- Email: [email protected]
Antibodies against human CMV-induced “early” antigens detected by immunoblotting are not exclusively associated with active infection
~) INSTITUTPASTEUR/ELsEVIER Paris 1991
Res. Virol. 1991, i42, 33-37
Antibodies against human CMV-induced "early" antigens detected by immunoblotting are not exclusively associated with active infection A. Hamann, W. Braun and H.-W. Doerr Zentrum der Hygiene, Abteilung fiir Medizir, ische Virologie, Universitiitskliniken Frankfurt, Paul-Ehrlich-Strasse 40, 6000 Frankfurt a. M. 70 (Cermany) SUMMARY We investigated the correlation between the prevalence of antibodies against human cytomegalovirus (HCMV} "early" proteins and the state of infection. Two hundred and twenty-eight HCMV-IgG-positive (HCMV late-antigen ELISA) sera, drawn from subjects with active and non-active HCMV infections, were tested for specific IgG antibodies against HCMV "early" antigens by a Western blotting micromethod. Applying virus strain AD-169, we found antibodies to HCMV "early" antigens in 59 % of all cases, irrespective of origin. Subjects with an elevated risk of cytomegelic inclusion disease (renal transplantation patients, AIDS patients) revealed significantly broader immune reactions when compared with healthy HCMV-antibody carriers. However, in follow-up studies we detected one patient who lacked an immune r~=pc:Ise to HCMV "early" antigens during active infection and subsequent convalescence. Our results indicate that active HCMV infection is not necessarily associated with the formation of antibodies against "'early'" :antlnAn=~ J~nrmnup_r_th~s;A antibodies may nersist durina non-active infection. . . . . .
O
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
•
=
v
Key-words: CMV, Pathogenicity, Diagnosis, Early protein; Immunoblotting, Active infection,
INTRODUCTION
Controversial information about the diagnostic evaluation of antibodies against the "early ~' antigens of human cytomegalovirus (HCMV) has often been published. The formation of antibodies against the "early" antigens of HCMV (EA-Ab) is considered to be a marker of an active or recently acquired HCMV infection (The et al., 1977; Musiani et al., 1978; Gerna et al., 1978; Griffiths et al., 1980; Musiani et aL, 1984; Numazaki et aL, 1980). Nevertheless, some
Submitted August 15, 1990, accepted November 19, 1990.
authors (Gergely et al., 1981 ; Friedman et al., 1982a,b; Middeldorp et al., 1984) assume that the presence of EA-Ab cannot be taken as presumptive evidence of active infection. In previous investigations of the prevalence of EA-Ab, immunofluorescenceassays (The et ai., 1977; Musiani et al., 1978; Griffiths et al., 1980; Friedman et al., 1980b), an immunoperoxidase technique (Gerna et aL, 1978) and ELISA (Middeldorp et aL, 1984) have been used for quantitative antibody detection. We have inves-
A. HAMANN
34
tigated the correlation between the prevalence o f EA-Ab and the state o f infection. E A - A b were detected by immunoblotting, a method that has not yet been used to this end.
M A T E R I A L S A N D METHODS
Serum samples We examined 96 HCMV-lgG-positive sera (as pretested by an HCMV late-antigen-specific ELISA) from immunosuppressed (HCMV risk groups) and immunocompetent persons: (1) 29 renal transplantation patients receiving immunosuppressive therapy, (2)33 AIDS patients, and (3)34 randomly selected, immunocompetent, healthy individuals. In addition, we tested 132 sera from 11 individuals, aged between 22-53 years, suffering from active cytomegalic inclusion disease (CID), to examine the course of infection and subsequent convalescence (5 cases of primary and 6 cases of reactivated infection as confirmed by: seroconversion, a 4-fold rise in IgG antibodies, a positive IgM titre to HCMV late antigen or isolation of virus from urine specimens).
ET AL.
FRG) by the Modi-blot method (Braun and Abraham, 1989). Blots (containing the infected and mockinfected cell preparations) were blocked against nonspecific binding with a "blocking" solution (BS=0.02 M Tris, 0 . 2 M NaC1, 0.2 070 NP-40, 0 . 0 1 % thimerosal, 2 070BSA, 5 070goat serum) and incubated with serum samples (diluted 1/40 in BS) overnight. Antigen-antibody reaction was visualized with a biotin/streptavidin amplification method using monoclonal mouse anti-human-IgG (Dianova, FRG; 1/1,000 in BS), biotin-labelled rat anti-mouse-IgG (Dianova, FRG; 1/2,000 in BS), streptavidin/horseradish peroxidase (Gibco, FRG; 1/3,000 in BS) and 3-amino-ethyl-carbazole (AEC)/H202 as enzyme substrate. The criteria used to identify virus-specific "early" proteins was the difference in electrophoretic mobility with respect to host cell polypeptides. Detection of antibodies to HCMV-specific late antigens All sera used in the immunoblot experiments were screened with a commercially available HCMV lateantigen-specific ELISA kit (Enzygnost-Zytomegalie, Behring, FRG) for IgG and IgM antibodies, as previously described (Doerr et al., 1987).
Detection of IgG-speeific antibodies to HCMV "early" antigens by immunoblotting Human foreskin fibroblasts were infected with HCMV AD-169 at an input multiplicity of 2 PFLl/ce!! and incubated in the presence of 100 t~g/ml phosphonoacetic acid to prevent virusspecific DNA synthesis. Cells were harvested at either 15, 25 or 35 h postinfection (p.i.). The resulting crude cell extracts were pooled and virus-specific "early" proteins were extracted from the nuclear fraction as described in detail elsewhere (Keil et al., 1985). The resulting cell lysates as well as crude cell extracts from mock-infected human embryonic fibroblasts were suspended in SDS sample buffer (1 mM Tris/Maleat, 1 mM EDTA, 2.5 070 SDS, 2.5 °70 DTE/bromophenol blue, pH 6.7) and heated to 100°C for 5 rain. Virus-specific "early" proteins and cell proteins were separated by"PhastSystems" eiectrophoresis system (Pharmacia, Sweden) and transferred to an "Immobilon" membrane (Millipore,
BS BSA CID EA-Ab ELISA
= = = = =
blocking solution. bovine serum albumin. cytomegalic inclusion disease. antibodies against HCMV "early" antigens. enzyme-linked immunosorbent assay.
RESULTS
The sera screened by immunoblotting reacted with a total o f 11 electrophoretically distinct HCMV-specific "early" proteins with a molecular weight ranging between 23 and 79 k D a (fig. 1). The prevalence o f E A - A b in immunosuppressed subjects at elevated risk o f CID (renal transplantation and A I D S patients) and in immunocompetent healthy subjects is shown in table I. Each group yielded approximately the same percentage o f positive immunoblots. Sera belonging to the group o f renal transplantation or A I D S patients recognized an average o f 6 + 1 polypeptides, and that o f i m m u n o c o m p e t e n t
HCMV PFU p.i. SDS
= = = =
human cytomegalovirus. plaque-forming unit. postinfection. sodium dodecyl sulphate.
CMV-INDUCED "'EARL Y'" A N T I G E N S A N D A CTIVE INFECTIO, q
A
B
C
D
E
,
F
,>
i~i!i~!~i~~! 4,.1
38 30
-%
2(;
iii
23
Fig. 1. Immune reactions against HCMV "early" proteins detected by immuno~otting. Sera drawn from 6 individuals. Immunoblots A-E are positive, F is negative; ~no!ecularweight~ (kDa) are marked on the side.
Table I. Prevalence of IgG antibodies to HCMV "early" antigens (EA-Ab) and a v e r s e numbc~ of recognized "early" proteins detected by immunob!otting in i~munosuppressed patients at elevated risk for CID and healthy control subjects.
Renal transplantation patients AIDS patients Immunocompetent, healthy subjects
No. pos./no, tested (070) for EA-Ab by immunoblotting
No. of recognized proteins (mean ± SD}
17/29 (59 %)
6 (_+ l)
19/33 (58 070)
6 (_+ 1)
20/34 (59 °70)
3 (+_ l)
35
A. H A M A N N E T A L .
36
healthy subjects, 3 _+1. All immunoblot-positive sera responded at least to the 66-kDa protein. Sera from the majority of the immunocompetent healthy subjects recognized only the 66, 59, 56, 50 or 26-kDa proteins. Sera from most of the immunosuppressed individuals responded, in addition, to the 79, 70, 43, 38, 30 or 23-kDa proteins. All sera tested by immunoblotting were screened with an HCMV late-antigen-specific ELISA for IgM antibodies. Among 30 ~70of the immunosuppressed persons belonging to the HCMV risk groups of renal transplantation and AIDS patients, IgM antibodies against HCMV late antigen were detected compared to 0 % among immunocompetent, healthy individuals. We examined 11 subjects suffering from active HCMV infection, during the course of infection and subsequent convalescence. One out of 11 patients, a 25-year-old immunocompetent woman suffering from primary typical CID (fever, hepatitis, atypical lymphocytes, positive IgM and IgA titres as determined by ! ~CMV lateantigen-specific ELISA), showed no positive immunoblots during the course of active infection or during subsequent convalescence. In all other patients, EA-Ab were detectable. Four out of 10 subjects continued to show positive immunoblots for up to 7 months after convalescence. During the active phase, the number of recognized polypeptides reached the highest value and decreased during convalescence. In 6 patients, EA-Ab were detected only during the active stage. In 5 individuals, HCMV could be isolated from urine, EA-Ab formation being associated with virus excretion. Formation of EA-Ab occurred both in primary and reactivated HCMV infections. DISCUSSION
We studied the immune response to "early" antigens in i m m u n o c o m p e t e n t and immunosupressed persons at elevated risk for CID (renal transplantation and AIDS patients; table I). If a correlation between the incidence of EA-Ab and active HCMV infection exists, it should be reflected by the results of this screen-
ing. Each group yielded approximately the same percentage of reactive immunoblots (59, 58 and 59 %). In 4 out of 10 patients suffering from an active HCMV-infection, EA-Ab could still be detected up to 7 months after convalescence. In agreement with other investigators (Gergely et al., 1981 ; Friedman et al., 1982a,b; Middeldorp et al., 1984), our data confirm that EA-Ab are present during non-active infection. The divergence from the results of those authors detecting EA-Ab exclusively during active infection might result from the higher sensitivity of the immunoblot used in our study. While investigating subjects suffering from an active HCMV infection during the course of infection and convalescence, 1 out of the 11 patients did not show a positive immunoblot during this time. We thus suppose that active HCMV infection does not lead to the formation of EA-Ab in all individuals, as determined by immunoblotting with strain-specific early antigen. The failure to detect EA-Ab in some individuals may be the result of antigenic heterogeneity between HCMV AD-169 (used in our experiments) and other HCMV strains. We have demonstrated that even quantitative EA-Ab detection does not prove active HCMV infection. In most cases, active HCMV infections could be differentiated from non-active infections by protein-specific immune reactions. Sera from the majority of individuals with non-active HCMV infection merely reacted with the 66, 59, 56, 50 or 26-kDa proteins, whereas sera from most of the subjects suffering from active HCMV infection recognized, in addition, the 79, 70, 43, 38, 30 or 23-kDa proteins. In common with Gergely et al., 1981, we also assume that the EA-Ab titre reaches the highest level at the active stage of infection and tends to decline thereafter. In some individuals, EA-Ab formation decreases to such an extent that EA-Ab cannot be detected after active infection. Finally, individual divergence of immune reactions to HCMV "early" antigens, in relation to pathogenesis and prognosis of HCMV infections, should be subject to further investigation.
C M V - I N D U C E D "'EARL Y'" A N T I G E N S A N D A C T I V E I N F E C T I O N
Acknowledgements This work was supported by grant 01ZR8606 from the Bundesminister for Forschung und Technologic, Germany.
Les anticorps anti-cytomdgalovirus humain induits par les antigbnes prdcoces et ddtectds par ~ immunoblot ~ ne sont pas exclucivement associds it une infection active
La corr61ation entre l'incidence des anticorps dirig6s contre les prot6ines pr6coces du cytom6galovirus humain (HCMV) et le stade de l'infection a 6t6 6tudi6e sur 228 s6rums pr6sentant dcs IgG-HCMV (HCMV-late-antigen ELISA). La pr6sence d'IgG sp6cifiques contre les antig~nes pr6coces de I'HCMV dans ces s6rums obtenus/l partir de personnes souffrant d'infections actives et non actives/~ HCMV, a 6t6 6tudi6e /l l'aide d'un ~micro Western blot~ (Modi-blot). En utilisant la souche AD-169, nous avons trouv6 des anticorps contre les antig6nes pr6coces de HCMV dans 59 °70 des cas, ind6pendemment de l'origine du s6rum. Les personnes A risque pour une infection A cytom6galovirus (transplant6s r6naux, personnes atteintes de SIDA) pr6sentent des r6ponses significativement plus 6tendues que les sujets sains HCMV-s6ropositifs. Pourtant, en comparant les r6actions immunes durant la phase active de l'infec",ion et pendant la convalescence, nous avons relev6 un sujet ne pr6sentant pas de r6ponse immune contre les antig6nes pr6coces de HCMV. Nos r6sultats sugg6rent que les infections actives ne sont pas n6cessairement accompagn6es par la formation d~anticorps contre !es antig~r~es pr6coces. Ces anticorps peuvent m~me persister pendant une infection non active. Mots-clds: CMV, Prot6ine pr6coce; Pathog6nie, Diagnostic, Immunoblotting, Infection active.
References Braun, W. & Abraham, R. (1989), Modified diffusion blotting for rapid and efficient protein transfer with PhastSystem. Electrophoresis, 10, 249-253. Doerr, H.W., Rentschler, M. & Scheifler, G. (1987), Sero-
37
logic detection of active infections with human herpes viruses (CMV, EBV, HSV, VZV): diagnostic potential of IgA class and IgG subclass-specific antibodies. Infection, 15, 93-98. Friedman, A.D., Furukawa, T. & Plotkin, S.A. (1982a), Detection of antibody to cytomegalovirus early antigen in vaccinated, normal volunteers and renal transplant candidates. J. infect. Dis., 146, 255-259. Friedman, A.D., Michelson, S. & Plotkin, S.A. (1982b), Detection of antibodies to pre-early nuclear amigen and immediate-early antigens in patients immunized with cytomegalovirus vaccine. Infect. Immun., 38, 1068-1072. Gergely, L., Czegledy, J. & Vaczi, L. (1981), Complementfixing antibodies to human cytomegalovirns-induced early nuclear antigens in mononucleosis. Med. Microbiol. Immunal., 170, 99-108. Gerna, G., Cereda, P.M., Cattaneo, E., Achilli, G. & Revello, M.G. (1978), Immunoglobulin G to virusspecific early antigens in congenital, primary, and reactivated human cytomegalovirus infections. Infect. Immun., 22, 833-841. Griffiths, P.D., Buie, K.J. & Heath, R.B. (1980), Persistence of high titre antibodies to the early antigens of cytomegalovirus in pregnant women. Arch. Virol., 64, 303-309. Keil, G.M., Fibi, M.R. & Koszinowski, U.H. (1985), Characterization of the major immediate-early polypeptides encoded by murine cytomegalovirus. J. Virol., 54, 422-428. Middeldorp, J.M., Jongsma, J., Ter Haar, A., Schirm, J. & The, T.H. (1984), Detection of immunoglobulin M and G antibodies against cytomegalovirus early and late antigens by enzyme-linked immunosorbept assay. J. clin. Microbiol., 20, 763-771. Musiani, M., Landini, M.P., Feletti, C. & La Placa, M. (1978), Serum antibodies against cytomegaiovirusinduc~-~d"e~ly" and "Lmm~_.iat¢early" antigens and presence of viruria in renal transplantation reci~.Jents. Microbiologica, 1, 101-105. Musiani, M., Zerbini, M., Carpi, C., Plazzi, M., Sermasi, G., Belletti, D. & Sacchi, R. (1984), Se.rological screening for the prevention of transfusion-acquired cytomegalovirus infection. J. Infect., 9, 148-152. Numazaki, Y.~ Oshima, T., Tanaka, A., Konno, T., Tazawa, Y., Karit, M., Ishii, A., Hirota, K., Watabe, N. & Ishida, N. (1980), Demonstration of IgG EA (early antigen) and IgM MA (membrane antigen) antibodies in CMV infection of healthy infants and in those with liver disease. J. Pediatr., 97, 545-549. The, T.H., Anderson, H.K., Spencer, E.S. & Klein, G. (1977), Antibodies against cytomegalovirus-induced early antigens (CMV-EA) in immunosuppressed renai-allograft recipients. Clin. exp. Immunol., 28, 502-505.
Res. Virol. 1991, i42, 33-37
Antibodies against human CMV-induced "early" antigens detected by immunoblotting are not exclusively associated with active infection A. Hamann, W. Braun and H.-W. Doerr Zentrum der Hygiene, Abteilung fiir Medizir, ische Virologie, Universitiitskliniken Frankfurt, Paul-Ehrlich-Strasse 40, 6000 Frankfurt a. M. 70 (Cermany) SUMMARY We investigated the correlation between the prevalence of antibodies against human cytomegalovirus (HCMV} "early" proteins and the state of infection. Two hundred and twenty-eight HCMV-IgG-positive (HCMV late-antigen ELISA) sera, drawn from subjects with active and non-active HCMV infections, were tested for specific IgG antibodies against HCMV "early" antigens by a Western blotting micromethod. Applying virus strain AD-169, we found antibodies to HCMV "early" antigens in 59 % of all cases, irrespective of origin. Subjects with an elevated risk of cytomegelic inclusion disease (renal transplantation patients, AIDS patients) revealed significantly broader immune reactions when compared with healthy HCMV-antibody carriers. However, in follow-up studies we detected one patient who lacked an immune r~=pc:Ise to HCMV "early" antigens during active infection and subsequent convalescence. Our results indicate that active HCMV infection is not necessarily associated with the formation of antibodies against "'early'" :antlnAn=~ J~nrmnup_r_th~s;A antibodies may nersist durina non-active infection. . . . . .
O
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
•
=
v
Key-words: CMV, Pathogenicity, Diagnosis, Early protein; Immunoblotting, Active infection,
INTRODUCTION
Controversial information about the diagnostic evaluation of antibodies against the "early ~' antigens of human cytomegalovirus (HCMV) has often been published. The formation of antibodies against the "early" antigens of HCMV (EA-Ab) is considered to be a marker of an active or recently acquired HCMV infection (The et al., 1977; Musiani et al., 1978; Gerna et al., 1978; Griffiths et al., 1980; Musiani et aL, 1984; Numazaki et aL, 1980). Nevertheless, some
Submitted August 15, 1990, accepted November 19, 1990.
authors (Gergely et al., 1981 ; Friedman et al., 1982a,b; Middeldorp et al., 1984) assume that the presence of EA-Ab cannot be taken as presumptive evidence of active infection. In previous investigations of the prevalence of EA-Ab, immunofluorescenceassays (The et ai., 1977; Musiani et al., 1978; Griffiths et al., 1980; Friedman et al., 1980b), an immunoperoxidase technique (Gerna et aL, 1978) and ELISA (Middeldorp et aL, 1984) have been used for quantitative antibody detection. We have inves-
A. HAMANN
34
tigated the correlation between the prevalence o f EA-Ab and the state o f infection. E A - A b were detected by immunoblotting, a method that has not yet been used to this end.
M A T E R I A L S A N D METHODS
Serum samples We examined 96 HCMV-lgG-positive sera (as pretested by an HCMV late-antigen-specific ELISA) from immunosuppressed (HCMV risk groups) and immunocompetent persons: (1) 29 renal transplantation patients receiving immunosuppressive therapy, (2)33 AIDS patients, and (3)34 randomly selected, immunocompetent, healthy individuals. In addition, we tested 132 sera from 11 individuals, aged between 22-53 years, suffering from active cytomegalic inclusion disease (CID), to examine the course of infection and subsequent convalescence (5 cases of primary and 6 cases of reactivated infection as confirmed by: seroconversion, a 4-fold rise in IgG antibodies, a positive IgM titre to HCMV late antigen or isolation of virus from urine specimens).
ET AL.
FRG) by the Modi-blot method (Braun and Abraham, 1989). Blots (containing the infected and mockinfected cell preparations) were blocked against nonspecific binding with a "blocking" solution (BS=0.02 M Tris, 0 . 2 M NaC1, 0.2 070 NP-40, 0 . 0 1 % thimerosal, 2 070BSA, 5 070goat serum) and incubated with serum samples (diluted 1/40 in BS) overnight. Antigen-antibody reaction was visualized with a biotin/streptavidin amplification method using monoclonal mouse anti-human-IgG (Dianova, FRG; 1/1,000 in BS), biotin-labelled rat anti-mouse-IgG (Dianova, FRG; 1/2,000 in BS), streptavidin/horseradish peroxidase (Gibco, FRG; 1/3,000 in BS) and 3-amino-ethyl-carbazole (AEC)/H202 as enzyme substrate. The criteria used to identify virus-specific "early" proteins was the difference in electrophoretic mobility with respect to host cell polypeptides. Detection of antibodies to HCMV-specific late antigens All sera used in the immunoblot experiments were screened with a commercially available HCMV lateantigen-specific ELISA kit (Enzygnost-Zytomegalie, Behring, FRG) for IgG and IgM antibodies, as previously described (Doerr et al., 1987).
Detection of IgG-speeific antibodies to HCMV "early" antigens by immunoblotting Human foreskin fibroblasts were infected with HCMV AD-169 at an input multiplicity of 2 PFLl/ce!! and incubated in the presence of 100 t~g/ml phosphonoacetic acid to prevent virusspecific DNA synthesis. Cells were harvested at either 15, 25 or 35 h postinfection (p.i.). The resulting crude cell extracts were pooled and virus-specific "early" proteins were extracted from the nuclear fraction as described in detail elsewhere (Keil et al., 1985). The resulting cell lysates as well as crude cell extracts from mock-infected human embryonic fibroblasts were suspended in SDS sample buffer (1 mM Tris/Maleat, 1 mM EDTA, 2.5 070 SDS, 2.5 °70 DTE/bromophenol blue, pH 6.7) and heated to 100°C for 5 rain. Virus-specific "early" proteins and cell proteins were separated by"PhastSystems" eiectrophoresis system (Pharmacia, Sweden) and transferred to an "Immobilon" membrane (Millipore,
BS BSA CID EA-Ab ELISA
= = = = =
blocking solution. bovine serum albumin. cytomegalic inclusion disease. antibodies against HCMV "early" antigens. enzyme-linked immunosorbent assay.
RESULTS
The sera screened by immunoblotting reacted with a total o f 11 electrophoretically distinct HCMV-specific "early" proteins with a molecular weight ranging between 23 and 79 k D a (fig. 1). The prevalence o f E A - A b in immunosuppressed subjects at elevated risk o f CID (renal transplantation and A I D S patients) and in immunocompetent healthy subjects is shown in table I. Each group yielded approximately the same percentage o f positive immunoblots. Sera belonging to the group o f renal transplantation or A I D S patients recognized an average o f 6 + 1 polypeptides, and that o f i m m u n o c o m p e t e n t
HCMV PFU p.i. SDS
= = = =
human cytomegalovirus. plaque-forming unit. postinfection. sodium dodecyl sulphate.
CMV-INDUCED "'EARL Y'" A N T I G E N S A N D A CTIVE INFECTIO, q
A
B
C
D
E
,
F
,>
i~i!i~!~i~~! 4,.1
38 30
-%
2(;
iii
23
Fig. 1. Immune reactions against HCMV "early" proteins detected by immuno~otting. Sera drawn from 6 individuals. Immunoblots A-E are positive, F is negative; ~no!ecularweight~ (kDa) are marked on the side.
Table I. Prevalence of IgG antibodies to HCMV "early" antigens (EA-Ab) and a v e r s e numbc~ of recognized "early" proteins detected by immunob!otting in i~munosuppressed patients at elevated risk for CID and healthy control subjects.
Renal transplantation patients AIDS patients Immunocompetent, healthy subjects
No. pos./no, tested (070) for EA-Ab by immunoblotting
No. of recognized proteins (mean ± SD}
17/29 (59 %)
6 (_+ l)
19/33 (58 070)
6 (_+ 1)
20/34 (59 °70)
3 (+_ l)
35
A. H A M A N N E T A L .
36
healthy subjects, 3 _+1. All immunoblot-positive sera responded at least to the 66-kDa protein. Sera from the majority of the immunocompetent healthy subjects recognized only the 66, 59, 56, 50 or 26-kDa proteins. Sera from most of the immunosuppressed individuals responded, in addition, to the 79, 70, 43, 38, 30 or 23-kDa proteins. All sera tested by immunoblotting were screened with an HCMV late-antigen-specific ELISA for IgM antibodies. Among 30 ~70of the immunosuppressed persons belonging to the HCMV risk groups of renal transplantation and AIDS patients, IgM antibodies against HCMV late antigen were detected compared to 0 % among immunocompetent, healthy individuals. We examined 11 subjects suffering from active HCMV infection, during the course of infection and subsequent convalescence. One out of 11 patients, a 25-year-old immunocompetent woman suffering from primary typical CID (fever, hepatitis, atypical lymphocytes, positive IgM and IgA titres as determined by ! ~CMV lateantigen-specific ELISA), showed no positive immunoblots during the course of active infection or during subsequent convalescence. In all other patients, EA-Ab were detectable. Four out of 10 subjects continued to show positive immunoblots for up to 7 months after convalescence. During the active phase, the number of recognized polypeptides reached the highest value and decreased during convalescence. In 6 patients, EA-Ab were detected only during the active stage. In 5 individuals, HCMV could be isolated from urine, EA-Ab formation being associated with virus excretion. Formation of EA-Ab occurred both in primary and reactivated HCMV infections. DISCUSSION
We studied the immune response to "early" antigens in i m m u n o c o m p e t e n t and immunosupressed persons at elevated risk for CID (renal transplantation and AIDS patients; table I). If a correlation between the incidence of EA-Ab and active HCMV infection exists, it should be reflected by the results of this screen-
ing. Each group yielded approximately the same percentage of reactive immunoblots (59, 58 and 59 %). In 4 out of 10 patients suffering from an active HCMV-infection, EA-Ab could still be detected up to 7 months after convalescence. In agreement with other investigators (Gergely et al., 1981 ; Friedman et al., 1982a,b; Middeldorp et al., 1984), our data confirm that EA-Ab are present during non-active infection. The divergence from the results of those authors detecting EA-Ab exclusively during active infection might result from the higher sensitivity of the immunoblot used in our study. While investigating subjects suffering from an active HCMV infection during the course of infection and convalescence, 1 out of the 11 patients did not show a positive immunoblot during this time. We thus suppose that active HCMV infection does not lead to the formation of EA-Ab in all individuals, as determined by immunoblotting with strain-specific early antigen. The failure to detect EA-Ab in some individuals may be the result of antigenic heterogeneity between HCMV AD-169 (used in our experiments) and other HCMV strains. We have demonstrated that even quantitative EA-Ab detection does not prove active HCMV infection. In most cases, active HCMV infections could be differentiated from non-active infections by protein-specific immune reactions. Sera from the majority of individuals with non-active HCMV infection merely reacted with the 66, 59, 56, 50 or 26-kDa proteins, whereas sera from most of the subjects suffering from active HCMV infection recognized, in addition, the 79, 70, 43, 38, 30 or 23-kDa proteins. In common with Gergely et al., 1981, we also assume that the EA-Ab titre reaches the highest level at the active stage of infection and tends to decline thereafter. In some individuals, EA-Ab formation decreases to such an extent that EA-Ab cannot be detected after active infection. Finally, individual divergence of immune reactions to HCMV "early" antigens, in relation to pathogenesis and prognosis of HCMV infections, should be subject to further investigation.
C M V - I N D U C E D "'EARL Y'" A N T I G E N S A N D A C T I V E I N F E C T I O N
Acknowledgements This work was supported by grant 01ZR8606 from the Bundesminister for Forschung und Technologic, Germany.
Les anticorps anti-cytomdgalovirus humain induits par les antigbnes prdcoces et ddtectds par ~ immunoblot ~ ne sont pas exclucivement associds it une infection active
La corr61ation entre l'incidence des anticorps dirig6s contre les prot6ines pr6coces du cytom6galovirus humain (HCMV) et le stade de l'infection a 6t6 6tudi6e sur 228 s6rums pr6sentant dcs IgG-HCMV (HCMV-late-antigen ELISA). La pr6sence d'IgG sp6cifiques contre les antig~nes pr6coces de I'HCMV dans ces s6rums obtenus/l partir de personnes souffrant d'infections actives et non actives/~ HCMV, a 6t6 6tudi6e /l l'aide d'un ~micro Western blot~ (Modi-blot). En utilisant la souche AD-169, nous avons trouv6 des anticorps contre les antig6nes pr6coces de HCMV dans 59 °70 des cas, ind6pendemment de l'origine du s6rum. Les personnes A risque pour une infection A cytom6galovirus (transplant6s r6naux, personnes atteintes de SIDA) pr6sentent des r6ponses significativement plus 6tendues que les sujets sains HCMV-s6ropositifs. Pourtant, en comparant les r6actions immunes durant la phase active de l'infec",ion et pendant la convalescence, nous avons relev6 un sujet ne pr6sentant pas de r6ponse immune contre les antig6nes pr6coces de HCMV. Nos r6sultats sugg6rent que les infections actives ne sont pas n6cessairement accompagn6es par la formation d~anticorps contre !es antig~r~es pr6coces. Ces anticorps peuvent m~me persister pendant une infection non active. Mots-clds: CMV, Prot6ine pr6coce; Pathog6nie, Diagnostic, Immunoblotting, Infection active.
References Braun, W. & Abraham, R. (1989), Modified diffusion blotting for rapid and efficient protein transfer with PhastSystem. Electrophoresis, 10, 249-253. Doerr, H.W., Rentschler, M. & Scheifler, G. (1987), Sero-
37
logic detection of active infections with human herpes viruses (CMV, EBV, HSV, VZV): diagnostic potential of IgA class and IgG subclass-specific antibodies. Infection, 15, 93-98. Friedman, A.D., Furukawa, T. & Plotkin, S.A. (1982a), Detection of antibody to cytomegalovirus early antigen in vaccinated, normal volunteers and renal transplant candidates. J. infect. Dis., 146, 255-259. Friedman, A.D., Michelson, S. & Plotkin, S.A. (1982b), Detection of antibodies to pre-early nuclear amigen and immediate-early antigens in patients immunized with cytomegalovirus vaccine. Infect. Immun., 38, 1068-1072. Gergely, L., Czegledy, J. & Vaczi, L. (1981), Complementfixing antibodies to human cytomegalovirns-induced early nuclear antigens in mononucleosis. Med. Microbiol. Immunal., 170, 99-108. Gerna, G., Cereda, P.M., Cattaneo, E., Achilli, G. & Revello, M.G. (1978), Immunoglobulin G to virusspecific early antigens in congenital, primary, and reactivated human cytomegalovirus infections. Infect. Immun., 22, 833-841. Griffiths, P.D., Buie, K.J. & Heath, R.B. (1980), Persistence of high titre antibodies to the early antigens of cytomegalovirus in pregnant women. Arch. Virol., 64, 303-309. Keil, G.M., Fibi, M.R. & Koszinowski, U.H. (1985), Characterization of the major immediate-early polypeptides encoded by murine cytomegalovirus. J. Virol., 54, 422-428. Middeldorp, J.M., Jongsma, J., Ter Haar, A., Schirm, J. & The, T.H. (1984), Detection of immunoglobulin M and G antibodies against cytomegalovirus early and late antigens by enzyme-linked immunosorbept assay. J. clin. Microbiol., 20, 763-771. Musiani, M., Landini, M.P., Feletti, C. & La Placa, M. (1978), Serum antibodies against cytomegaiovirusinduc~-~d"e~ly" and "Lmm~_.iat¢early" antigens and presence of viruria in renal transplantation reci~.Jents. Microbiologica, 1, 101-105. Musiani, M., Zerbini, M., Carpi, C., Plazzi, M., Sermasi, G., Belletti, D. & Sacchi, R. (1984), Se.rological screening for the prevention of transfusion-acquired cytomegalovirus infection. J. Infect., 9, 148-152. Numazaki, Y.~ Oshima, T., Tanaka, A., Konno, T., Tazawa, Y., Karit, M., Ishii, A., Hirota, K., Watabe, N. & Ishida, N. (1980), Demonstration of IgG EA (early antigen) and IgM MA (membrane antigen) antibodies in CMV infection of healthy infants and in those with liver disease. J. Pediatr., 97, 545-549. The, T.H., Anderson, H.K., Spencer, E.S. & Klein, G. (1977), Antibodies against cytomegalovirus-induced early antigens (CMV-EA) in immunosuppressed renai-allograft recipients. Clin. exp. Immunol., 28, 502-505.