of proteins in cells infected with herpesvirus. X. Proteins excreted by cells infected with herpes simplex virus types 1 and 2. Virology 64: Martin, M. L., E. L. Palmer, and R. E. Kissling Complement-fixing antigens of herpes simplex virus types 1 and 2: reactivity of capsid, envelope, and soluble antigens. Infect. Immun. 5: Roane, P. R., Jr., and B. Roizman Studies of the determinant antigens of viable cells. H. Demonstration of altered antigenic reactivity of HEp-2 cells infected with herpes simplex virus. Virology 22: Roizman, B Herpesviruses, membranes, and the social behavior of infected cells, p In M. Sanders (ed.), Viruses affecting man and animals. W. H. Green, Inc., St. Louis, Mo. 9. Roizman, B Biochemical features ofherpesvirusinfected cells, p In W. Nakahara, K. Nishioka, T. Hirayma, and Y. Ito (ed.), Recent advances in human tumor virology and immunology (Proc. 1st Int. Symp. Princess Takamatsu Cancer Res. Fund). University of Tokyo Press, Tokyo. 10. Schwartz, J., and T. S. Elizan Altered immunologic specificity of cells infected with herpes simplex virus: recognition by human antiserum. Z. Immunitaetsforsch. 148: Smith, J. W., S. P. Lowry, J. L. Melnick, and W. E. Rawls Antibodies to surface antigens ofherpesvirus type 1- and type 2-infected cells among women with cervical cancer and control women. Infect. Immun. 5: Viljanen, M. K., K. Granfors, and P. Toivanen, Radioimmunoassay of class-specific antibodies (RIACA): chicken antibodies to bovine serum albumin. Immunochemistry 12: