Serum cytokine levels and antibody response to influenza vaccine in the elderly

Cytokines play critical roles in regulating the antibody response to vaccines. We sought to understand the role of endogenous cytokines in the determination of antibody production in the elderly, a group of subjects known to have a lower response rate to vaccination. We found that in a healthy elderly group, only 52% of whom responded to the influenza vaccine, endogenous levels of interleukin 6 (IL-6), IL-10 and gamma interferon (IFNgamma) did not differ statistically significantly between responders and non-responders (responders: n = 27, IL-6 = 293 +/- 101 pg/ml, IL-10 = 882 +/- 240 pg/ml; nonresponders: n = 26, IL-6 = 223 +/- 71 pg/ml, P = 0.57, IL-10 = 445 +/- 148 pg/ml, mean +/- SE, P = 0.14, respectively, and undetectable IFNgamma). Serum levels of these three cytokines were not changed significantly four weeks after vaccination (P < 0.05 for IL-6 and P < 0.01 for IL-10). In addition, there were also no age-dependent differences in serum IL-6 and IL-10 levels.     

Influence of endotoxin-protein in immunoglobulin G isotype responses of mice to Brucella abortus lipopolysaccharide

Brucella abortus endotoxin preparations, containing approximately 5 to 6% protein, induce strong immune and adjuvant immunoglobulin G (IgG) responses as compared with Escherichia coli endotoxin preparations, with equivalent amounts of protein, which induce responses in which IgM antibody predominates. Using an enzyme-linked immunoassay with isotype-specific conjugates, we found that antibody of all four subclasses of IgG were evoked during the course of the immune responses of C3H/HeAu mice to B. abortus endotoxin. Secondary responses of endotoxin-hyporesponsive C3H/HeJ mice were similar to those seen in C3H/HeAu mice, although lower levels of antibody were produced during their primary responses. The primary responses of BALB/c athymic mice consisted almost entirely of IgG3, and IgG1 appeared following a second injection. The effects of lipopolysaccharide (LPS)-associated protein on the immunogenic properties of B. abortus endotoxin were examined by comparing responses to endotoxin with those to a purified B. abortus LPS containing less than 1% protein. The endotoxin evoked strong primary and secondary responses in which antibody directed to LPS determinants consisted mainly of IgG3 and those to the protein determinants were largely IgG1 antibody. Primary and secondary responses to purified LPS consisted mainly of IgG3 antibody. The potential mechanism of the contribution of protein to the immunogenic properties of the endotoxin as well as possible immune mechanisms involved in these responses are discussed.     

Regulation of intestinal immunoglobulin production in response to dietary ovalbumin.

The effects of oral administration of ovalbumin (OVA) on intestinal immunoglobulin production was examined. Balb/c mice, bred and reared on an OVA-free diet, received either one dose or 14 consecutive daily intragastric doses of 25 mg OVA/dose. Single dose administration of OVA resulted in significant suppression of total immunoglobulins in the intestinal mucosa, particularly of the IgA isotype, although a very low titre anti-OVA IgG class antibody response was induced. After multiple peroral immunisations, there was more intestinal anti-OVA antibody induction and less suppression of total immunoglobulins. However, all the anti-OVA antibody was of the IgG isotype. In vitro production of mucosal immunoglobulins was not significantly reduced over 5 days, compared with controls, in either single or multiple administration groups, suggestive of a loss of T suppressor cell function in culture. Prior adoptive transfer of splenic and lymph node cells from mice preimmunised with OVA was capable of abrogating the local suppression of immunoglobulin production in vivo. Although adoptive transfer of bovine serum albumin-sensitised cells could also overcome some of the local suppression, complete restoration of normal immunoglobulin levels was not achieved. These data suggest that single oral administration of a novel dietary antigen induces a transient, non-specific suppression of intestinal immunoglobulin production, which can be overcome by antigen-specific T cells.     

The role of interleukin-4 in the regulation of sequential isotype switch from immunoglobulin G1 to immunoglobulin E antibody production

The immunoglobulin (Ig)E immune response against protein antigens is profoundly influenced by the antigen dose used for immunization. Whereas immunization of CBA/J mice with low antigen doses results in the production of large amounts of IgE antibody, priming with high antigen doses leads to only marginal IgE antibody production in animals. However, in vitro restimulation of spleen cells from mice primed with high antigen doses leads to considerable activation of IgE-producing B cells, which suggests that B cells primed for IgE antibody production do exist among spleen cells. We investigated the modalities of activation of these memory B cells. The data presented here reveal that the anamnestic IgE immune response in vitro is strictly dependent on the presence of IgG1-expressing B cells, which differentiate after a sequential isotype switch into IgE-producing plasma cells with the help of primed CD4+ T cells. The induction of IgE-producing plasma cells requires a cognate interaction between B cells and CD4+ T cells. Interleukin (IL)-4 seems not to be involved in this process, since IgE production in vitro is resistant to suppression by anti-IL-4 monoclonal antibody. Finally, we show that IgG1-expressing B cells represent a relevant memory cell population in vivo also, but in contrast to the in vitro situation the final differentiation into IgE-producing plasma cells is dependent on IL-4.     

Effect of antigen form on local immunoglobulin A memory response of intestinal secretions to Shigella flexneri.

An enhanced memory response, as shown by increased titers of specific immunoglobulin A (IgA), was seen in intestinal secretions from isolated Thiry-Vella loops in rabbits primed orally with live, locally invasive Shigella sp. X16 and challenged 60 days later with a single oral dose of the same antigen. Heat-killed shigella preparations, when used as either the priming or challenge antigen, did not elicit such a memory response in this system. In the present study, the role of antigen form and dosage in eliciting the enhanced local IgA response was investigated. A noninvasive strain, Shigella flexneri 2457-0, was capable of significantly enhancing the mucosal IgA memory response, whereas heat-killed Shigella sp. X16 was unable to augment the local IgA response, even when the priming dose was increased 100-fold. A proposed mucosal adjuvant, DEAE-dextran, given orally with live Shigella sp. X16, did not enhance the local IgA response. Viable, noninvasive shigellae were effective priming agents in enhancing the local IgA memory response. The poor mucosal response to heat-killed shigella preparations is thought to be related to an ineffective delivery of nonviable bacterial antigens into gut-associated lymphoid tissues. The ability of the live, noninvasive strain to elicit a vigorous local IgA memory response when given orally to rabbits was consistent with previous findings that live preparations elicit the best mucosal IgA response.     

Prolonged and preferential production of polymeric immunoglobulin A in response to Streptococcus pneumoniae capsular polysaccharides.

Streptococcus pneumoniae is an invasive mucosal pathogen for which host defense is dependent on capsular polysaccharide-specific antibody. Capsule-specific immunoglobulin G (IgG), IgM, and IgA are produced following pneumococcal vaccination and infection. Serum IgA has two molecular forms, polymeric and monomeric. These forms may modulate the avidity of antigen binding and evolve over time as the immune response matures. Therefore, we sequentially characterized the molecular forms of serum IgA to three serotypes of pneumococcal capsular polysaccharides (types 8, 12F, and 14) after pneumococcal vaccination and after natural infection with type 14 S. pneumoniae. Although typically the form of IgA in antigen-specific systemic responses to protein antigens is predominantly polymeric in sera of patients shortly after exposure and shifts to the monomeric form in sera obtained several weeks later, the form of IgA in response to each pneumococcal capsular polysaccharide remained predominantly polymeric 1 month after natural infection and up to I year following vaccination. In contrast, IgA to pneumococcal cell wall polysaccharide was both polymeric and monomeric. Moreover, the form of IgA in response to polyribosyl-ribitol-phosphate (PRP), the capsular polysaccharide of Haemophilus influenzae type b, was predominantly monomeric in the sera of 8 of 10 subjects tested 1 to 3 months after vaccination with either PRP alone or the diphtheria toxoid conjugate of PRP. We conclude that systemic responses to pneumococcal capsular polysaccharides are distinct in the production of predominantly polymeric IgA over time. The persistence of polymeric IgA may facilitate binding and clearance of pneumococci from the systemic circulation or reflect limited maturation of the immune response to pneumococcal capsular polysaccharides.     

Influence of antigen processing on thymic T-cell selection.

The design of a specific blocking peptide for the immunosuppressive therapy of an autoimmune disease requires the identification of peptides of an autoantigen that are physiologically processed in vivo and bind to MHC-encoded membrane glycoproteins. However, knowledge of how an antigen is physiologically processed by antigen-presenting cells (APC) in vivo, particularly in the thymus, is lacking. It is also unknown whether the processing of an antigen by different APC in the thymus can influence thymic T-cell selection. This is an important consideration for attempts to delete or inactivate autoreactive T cells that elicit autoimmune disease. To address these issues, we investigated the processing of biosynthetically labelled recombinant human insulin (rHI), a model autoantigen, injected into mice and characterized the insulin peptides associated with MHC class II molecules on thymic epithelial cells and dendritic cells. These APC were found to differ in the way they process insulin. The detection of MHC-class-II-bound insulin peptides on the surface of the epithelial cells but not the dendritic cells correlated with their capacity to either present or not present insulin to T cells, respectively. Thus, antigen processing may control the appearance of different peptide-MHC class II complexes on thymic APC that mediate positive and negative selection, and thereby influence the development of the T-cell repertoire. Our findings could have important bearing on the future design of synthetic blocking peptides that reduce or eliminate the onset of autoimmune disease.     

Anti-Gal binds to pili of Neisseria meningitidis: the immunoglobulin A isotype blocks complement-mediated killing.

alpha 1,3-Galactosyl antibodies (anti-Gal) are ubiquitous natural human serum and secretory polyclonal antibodies that bind to terminal galactose-alpha 1,3-galactose (alpha-galactosyl) residues. Serum immunoglobulin G (IgG) anti-Gal can block alternative complement pathway-mediated lysis of representative gram-negative enteric bacteria that bind it to lipopolysaccharide alpha-galactosyl structures, thereby promoting survival of such bacteria in the nonimmune host. We wanted to know whether anti-Gal also could bind to the lipooligosaccharides (LOS) of Neisseria meningitidis. To our surprise, we found that serum and secretory anti-Gal bound to pili but not to LOS of certain strains. This suggested the presence of an immunogenic pilus carbohydrate epitope. Mild periodate oxidation of sodium dodecyl sulfate-polyacrylamide gel electrophoresis-separated outer membrane preparations from strains that bound anti-Gal followed by labeling of the neoaldehyde groups resulted in the labeling of bands that corresponded to pilin and LOS, confirming that pilin contains carbohydrate structures. A Bandeiraea simplicifolia lectin that also binds terminal alpha 1,3-galactosyl residues also bound to pilin. Serum IgG, IgA, and IgM anti-Gal as well as colostral secretory IgA anti-Gal bound to pilin, as judged by immunoblotting, and to the pili of intact piliated organisms, as judged by immunoelectron microscopy. Total serum anti-Gal (IgG, IgA, and IgM) and purified serum IgA1 anti-Gal, but not its purified IgG isotype, blocked complement-mediated lysis of a piliated meningococcal strain that bound anti-Gal to its pili. Colostral anti-Gal secretory IgA blocked killing of the same strain. Thus, anti-Gal IgA may promote disease when it binds to the pili of N. meningitidis strains.     

Anti-viral immunity induced by recombinant nucleoprotein of influenza A virus. III. Delivery of recombinant nucleoprotein to the immune system using attenuated Salmonella typhimurium as a live carrier

A plasmid encoding the influenza nucleoprotein gene from A/NT/60/68 virus was transduced into the attenuated Salmonella typhimurium aroA- strain SL3261. The bacterial vector expressing the viral gene product was able to induce both humoral and cell-mediated immune responses to the nucleoprotein antigen. CD4+ virus-specific T cells capable of proliferation were readily induced and, in some circumstances, class II major histocompatibility complex (MHC)-restricted cytotoxicity was detected. However, virus-specific class I MHC-restricted cytotoxic T lymphocytes (CTL) were not detected after such immunization. Mice immunized orally with the nucleoprotein-expressing bacteria mounted a strong anti-viral antibody response and spleen cells from such mice proliferated specifically to virus challenge in vitro, producing interferon-gamma (IFN-gamma) and interleukin-2 (IL-2). Orally immunized mice showed significant protection from challenge infection with influenza virus if the mice were also boosted intranasally before infection.     

Defective in vitro antibody production in response to pokeweed mitogen and influenza antigen in patients with Hodgkin's disease

Cultures of peripheral blood mononuclear cells (PBM) from 33 patients with Hodgkin's disease, were stimulated in vitro with pokeweed mitogen (PWM) or influenza antigen. Impaired production of immunoglobulin (Ig) of one or more of the three main classes (IgG, IgM and IgA) in PWM stimulated cultures was found in 22 patients and in 11 patients no Ig of any class was produced. Antibody to influenza virus was detected in PWM stimulated PBM cultures in 13 of 14 normal individuals, but in only four of 25 patients with treated Hodgkin's disease though IgG was produced in 16 of 25. Influenza antigen induced anti-influenza antibody production in 10 of 12 cultures from normal individuals but in only two of 22 from patients. The results confirm our earlier report of defective antibody production in vitro by PBM from patients with Hodgkin's disease and indicate that polyclonally activated production of immunoglobulins of several classes is defective, though in vivo humoral immunity is normal.     

Immunoassay for serologic diagnosis of influenza type A using recombinant DNA produced nucleoprotein antigen and monoclonal antibody to human IgG

Influenza type A nucleoprotein (NP) derived from the full length cloned gene expressed in E. coli was evaluated in a solid phase enzyme immunoassay (EIA) for detection of human antibody to influenza. Monoclonal antibody to human IgG was used for detection. Direct and indirect assays were developed and sera were tested in serial and single dilution formats. Preliminary results indicated that recombinant-and virion-derived NP antigens were comparable in binding ability to plastic and binding human antibody. Eighty-seven paired sera from influenza patients were tested. The most sensitive assay (indirect-serial dilution) detected 56 (64%) rises and the simplest assay (direct-single dilution) detected 43 (49%) rises, compared to 36 (41%) for complement fixation. Paired sera from 18 control patients showed no evidence of antibody rises by any of the assays. Forty-nine paired sera from influenza B infected patients were negative for antibody rises except for one borderline rise by the indirect-serial dilution assay. These results indicate that the use of recombinant DNA derived nucleoprotein for immunoassay is feasible. The sensitivity of immunoassays using NP adsorbed to the solid phase and monoclonal antibody specific for human IgG to detect bound antibody exceeded that of conventional complement fixation testing for establishing serologic evidence of influenza type A infection.     

Development of gut immunoglobulin A production in piglet in response to innate and environmental factors

The current review focuses on pre- and post-natal development of intestinal immunoglobulin A (IgA) production in pig. IgA production is influenced by intrinsic genetic factors in the foetus as well as extrinsic environmental factors during the post-natal period. At birth, piglets are exposed to new antigens through maternal colostrums/milk as well as exogenous microbiota. This exposure to new antigens is critical for the proper development of the gut mucosal immune system and is characterized mainly by the establishment of IgA response. A second critical period for neonatal intestinal immune system development occurs at weaning time when the gut environment is exposed to new dietary antigens. Neonate needs to establish oral tolerance and in the absence of protective milk need to fight potential new pathogens.     

Predominant immunoglobulin A response to phase II antigen of Coxiella burnetii in acute Q fever

Diagnosis of acute Q fever is usually confirmed by serology, on the basis of anti-phase II antigen immunoglobulin M (IgM) titers of >/=1:50 and IgG titers of >/=1:200. Phase I antibodies, especially IgG and IgA, are predominant in chronic forms of the disease. However, between January 1982 and June 1998, we observed anti-phase II antigen IgA titers of >/=1:200 as the sole or main antibody response in 10 of 1,034 (0.96%) patients with acute Q fever for whom information was available. In order to determine whether specific epidemiological or clinical factors were associated with these serological profiles, we conducted a retrospective case-control study that included completion of a standardized questionnaire, which was given to 40 matched controls who also suffered from acute Q fever. The mean age of patients with elevated phase II IgA titers was significantly higher than that usually observed for patients with acute Q fever (P = 0.026); the patients were also more likely than controls to live in rural areas (P = 0.026) and to have increased levels of transaminase in blood (P = 0.03). Elevated IgA titers are usually associated with chronic Q fever and are directed mainly at phase I antigens. Although the significance of our findings is unexplained, we herein emphasize the fact that IgA antibodies are not specific for chronic forms of Q fever and that they may occasionally be observed in patients with acute disease. Moreover, as such antibody profiles may not be determined by most laboratories, which test only for total antibody titers to phase I and II antigens, the three isotype-specific Ig titers should be determined as the first step in diagnosing Q fever.     

Immunoglobulin isotype production by cycling human B lymphocytes in response to recombinant cytokines and anti-IgM

Actively cycling populations of purified human tonsilar B lymphocytes were examined for their capacity to secrete IgM, IgA, IgE and IgG of all four subclasses in direct response to recombinant cytokines; in some experiments, monoclonal antibody to IgM (anti-mu) was included in order to explore the influence of antigen receptor ligation on immunoglobulin (Ig) production. Enhanced IgM release was seen on culture of the cycling cells with either interleukin-2 (IL-2), IL-4 or interferon-alpha (IFN-alpha). IL-2 and IFN-alpha also augmented IgA production, whereas IL-4 had no effect on this isotype. IL-4 did, however, encourage the production of the IgG subclasses IgG1, IgG2 and IgG3, while IL-2 augmented IgG1 and IgG3 release and IFN-alpha increased IgG1 levels. IgG4 production, and that of IgE, failed to be perturbed by any of the cytokines assayed. Neither IL-1 alpha, IL-1 beta, IL-5 nor IFN-gamma significantly altered the profile of Ig isotype release. When confronted with anti-mu, cycling B cells demonstrated a marked suppression in IgM production. Suppression could not be overcome by the addition to culture of the normally IgM-promoting IL-4. Concomitant with the reduction in IgM levels, an increase in IgG release was observed. This was comprised of elevations in IgG1 and IgG3. Although not influencing IgA release directly, anti-mu was found to promote increased IgA production in co-culture with either IL-2 or IFN-alpha. The findings are discussed in the context of recent findings on Ig isotype control in both human and murine systems.     

Modulation of immunoglobulin production and cytokine mRNA expression in peripheral blood mononuclear cells by intravenous immunoglobulin

Intravenous immunoglobulin (IVIG) has the potential to regulate Ig production, but the mechanism(s) responsible for this effect is unknown. In experiments reported here, we examined the ability of IVIG to regulate Ig production in human peripheral blood mononuclear cells (PBMCs) stimulated with pokeweed mitogen (PWM). IVIG (2–10 mg/ml) showed a potent (80–85%) inhibition of PWM-stimulated IgG, IgM, and IgA production. To determine more precisely how IVIG mediated the inhibition of Ig production, we studied Ig promoting cytokine gene expression after PWM stimulation with or without IVIG (2 and 10 mg/ml) using dot-blot techniques. RNA was isolated from PBMCs at predetermined time points and probed with cDNAs specific for human cytokines (IL-1-, IL-2, IL-2R, IL-4, IL-5, IL-6, -IFN, and TNF-). IL-6 mRNA accumulation was maximal at 4.5 hr post-PWM stimulation and was inhibited 64–75% when IVIG (10 mg/ml) was present. -IFN mRNA levels peaked at 72 hr poststimulation and were also 68–75% inhibited by IVIG. IL-2 mRNA levels peaked at 4.5 hr and were 23–46% inhibited by IVIG. The inhibitory effect of IVIG on production of these cytokines (IL-6 and -IFN) was also observed at the protein level in sonicated PBMCs after incubation with PWM and IVIG. The mRNA levels for other cytokines were not or only minimally inhibited by IVIG. Addition of IL-6, -IFN, or IL-2 partially restored Ig production in IVIG-treated PWM-stimulated cultures, suggesting that inhibition of other cytokines or another mechanism(s) independent of cytokine inhibition might also be involved, although inhibition of IL-6, -IFN, and IL-2 may be one of the critical factors in the suppression of Ig production by IVIG.     

Bile immunoglobulin of the duck (Anas platyrhynchos). II. Antibody response in influenza A virus infections

The capacity of the IgM-like bile immunoglobulin (IgX) of the duck (Anas platyrhynchos) to express antibody activity to H3N2 influenza A viruses, and the dependence of this activity on the co-existence of serum IgM antibodies were investigated. Ducklings infected orally and intranasally at 15-29 days of age with viruses isolated from different host species were examined for haemagglutination-inhibiting (HI) antibodies in biles and sera 16-29 days after infection (p.i.). All biles had antibodies associated with IgX; all sera had antibodies associated only with the 7.8S IgG. Following oral infection of birds 42-days-old with influenza A/duck/HK/7/75 virus, serum HI antibodies were an initial IgM response occurring from 5-12 days p.i., followed by the appearance of 7.8S IgG antibodies. Virus-neutralizing (VN) antibodies in serum were also biphasic; isotype classification was not attempted. Bile IgX developed HI and VN activity. HI antibodies reached peak titres 12 days p.i. and fell to low levels by 24 days p.i. VN antibodies also reached peak titres 12 days p.i., but thereafter persisted at quite high levels throughout the experiment. Development of high titres of antibody in bile coincided with the termination of virus excretion in faeces. These experiments confirm that bile IgX of the duck can function as antibody in response to influenza A viruses, and that its activity appears to be independent of serum IgM. Its possible relevance in determining survival of virus in the intestine is discussed.     

Antibody-binding epitope differences in the nucleoprotein of avian and mammalian influenza A viruses

Abstract Influenza virus nucleoprotein (NP) binds to the viral genome RNA and forms the internal ribonucleoprotein complex of the virus particle. Avian and human influenza virus NP have characteristic differences at several amino acid positions. It is not known whether any of these differences can be recognized by antibodies. In the present study five monoclonal antibodies (MAbs) were produced against NP of A/Duck/Novosibirsk/56/05 (H5N1) influenza virus. Two MAbs discerned human and avian influenza strains on ELISA testing. The NP expressed in a prokaryotic system was used for the analysis of site-specific mutants carrying amino acid substitutions in the relevant positions. Amino acid residues in positions 100 and 101 were shown to be recognized by the MAbs. The residue in position 100 is host-specific, and its recognition by the MAb 2E6 may be useful for the differentiation of human and avian viruses. The data are discussed in view of the effects of amino acid substitutions in influenza virus NP affecting both host range and antibody-binding specificity.     

Influence of mode of delivery on cytokine expression in cord blood

The mode of delivery is a known risk factor for immune-related disorders. Normal term vaginal delivery is an inflammatory process and several cytokines are suggested to be involved. The purpose of the study was to evaluate differences in cord blood cytokine expression between modes of delivery in term-born children.Cord blood was collected from 49 elective Caesarean section (C-section) cases and from 49 normal vaginal term deliveries. Plasma was tested for 17 cytokines with Bio-Plex®-200-system. Mann-Whitney test was used for comparing the groups with Bonferroni correction for multiple testing.Four cytokines showed significant differences between the modes of delivery. Interleukin-6, Interleukin-8 showed a significantly higher expression in the vaginal delivery group, while Tumor-Necrosis Factor-a, Granulocyte-Colony Stimulating Factor showed a significantly higher level of expression in the C-section cord blood.Our study shows that there is differential expression of pro-inflammatory cytokines in elective C-section compared with normal term vaginal delivery.     

The immune response to measles virus in mice. T-helper response to the nucleoprotein and mapping of the T-helper epitopes.

The T-helper response to the measles virus nucleoprotein (NP) has been studied in mice. The T-cell proliferative response was measured in lymphocytes from mice immunized with either a vaccinia measles-NP recombinant virus or a mouse neuro-adapted measles virus. A T-cell response was obtained with lymphocytes from H2d or H2k mice when stimulated with either measles virus or the NP expressed in bacteria. The response was CD4+ specific. The T-helper epitopes were mapped using truncated NP peptides. The major epitopes in both H2d and H2k mice were determined to be between amino acids 67-98. A further T-cell epitope (between amino acids 457-525) was identified when H2d mice were immunized with measles virus. Studies to quantitate the precursor cells for these epitopes confirmed that the region 67-98 of NP was immunodominant in both haplotypes immunized with the vaccinia-NP recombinant virus, whereas an additional major epitope was observed in the measles virus-infected H2d mice. The primary structure of the epitopes determined here are compared to predicted T-cell epitope motifs.     

The antibody response to HBs antigen is regulated by coordinated Th1 and Th2 cytokine production in healthy neonates

A proportion of healthy neonates fail to produce protective levels of anti-HBs antibody following vaccination with recombinant hepatitis B vaccine. This study was undertaken to investigate contribution of Th1 and Th2 responses to anti-HBs antibody production and to explore the mechanism(s) of unresponsiveness to HBsAg in human neonates. Peripheral blood manonuclear cells (PBMCs) were isolated form 28 nonresponder (anti-HBs antibody <10 IU/l) and 25 responder neonates. The cells were stimulated in vitro with recombinant HBsAg and PHA mitogen and concentrations of IL-4, IL-10 and IFN-gamma were quantified in culture supernatants by sandwich ELISA. Our results demonstrated significantly increased production of all cytokines, including IL-4 (P < 0.001), IL-10 (P < 0.002) and IFN-gamma (P < 0.01) in responder compared to nonresponder vaccinees. No significant differences, however, were observed between the two groups of neonates in the levels of cytokines induced by PHA or secreted in absence of antigen and mitogen. Our findings suggest that unresponsiveness to recombinant HBsAg in healthy neonates is linked to inadequate secretion of both Th1 and Th2 cytokines.