Human immune responses to oral micro-organisms. I. Association of localized juvenile periodontitis (LJP) with serum antibody responses to Actinobacillus actinomycetemcomitans.

The association between periodontal disease in humans and serum and salivary antibody to Actinobacillus actinomycetemcomitans strain Y4 was determined. An Elisa was used to examine anti-Y4 antibody of the IgM, IgG, IgA and IgE isotypes in serum from 127 individuals and IgA in parotid saliva. Patients diagnosed as having localized juvenile periodontitis (n = 37) had significantly higher levels and frequency of serum IgG antibodies to Y4 than all other groups. Serum and salivary IgA and serum IgE antibody levels were significantly increased in patients with both localized and generalized types of juvenile periodontitis (n = 48) when compared to all other patient groups. Specificity studies suggested that the antigenic determinants that were differentiating the group responses were unique to the Y4 organism. These results indicate that serum antibodies to Y4 may reflect an infectious process with this micro-organism and that these responses may provide some diagnostic value in delineating different types of periodontal diseases.     

Induction of Salivary Antibody Responses in Rats after Immunization in Peyer''s Patches

We examined salivary, milk, and serum antibody levels after immunization in the Peyer's patches (Pp) of rats with horse spleen ferritin. Priming of the Pp one day after parturition led to the appearance of IgG, but not IgA or IgM, anti-ferritin antibodies in saliva 9 days later. IgG and IgM antibodies were detected both in milk and in serum, whereas IgA antibodies could only be demonstrated in milk. During a second lactation period the salivary antibodies had vanished but IgG antibodies could still be detected in milk and serum. During a third lactation period, when the rats were immunized in the Pp a second time, not only IgG but also IgA anti-ferritin antibodies appeared in the saliva. Salivary IgG antibody levels and milk IgG, IgM, and IgA antibody levels were higher than those observed after primary immunization in the Pp. The IgG antibody activity in the saliva was positively correlated to the serum IgG antibody activity. It is concluded that salivary IgA antibody responses can be induced by immunization in the Pp. The results of this study suggests that IgA antibodies detected in saliva are produced locally by cells that have migrated from the intestinal lymphoid tissue to the salivary glands.     

Specificity of secretory antibodies to bacterial immunogens.

The present investigation examined the specificity of the salivary immune response of axenic and conventional mice to topically administered Salmonella typhi, S. gallinarum, and S. typhimurium. Specific antibacterial antibodies were determined by passive hemagglutination and bacterial agglutination. Reciprocal antibody titers up to 320 were detected in saliva from mice immunized and assayed with homologous antigens. Antibodies to heterologous immunogens exhibited lower mean titers of 10 to 20 under identical conditions. High concentrations of specific antibodies to the somatic (O) antigen were detected in the saliva of mice administered these microorganisms; however, no significant differences in serum antibody levels were detectable after oral immunization. Only low levels of specific antiflagellar (H) antibodies were demonstrated in the saliva of immunized mice, whereas mean reciprocal titers of 20 were observed in the serum. Antibodies to the Vi antigen of S. typhi were detected in the saliva and serum of only those mice administered formalin-treated S. typhi. Examination of the classes of antibody elicited by these organisms indicated that immunoglobulin A (IgA) was the predominant class in saliva against the O antigens. The salivary response to the H antigens was comprised of both IgG and IgA, whereas the specific serum immunoglobulins were consistent with a primary humoral immune reaction. Local antibodies formed in response to the Vi antigen were exclusively IgG. Serum immunoglobulins produced after peroral administration of the somatic and virulence antigens were limited to the IgM class.     

Class and subclass-associated specificity differences of anti-gliadin antibodies from mucosa and serum.

The class and subclass distribution of antibodies against gliadin in intestinal lavage fluid, saliva and serum was investigated in individuals with coeliac disease. Serum antibodies against gliadin were mainly or even exclusively of the IgA1 subclass. In intestinal lavage fluid and saliva, antibodies of both IgA1 and IgA2 subclasses were found. In patients with and without IgA deficiency, an IgG response was detected both in serum and intestinal lavage fluid with a predominance of IgG1 in selected patients. Specific IgG2, IgG3 and IgG4 antibodies were also detected in intestinal lavage fluid, while no specific IgG2, IgG3 or IgG4 antibodies were found in serum, suggesting a local production of specific IgG antibodies. In Western blot analysis, intestinal lavage fluid and serum IgA antibodies reacted against gliadin components with a MW between 33,000 and 42,000. Serum IgA1 antibodies directed against a gliadin component with a MW slightly higher than 42,000 were also observed. Specific IgG and IgM antibodies in both the secretion and serum against gliadin components with a MW between 33,000 and 42,000 were also detected. This study shows that mucosa-derived gliadin-specific IgA and IgG antibodies may be produced even when there is an absence of specific antibodies of the corresponding immunoglobulin subclass in serum. Furthermore, the specificity of serum and intestinal lavage fluid anti-gliadin IgA1 antibodies may differ.     

Effective immunity to dental caries: gastric intubation of Streptococcus mutans whole cells or cell walls induces protective immunity in gnotobiotic rats.

Gnotobiotic rats were given Streptococcus mutans 6715 whole cells (WC), purified cell walls (CW), or cell wall lysate by gastric intubation (GI), and assessments were made of humoral immune responses in serum and saliva and of caries protection. Levels of secretory immunoglobulin A (IgA) and IgG antibodies to S. mutans WC in saliva samples from experimental rats were determined by an enzyme-linked immunosorbent assay. Serum antibody levels of the IgM, IgG, and IgA isotypes were also determined. Similar levels of salivary antibodies were induced in rats given S. mutans WC or CW by GI, whereas lower salivary antibody titers were observed in rats given cell wall lysate by the oral route. The level of serum antibodies in the various groups of rats also reflected the oral antigen used. The specificity of salivary IgA and serum IgG antibodies in the various groups of rats was determined by enzyme-linked immunosorbent assay with lipoteichoic acid, serotype g carbohydrate, dextran, CW, and WC as coating antigens. Salivary IgA and serum IgG antibodies in rats given S. mutans WC or CW by GI were primarily directed to lipoteichoic acid and serotype g carbohydrate. The presence of salivary IgA antibodies to S. mutans in rats given either S. mutans WC or CW by GI correlated with a significant reduction in the levels of plaque, numbers of viable S. mutans in plaque, and caries scores when compared with the control animals (infected only). These results demonstrate that particulate antigens of S. mutans induce salivary immune responses when given by GI to gnotobiotic rats and that the presence of these antibodies correlates with caries protection.     

IgA and IgM rheumatoid factors in serum, saliva and other secretions: relationship to immunoglobulin ratios in systemic sicca syndrome and rheumatoid arthritis.

Paired serum and saliva samples from seven patients with systemic sicca syndrome (SSS), 15 patients with rheumatoid arthritis and a positive Schirmer's test (RA+), 15 patients with rheumatoid arthritis and negative Schirmer's test (RA-) and 14 normal individuals were analysed for albumin and immunoglobulin concentration as well as IgA and IgM rheumatoid factor (RF) activity. Protein levels in saliva were higher in SSS and RA+ but, when corrected for serum concentration and salivary flow rate, only the IgG ratio remained significantly elevated in SSS (P less than 0.01) and RA- (P less than 0.05) and the IgM ratio was reduced in RA- (P less than 0.05) compared to controls. Although IgM RF activity in serum and saliva was strongly correlated (P less than 0.001) in all three patient groups, the activity in saliva was considerably lower than serum activity. In the two (RA) patients tested, IgM RF in saliva contained secretory component. Mean salivary IgA RF activity varied between 34% (RA-) and 84% (SSS) of serum activity and correlated with serum activity in SSS (P less than 0.001) and RA- (P less than 0.01). IgA RFs in saliva, but not in serum, contained secretory component. Additional demonstration of IgA RF activity in nasal and duodenal secretions in SSS may be related to involvement of the common mucosal immune system.     

Immunoglobulin levels in saliva in individuals with selective IgA deficiency: Compensatory IgM secretion and its correlation with HLA and susceptibility to infections

Total levels of IgA, IgM, and IgG were measured in unstimulated whole saliva and serum from 63 individuals with selective IgA deficiency. Values were compared with the incidence of upper respiratory tract infections, antibiotic treatments (necessitated by upper respiratory tract infections), and HLA antigens. A statistically significant increase in salivary IgM and IgG levels was noted in individuals with selective IgA deficiency compared to healthy normal individuals. Healthy individuals with selective IgA deficiency did not have increased concentrations of salivary IgM compared to infectious-prone patients. Nor was there any correlation found between proneness to infections and HLA antigens or between salivary IgM or IgG levels and HLA antigens in this patient material.     

Antibody responses in arthritic and uncomplicatedYersinia enterocolitica infections

Concentrations of antiyersinia antibody isotypes IgG1, IgG2, IgG3, IgG4, IgA and IgM were measured in 33 patients with yersiniosis using a solid-phase radioimmunoassay. Sixteen patients had a complicating reactive arthritis. Throughout the observation period IgG1 and IgM antibodies both constituted approximately one-third of the total antibodies, while IgA accounted for 10%, IgG3 accounted for 1%, and IgG4 antibodies could not be detected. IgG1, IgM, and IgA antibodies (and the total titer) had reached their peak at the beginning of the observation period (ca. day 20 after the onset of symptoms). The levels then gradually decreased; the total titers averaged 40 times the background at the beginning of the observation period and 4 times the background on day 350. IgM antibodies could be detected as late as a year after the infection. The concentration of IgG2 antibodies varied greatly from patient to patient. In most patients it increased until a plateau was reached approximately 2 months after the onset of symptoms. A decline was observed later. Five arthritic but no nonarthritic patients had a pronounced IgG2 response (more than half of the IgG antibodies were IgG2 in one or several samples).     

Detection of anti-SS-A/Ro and anti-SS-B/La antibodies of IgA and IgG isotypes in saliva and sera of patients with Sj?gren's syndrome.

To assess the frequency and the possibility of local production of autoantibodies against SS-A/Ro and SS-B/La in patients with primary Sj?gren's syndrome (SS), serum and saliva samples were obtained from 42 patients with SS, 10 with rheumatoid arthritis without sicca syndrome, and 12 healthy volunteers. Autoantibodies were detected using enzyme-linked immunosorbent assay and immunoblotting. The frequencies of IgA anti-SS-A antibody, IgA anti-SS-B antibody, IgG anti-SS-A antibody and IgG anti-SS-B antibody in serum from SS patients were 45%, 50%, 43% and 21%, respectively. The frequencies of IgA anti-SS-A antibody, IgA anti-SS-B antibody, IgG anti-SS-A antibody and IgG anti-SS-B antibody in saliva from SS patients were 31%, 33%, 40%, and 19%, respectively. We also found secretory IgA anti-SS-A and anti-SS-B antibodies accompanying secretory components in saliva and sera in representative SS patients. Significant correlations were found between serum and salivary levels of IgA anti-SS-A antibodies, and between serum and salivary levels of IgA anti-SS-B antibodies in SS patients. Significant correlations were also found between serum and salivary levels of IgG anti-SS-A antibodies, and between serum and salivary levels of IgG anti-SS-B antibodies in SS patients. Immunoblot analysis confirmed the presence of IgA-class autoantibodies against SS-A and SS-B in saliva and serum from representative patients. The presence of IgA- and IgG-class autoantibodies against SS-A and SS-B and those accompanying secretory components in saliva from SS patients suggests the local production of these antibodies and the relationship between local and systemic antibody responses.     

Profiling antibodies to class II HLA in transplant patient sera

Immunizing events including pregnancy, transfusions, and transplantation promote strong alloantibody responses to HLA. Such alloantibodies to HLA preclude organ transplantation, foster hyperacute rejection, and contribute to chronic transplant failure. Diagnostic antibody-screening assays detect alloreactive antibodies, yet key attributes including antibody concentration and isotype remain largely unexplored. The goal here was to provide a detailed profile of allogeneic antibodies to class II HLA. Methodologically, alloantibodies were purified from sensitized patient sera using an HLA-DR11 immunoaffinity column and subsequently categorized. Antibodies to DR11 were found to fix complement, exist at a median serum concentration of 2.3 μg/mL, consist of all isotypes, and isotypes IgG2, IgM, and IgE were elevated. Because multimeric isotypes can confound diagnostic determinations of antibody concentration, IgM and IgA isotypes were removed and DR11-IgG tested alone. Despite removal of multimeric isotypes, patient-to-patient antibody concentrations did not correlate with MFI values. In conclusion, allogeneic antibody responses to DR11 are comprised of all antibody isotypes at differing proportions, these combined isotypes fix complement at nominal serum concentrations, and enhancements other than the removal of IgM and IgA multimeric isotypes may be required if MFI is to be used as a means of determining anti-HLA serum antibody concentrations in diagnostic clinical assays.     

Antibody responses of monkeys to oral and local immunization with Streptococcus mutans.

Monkeys were immunized with Streptococcus mutans by a number of routes in an attempt to elicit exclusively a secretory immunoglobulin A (IgA) response. Antibody responses were detected by a sensitive radioimmunoassay. Monkeys primed subcutaneously and boosted submucosally with formolized cells of S. mutans had high serum IgG, IgA, and IgM radioimmunoassay titers and only slight salivary IgG titers. Instillation of killed cells of S. mutans into the right parotid salivary duct elicited good IgG, IgA, and IgM responses in both the right parotid saliva and serum, but only a weak IgM response was detected in the left parotid saliva. Administration of killed cells of S. mutans in enterically coated capsules did not elicit a detectable antibody response or have a discernible effect on the antibody response to subsequent immunization by instillation. No increase in antibody titer was detected in the serum or whole saliva from monkeys orally immunized with enterically coated capsules containing viable S. mutans or in the serum, whole saliva, or intestinal contents from monkeys immunized with uncoated capsules containing killed cells of the same organism. These results do not support the concept that oral immunization with S. mutans is effective in stimulating a generalized secretory IgA response in primates.     

Salivary immunoglobulin A and serum antibodies to Streptococcus mutans ribosomal preparations in dental caries-free and caries-susceptible human subjects

Caries-free subjects or individuals with low caries susceptibility exhibited significantly higher (P less than 0.001) levels of naturally occurring salivary immunoglobulin A (IgA) and serum IgG, IgA, and IgM antibodies to a Streptococcus mutans ribosomal preparation than subjects with high caries susceptibility. Absorption of saliva and serum samples with S. mutans ribosomal preparations, but not with other S. mutans antigens or with Escherichia coli and Neisseria gonorrhoeae ribosomal preparations, removed the antibody activity. Absorption with Streptococcus sanguis ribosomes and NH4Cl-washed S. mutans ribosomes partially removed the anti-S. mutans ribosome antibody activity. These results provide evidence that naturally occurring salivary and serum antibodies to the S. mutans ribosomal preparation correlate with susceptibility to dental caries.     

Salivary sIgA response in HIV-1 infection.

Local and systemic production of total and HIV-1 specific IgA was determined in whole saliva and serum from 45 HIV-1-infected individuals and 15 healthy controls. The antigenic domains important for sIgA and IgG binding, respectively, was investigated with epitope mapping using synthetic peptides of HIV-1 proteins. Decreased levels of total sIgA in saliva were found among patients with low CD4+ cell counts in advanced stages of acquired immunodeficiency. HIV-1 specific IgA response, predominantly directed to the envelope proteins, was found in saliva and serum also at later stages of disease. Analyses using peptide enzyme-linked immunosorbent assays (ELISA) showed that the sIgA antibody response in saliva was mainly directed to the V4 region (aa 385-409) and a more C-terminal part of the V3-region (aa 325-344) compared with the IgG response, which predominantly was directed to a more central part of the V3 loop (aa 308-325). A similar picture was seen for immunoglobulins of the two isotypes derived from serum. We have in this study shown IgA epitope-specific immune response within HIV-1 gp160, indicating that antibodies of IgA isotype may recognize somewhat different antigenic domains compared to IgG antibodies.     

Salivary antigliadin and antiendomysium antibodies in coeliac disease

The definitive diagnosis of coeliac disease is based on typical changes in the small intestine biopsy specimens. To screen individuals for coeliac disease serum IgA and IgG antigliadin (AGA), IgA antireticulin (ARA) and IgA antiendomysium (EmA) antibodies are used. The aim of this study was to investigate whether these antibodies can also be detected in saliva as diagnostic markers of coeliac disease. The study population comprised 30 patients with coeliac disease treated with a gluten-free diet, 14 patients with untreated coeliac disease and 13 healthy control subjects. Sera and saliva were tested simultaneously for the presence of IgA and IgG AGA and IgA EmA. None of patients studied had a selective IgA deficiency. There was no significant difference in salivary IgA AGA levels between the three groups tested and there was no correlation between the individual serum and salivary values of IgA AGA. Salivary IgG AGA levels were very low or undetectable. Serum IgA AGA showed a low sensitivity (36.4%) to detect an untreated patient with coeliac disease. All salivary samples, regardless of the study group were negative for IgA EmA. Serum IgA EmAs were universally detected in the sera of patients with newly diagnosed coeliac disease and also in the sera of five of 30 patients with treated coeliac disease. No IgA EmA was detected in the sera of controls. None of the patients studied had a selective IgA deficiency either. Serum IgA EmA is the most sensitive, and IgA and IgG AGA are good indicators for coeliac disease, but salivary IgA or IgG AGA and salivary IgA EmA are not helpful for the diagnosis or follow-up of coeliac disease patients.     

Mucosal immune responses to meningococcal group C conjugate and group A and C polysaccharide vaccines in adolescents

Previous studies in children have shown that Haemophilus influenzae type b (Hib) polysaccharide conjugate vaccines can reduce nasopharyngeal carriage of H. influenzae and provide herd immunity and suggest that this effect is mediated through mucosal antibodies. As this phenomenon may operate in other invasive bacterial infections which are propagated by nasopharyngeal carriage, mucosal antibody responses to meningococcal C conjugate and A/C polysaccharide vaccines were investigated. A total of 106 school children aged 11 to 17 years were randomized to receive a single dose of either conjugate or polysaccharide vaccine in an observer-blind study. Before and at 1, 6, and 12 months after immunization, samples of unstimulated saliva were collected and assayed by enzyme-linked immunosorbent assay for group C polysaccharide-specific immunoglobulin A (IgA), IgA1, IgA2 and secretory component, IgG antibodies, and total IgG and IgA. A subset of serum samples were also assayed for specific IgA and IgG antibodies. The concentrations of specific IgA and IgG in saliva were expressed both as nanograms per milliliter and as nanograms per microgram of total IgA or IgG. One month after immunization, significant increases in antibody titers (both IgA and IgG) were observed in saliva in both groups. There were significant subsequent falls in antibody titers by 6 months. Anti-meningococcal C-specific secretory component and IgA antibody titers were closely correlated (r = 0.85, P < 0.001), but there was no significant correlation between salivary and serum IgA titers, suggesting that IgA antibodies are locally produced. Significant correlation was found between salivary and serum IgG titers (r = 0.52, P < 0.01), suggesting that salivary IgG may be serum derived. Compared with polysaccharide vaccine, the conjugate vaccine induced significantly higher salivary IgG responses (P < 0.05), although there were no significant differences between salivary IgA responses to the two vaccines. The conjugate vaccine induced greater salivary IgG responses than a polysaccharide vaccine. Both vaccines induced significant salivary IgA antibodies. Further studies are needed to establish the functional significance of these mucosal responses.     

IgA-associated renal diseases: Antibodies to environmental antigens in sera and deposition of immunoglobulins and antigens in glomeruli

Levels of IgA1, IgA2, IgM, and IgG antibodies specific for 10 ubiquitous food and bacterial antigens were examined by radioimmunoassay in the sera of 29 patients with IgA-associated renal diseases and 22 normal individuals. No significant differences were observed between patient and normal groups in the levels of IgA1 antibodies, and IgA2 antibodies were detected in only a few individuals in either group. Minor differences in IgM or IgG antibodies were seen against some antigens. Significant positive correlations between IgA1 and IgG and between IgA1 and IgM antibodies to casein were found in the patient group. Analysis of the molecular form of serum IgA1 antibodies revealed that although the pattern of polymeric and monomeric forms varied between individuals and between antibody specificities, there was no preponderance of one form in either patient or normal groups. Examination of kidney biopsies from 50 patients with IgA-associated renal diseases revealed that IgA1 represented the predominant subclass deposited in the glomerular mesangium; glomeruli from three patients contained both IgA1 and IgA2. Seventy-eight percent of the patients also had deposits of IgM, although IgA and IgM deposits did not always coincide. When IgG was present in glomeruli (45% of patients), the IgG1 subclass predominated. J chain was detectable in glomeruli of only four patients. C3 was detected in glomeruli of 95% of the patients, although the distribution of C3 did not always coincide with that of IgA. Indirect immunofluorescence staining with rabbit antisera to various environmental antigens showed that milk protein antigens could be deposited in association with IgA in the glomerular mesangium.     

Antibodies in different immunoglobulin classes to Candida albicans in allergic respiratory disease

Measurements by radiometric assay of specific IgE, IgA, IgM, igD and IgG antibodies, in fifty-six patients with asthma, fifty-six with rhinitis and twenty with allergic bronchopulmonary aspergillosis (ABPA), showed differences in the patterns of antibody response between the three patient groups, both in terms of levels of class specific antibodies and the proportion of patients giving positive results. Specific IgE antibody values were similar in all three groups and positive results were found in three-quarters of the patients in each group. No other class of antibody to Candida albicans gave values which were similar in all three patient groups. Specific IgG and IgA antibody values were highest in ABPA and lowest in rhinitis; specific IgM antibody values were highest in asthma and lowest in ABPA; specific IgD antibody values were higher in rhinitis than asthma and very low in ABPA. Specific IgG1 and lgG2 antibodies to C. albicans were higher in rhinitis than in asthma, whereas specific IgG3 and lgG4 values were significantly higher in asthma than in rhinitis. There was a greater number of positive results for IgG1, IgG2 and IgG4 in patients with ABPA than in the other two patient groups where values for IgG1 and IgG2 antibody to C.albicans were mainly negative. Positive IgG3 antibody results occurred with greatest frequency in patients with asthma, half of them giving values greater than the mean + 2 s.d. of the control value; positive specific IgG4 antibody values occurred more frequently in asthma than in rhinitis.     

Kinetics of antigen-specific IgM/IgG/IgA antibody responses during Zika virus natural infection in two patients

Understanding of kinetics of antibody responses is crucial for developing rapid serological tests and studying the mechanisms of Zika virus (ZIKV) infection. Most of the serological diagnostic assays previously published are based on either IgM or IgG titer, little is known on the level of IgA antibody in saliva and urine. In this study, we investigated the kinetics of IgM/IgG/IgA antibody responses in serum, saliva, and urine obtained from two ZIKV infected individuals from as early as the second day of onset of symptoms to as long as 2 years postinfection. Other than detecting robust early IgM response, long lasting IgG response, we discovered strong early IgA response specific for ZIKV in saliva in both patients. This unique observation provides a novel strategy and scientific basis for the development of noninvasive rapid tests for ZIKV infection.     

Radioimmunoassay of class-specific antibodies to Streptococcus mutans in monkey serum and saliva.

A radioimmunoassay (RIA) has been developed to measure class-specific antibodies to Steptococcus mutans in the serum and saliva of monkeys (Macaca fascicularis). Anti-human immunoglobulin antibodies purified by affinity chromatography on immobilised monkey immunoglobulins and labelled with 125I were employed. Formolised cells of S. mutans and an extract of culture supernatant adsorbed to polystyrene wells were used as solid-phase antigens. The coefficients of variation for IgG, IgA and IgM assays were less than or equal to 10% for both antigen systems. Two monkeys were immunised with formolised cells of S. mutans by subcutaneous injection and subsequent instillation of bacterial cells into their right parotid ducts. IgG, IgA and IgM antibody responses to S. mutans in samples of serum and saliva were quantitated by RIA. Immobilisation of purified components of S. mutans on polystyrene wells enabled the measurement of antibody response to a number of antigens to be made. The RIA is a sensitive, reproducible and quantitative method of measuring serum and salivary antibody responses in monkeys.     

Antibody response in the parotid fluid and serum of Irus monkeys (Macaca fascicularis) after local immunization with Streptococcus mutans.

The antibody response of Macaca fascicularis in parotid saliva and serum to local immunization by two routes with Streptococcus mutans was studied and compared over 1 year. Antibodies were titrated and classified by indirect immunofluorescent staining using specific antiglobulin conjugates. Antiglucosyltransferase activity was assayed by an enzyme inhibition test. Animals were immunized first by injecting formalin-killed bacterial cells and cell products subcutaneously into the vicinity of the four major salivary glands. The monkeys were next immunized by retrograde instillation of antigen into the parotid duct. Extensive subcutaneous local immunization gave a serum response only. After parotid duct immunization, high titers of immunoglobulin A (IgA) antibody, along with traces of immunoglobulin G (IgG) immunoglobulin M (IgM) antibody, appeared in the parotid saliva, and in the serum high titers of IgG antibody were present along with lower titers of IgA and IgM. IgA antibodies in parotid fluid were shown by double immunofluorescent staining to be associated with antigenic determinants which cross-reacted with an antiserum directed to human secretory component. Titers in parotid fluids and sera fell sharply when immunization was stopped. This response pattern was reproducible. High concentrations of antibody capable of inhibiting glucosyltransferase prepared from S. mutans were found in the sera, but relatively little was detected in the parotid fluids. Extensive immunization via the parotid duct resulted in transient functional impairment of the gland, as evidenced by diminished salivary flow rates. We conclude that parotid ductal immunization can be an effective method for stimulating a salivary secretory IgA antibacterial antibody response.