Share this article on:

Salivary production of IgA and IgG to human herpes virus 8 latent and lytic antigens by patients in whom Kaposi's sarcoma has regressed

Mbopi-Keou, Francois-Xaviera,b; Legoff, Jeromec; Piketty, Christophed; Hocini, Hakimc; Malkin, Jean-Eliee; Inoue, Naokif; Scully, Crispian Ma; Porter, Stephen Ra; Teo, Chong-Geeb; Belec, Laurentc

Research Letters

aDepartment of Oral Medicine, Eastman Institute for Oral Health Care Sciences, University College London, London, UK; bSexually Transmitted and Blood Borne Virus Laboratory, Health Protection Agency, London, UK; cLaboratoire de Virologie, Hôpital Européen Georges Pompidou, Paris, France; dINSERM U430, Institut de Recherches Biomédicales des Cordeliers, Paris, France; eCentre Médicale de l'Institut Pasteur, Paris, France; and fHerpesvirus Section, Centers for Disease Control and Prevention, Atlanta, a, GA, USA. *Current address: National Institute of Infectious Diseases, Tokyo, Japan.

Sponsorship: This work was supported by grants from the French National Agency for AIDS Research (ANRS).

Received: 14 July 2003; revised: 30 July 2003; accepted: 12 August 2003.

IgG and IgA antibodies with specificities to a latent and a lytic antigen of human herpes virus 8 (HHV-8) were detectable in the saliva and serum of eight patients whose Kaposi's sarcoma had regressed, seven of whom were HIV-1 infected. The measurement of antibody-specific activity and secretion rate, and the detection of secretory IgA all indicate anti-HHV-8 antibody activity in saliva. The specific humoral responses possibly influence mucosal replication of HHV-8, and in turn, that of HIV.

The mucosal production of virus-specific IgA may be viewed as a host response that contains or eradicates the virus, as IgA often possesses virus-neutralization capabilities [1]. Such IgA may paradoxically broaden the tropic range of the virus, however [2]. As human herpes virus 8 (HHV-8) has the potential to enhance HIV replication in the mouth [3], we wondered whether anti-HHV-8 IgA responses may influence the replicative activity of HHV-8, and indirectly, that of HIV. We report here the findings of a search for HHV-8-specific IgA in both saliva and serum.

Eight men (age range 30–55 years; seven HIV-1 seropositive and one HIV-1 seronegative) in whom the regression of Kaposi's sarcoma (KS) had been clinically documented were prospectively recruited for the study. Informed consent was obtained. The HIV-infected patients were under highly active antiretroviral therapy (HAART). The single HIV-seronegative patient was being followed up after postoperative radiation and chemotherapy for thymic carcinoma. All were HHV-8 seropositive, as tested by the HHV-8 Check IgG immunoassay (All Diag, Strasbourg, France). None manifested oral lesions. Specimens from 15 age-matched healthy HHV-8-seronegative laboratory workers were used as controls.

Peripheral blood was collected, and serum was separated. Dribbled whole saliva was collected without stimulation for 5 min, and the acellular fraction was separated. Traces of haemoglobin in whole saliva were tested by second derivative spectrophotometry [4]. Concentrations of IgA and IgG in serum and whole saliva were measured using an asymmetric sandwich enzyme-linked immunosorbent assay (ELISA), and human serum albumin (HSA) concentrations in serum and whole saliva were measured by capture ELISA [4].

Anti-HHV-8-specific IgA and IgG antibodies were assayed by indirect ELISA, using as antigens purified extracts of BHK21 cells that had been infected with recombinant Semliki Forest virus expressing the HHV-8 open reading frame (ORF) 73 and ORF K8.1 [5]. In HHV-8-infected primary effusion lymphoma cell lines, the ORF73 protein is a constituent of nuclear antigens expressed during the latent phase of HHV-8 infection, whereas the ORF K8.1 protein is an immunodominant envelope glycoprotein expressed during viral lytic replication. Assays were carried out essentially as previously described [5]. Solid-phase antigen density and sample dilutions (serum was diluted 1 : 100 and whole saliva was diluted 1 : 2 in phosphate-buffered saline-Tween 20 containing uninfected BHK21 cell lysate) were optimized to yield optical density (OD) readings that linearly reflect concentrations of the analyte contained in the well.

Specific activities of IgA to the ORF 73 and ORF K8.1 proteins in serum (AIgA/S) and whole saliva (AIgA/wSa) were calculated with respect to the weight in micrograms of total IgA in serum or whole saliva, respectively, according to the formula: A = OD × d/[IgA], where d is the dilution of the specimen. IgG-specific activities were similarly evaluated. To measure the rate of secretion of immunoglobulins (Ig) from the blood to salivary fluid, the relative coefficients of excretion (RCE) of IgA and IgG in saliva were calculated, using HSA as a reference, according to the formula: RCE = {[Ig]wSa/[HSA]wSa} × {[Ig]S/[HSA]S}.

Molecular forms of IgA antibodies were assessed after the separation of serum and whole saliva on sephacryl S300 columns [4]. Selected fractions containing monomeric, dimeric and secretory IgA were then assayed for specificity to the ORF 73 antigen.

Nested polymerase chain reaction was applied to detect HHV-8 DNA in whole saliva [6].

Fig. 1 compares the specific activity and RCE between antibodies against the ORF 73 and ORF K8.1 antigens. All whole saliva samples were devoid of haemoglobin. IgG antibodies specific to the two antigens were detected in all whole saliva and serum samples. For the ORF 73 antigen, the mean AIgG/whole saliva was 3.4-fold higher than AIgG/serum, whereas for the ORF K8.1 antigen, it was 5.8-fold higher. IgA antibodies specific to the two antigens were also detected in the whole saliva of all patients, and in the serum of all except the one who was HIV seronegative. The mean AIgA/whole saliva values were similar to AIgA/serum (Fig. 1a). RCEIgG and RCEIgA values were greater than 1 for all patients. For the ORF 73 antigen, the mean RCEIgA compared with RCEIgG was 42-fold higher, whereas for the ORF K8.1 antigen, it was 55-fold higher (Fig. 1b). In all patients, secretory IgA antibody to the ORF 73 antigen was present in sephacryl-fractionated whole saliva but not serum (data not shown). HHV-8 DNA was amplifiable in the whole saliva of two patients; there was no correlation between HHV-8-DNA positivity with whole saliva IgG or IgA-specific activity and RCE.

Fig. 1. Anti-human herpesvirus 8 IgG and IgA in serum and saliva.

Fig. 1. Anti-human herpesvirus 8 IgG and IgA in serum and saliva.

This study reveals the presence of antibodies to two HHV-8 antigens in both the blood and saliva of people in whom KS had regressed. In serum, high anti-HHV-8 IgG compared with IgA titres were found, and the IgA was predominantly monomeric, a profile typical of a classic humoral response to persistent viral infection [4]. In whole saliva, the high RCE values obtained for IgA and the presence of secretory IgA point to the oral mucosa as a site of active IgA production. Salivary IgG specific to the two antigens was also being locally produced, because AIgG/whole saliva values were higher than AIgG/serum values, and the RCE for IgG were greater than 1; the lower RCE for IgG than for IgA suggest that the transudation of HHV-8-specific monomeric antibodies from the circulation to the mouth also contributes to salivary IgG carriage.

In our study, the magnitude of IgG responses to the latent ORF 73 antigen was similar to those to the lytic ORF K8.1 antigen. Such a pattern is reminiscent of the heightened response to Epstein–Barr virus (EBV) observed in EBV-associated malignancies, i.e. Burkitt's lymphoma and nasopharyngeal carcinoma [7]. The levels of specific IgA were not particularly high, however, contrasting with the high titres of anti-EBV IgA antibodies observed in nasopharyngeal carcinoma.

Immune restoration after HAART [8] probably contributed to KS clearance in the seven HIV-infected patients. That anti-HHV-8-specific IgA and IgG can be produced in the saliva of the single HIV-uninfected patient raises the possibility that processes independent of HAART also operate to clear KS. Further studies employing a wider array of patients and HHV-8 antigens should lead to a greater understanding of how oral HHV-8 replication and oral HIV shedding may be controlled.

Back to Top | Article Outline


1. Valimaa H, Waris M, Hukkanen V, Blankenvoorde MF, Nieuw Amerongen AV, Tenovuo J. Salivary defense factors in herpes simplex virus infection. J Dent Res 2002; 81:416–421.
2. Sixbey JW, Yao QY. Immunoglobulin A-induced shift of Epstein–Barr virus tissue tropism. Science 1992; 255:1578–1580.
3. Mbopi-Keou FX, Belec L, Teo CG, Scully C, Porter SR. Synergism between HIV and other viruses in the mouth. Lancet Infect Dis 2002; 2:416–424.
4. Belec L, Legoff J, Si-Mohamed A, Bloch F, Mbopi-Keou FX, Becquart P, et al. Mucosal humoral immune response to hepatitis C virus E1/E2 surface glycoproteins and HCV shedding in saliva and cervicovaginal fluids from chronically HCV-infected patients. J Hepatol 2003; 38:833–842.
5. Corchero JL, Mar E-C, Spira TJ, Pellett PE, Inoue N. Comparison of serologic assays for detection of antibodies against human herpesvirus 8. Clin Diagn Lab Immunol 2001; 8:913–921.
6. Moore PS, Kingsley LA, Holmberg SD, Spira T, Gupta P, Hoover DR, et al. Kaposi's sarcoma-associated herpesvirus infection prior to onset of Kaposi's sarcoma. AIDS 1996; 10:175–180.
7. Henle W, Henle G. Epstein–Barr virus and human malignancies. Cancer 1974; 34:1368–1374.
8. Boivin G, Gaudreau A, Routy JP. Evaluation of the human herpesvirus 8 DNA load in blood and Kaposi's sarcoma skin lesions from AIDS patients on highly active antiretroviral therapy. AIDS 2000; 14:1907–1910.
© 2004 Lippincott Williams & Wilkins, Inc.