Expression of Homing Receptors on IgA1 and IgA2 Plasmablasts in Blood Reflects Differential Distribution of IgA1 and IgA2 in Various Body Fluids

Transcription

1 CLINICAL AND VACCINE IMMUNOLOGY, Mar. 2010, p Vol. 17, No /10/$12.00 doi: /cvi Copyright 2010, American Society for Microbiology. All Rights Reserved. Expression of Homing Receptors on IgA1 and IgA2 Plasmablasts in Blood Reflects Differential Distribution of IgA1 and IgA2 in Various Body Fluids Sari H. Pakkanen, 1,2 Jussi M. Kantele, 3 Zina Moldoveanu, 4 Spencer Hedges, 4 Miikka Häkkinen, 2 Jiri Mestecky, 4 and Anu Kantele 1,2 * Division of Infectious Diseases, Department of Medicine, Helsinki University Central Hospital, Helsinki, Finland 1 ; Department of Bacteriology and Immunology, Haartman Institute, University of Helsinki, Helsinki, Finland 2 ; Department of Medical Microbiology and Immunology, University of Turku, Turku, Finland 3 ; and Departments of Microbiology and Medicine, University of Alabama at Birmingham, Birmingham, Alabama 4 Received 10 November 2009/Returned for modification 10 December 2009/Accepted 12 January 2010 Although secretory IgA is the most abundantly produced Ig isotype, the mechanisms underlying the differential distribution of IgA subclasses in various body fluids remain unclear. To explore these mechanisms, we examined the distribution of IgA subclasses, the influence of the nature and sites of encounters with antigens, and the correlation between IgA subclass distribution and homing potentials of circulating IgA plasmablasts. IgA1 predominated in serum, tears, nasal wash fluid, and saliva; the levels of IgA1 and IgA2 were comparable in vaginal wash fluid; and IgA2 predominated in intestinal lavage fluids. Seventy-one percent of circulating IgA plasmablasts secreted IgA1. The intestinal homing receptor (HR), 4 7, was expressed more frequently on IgA2 than on IgA1 plasmablasts, with no differences in the expression of other HRs. IgA subclass distribution among circulating antigen-specific antibody-secreting cells (ASC) was dependent on the nature of the antigen: following vaccination with Salmonella enterica serovar Typhi, unconjugated pneumococcal polysaccharide, or Haemophilus influenzae polysaccharide-diphtheria toxoid conjugate, the proportions of specific IgA1 ASC were 74%, 47%, 56%, and 80%, respectively. HR expression depended on the route of administration: expression of HRs was different after oral than after parenteral vaccination, while no difference was seen between HR expression of antigen-specific IgA1 and IgA2 ASC induced via the same route. The key factors determining IgA subclass distribution in a given secretion are the nature of the antigens encountered at a particular site and the site-specific homing instructions given to lymphocytes at that site. These two factors are reflected as differences in the homing profiles of the total populations of circulating IgA1 and IgA2 plasmablasts. Because humoral immunity is most frequently evaluated by titers of serum antibodies, which are dominated by IgG, the role of IgA as the major Ig isotype produced in humans has not been adequately appreciated: the daily production of IgA (66 mg/kg of body weight/day) exceeds that of all other immunoglobulin classes combined (10, 14, 40). Even if IgA levels in serum are lower than those of IgG, due to a shorter half-life in the circulation, IgA is the predominant immunoglobulin in most external secretions (3, 14, 40). IgA, as the mucosal Ig isotype, plays a dominant role as the first immunological defense barrier in the body, since the mucosal sites act as a portal of entry to the majority of human pathogens (3, 40, 53). In humans, IgA comprises two subclasses, IgA1 and IgA2, which are unequally distributed in the body fluids (4, 13, 34, 36, 41). In serum, IgA1 is dominant, while in secretions, there is a significant contribution of IgA2. In external secretions, both IgA1 and IgA2 are present as secretory IgA (S-IgA), a polymeric form which is more resistant to proteolytic enzymes than any of the other isotypes (40). S-IgA is known to provide protection of mucosal membranes in several complementary * Corresponding author. Mailing address: Helsinki University Central Hospital, Department of Medicine, Division of Infectious Diseases, Aurora Hospital, Building 5, 3rd floor, P.O. Box 348, FIN HUS Helsinki, Finland. Phone: Fax: E- mail: anu.kantele@hus.fi. Published ahead of print on 20 January ways: it can effectively neutralize viruses (46, 47, 49) or toxins (47) and displays antibacterial activity (47, 51). To evade the protective effect of S-IgA at the mucosal sites, some pathogenic bacteria (e.g., Streptococcus pneumoniae, Haemophilus influenzae, Neisseria meningitidis, and Neisseria gonorrhoeae) produce proteases which cleave the IgA1 molecule (31), leaving the IgA2 intact. The unequal distribution of IgA subclasses in different body fluids (5, 13, 28) and among IgA-secreting cells in the circulation (25) and mucosal and systemic lymphoid tissues (11, 30) has been reported in several independent studies. The differential distribution of the two IgA subclasses in secretions has been shown to be accompanied by a similar distribution of IgA1- and IgA2-producing cells at those sites (11, 30). IgA1 cells predominate in the bone marrow, and they are also present in higher proportions in upper parts of the orogastric and the respiratory tracts, while IgA2 cells predominate in the lower part of the gastrointestinal tract (11, 30). The basis for this differential distribution of IgA subclasses in various mucosal sites is not understood. More than two-thirds of all Ig-producing cells in the body reside in the various mucosal and exocrine sites (40), particularly along the gastrointestinal tract, and most of them produce IgA specific for antigens encountered at mucosal surfaces. The major source of the precursors of the intestinal IgA plasma cells is the naïve B cells in the organized lymphoepithelial tissues, such as the Peyer s patches (12). B cells in the germinal 393

2 394 PAKKANEN ET AL. CLIN. VACCINE IMMUNOL. centers of Peyer s patches switch from IgM to IgA under the influence of resident cells and various cytokines (8, 14, 38). Subsequently, the cells migrate to the draining mesenteric lymph node, where they continue to divide and differentiate into plasmablasts (12, 15, 16). Finally, they exit the lymph nodes and migrate via thoracic duct lymph into the blood, which carries them to the target tissues, such as the lamina propria of the gut, to secrete IgA. The homing of these activated lymphocytes from the circulation into mucosal tissues involves multiple steps (33, 43, 48) in which the two central events are (i) the interactions between the chemokine receptors (CCR) on the lymphocyte surface and the chemokines produced in the tissue and (ii) the recognition by the lymphocyte surface homing receptors (HR) of their tissue-specific ligands, called addressins, on the surface of the endothelial cells in the mucosa. The key homing receptor-addressin interaction in lymphocyte homing to the intestinal tissue is believed to be the binding of the gut HR, 4 7, on the lymphocyte surface to its ligand, MAdCAM-1, on the high endothelial venules of the intestinal lamina propria (2). A fraction of IgA plasmablasts in the circulation expresses 4 7 (21, 27), consistent with their ability to home to intestinal lamina propria. Some IgA plasmablasts express other HR (20 22, 26), such as the peripheral lymph node HR, L-selectin (6, 32), and the skin HR, cutaneous lymphocyte-associated antigen (CLA) (44), consistent with the ability of a portion of IgA plasmablasts to home to extraintestinal lymphoid tissues at systemic and mucosal sites (48). The HR guiding the cells to the different extraintestinal mucosal sites have not been identified, but it has been suggested that there would be several different HR and CCR contributing to the selective homing to extraintestinal mucosal sites (4, 54). In humans, it is possible to determine the expression of HR on circulating plasmablasts and interpret these data as homing potentials or homing profiles of these cells (22, 26, 27, 45). Furthermore, it might be assumed that the differential distribution of IgA1 and IgA2 plasma cells in the various sites in the body would depend on a differential homing potential of circulating IgA1 and IgA2 plasmablasts and, thus, could be predicted by the differential expression of HR. Although there are data concerning the differential distribution of IgA subclasses in various body fluids (5, 9, 13, 34, 36, 41, 42), studies involving parallel distribution of IgA subclasses in various body fluids and circulating plasmablasts expressing mucosal and/or systemic HR are not available. In the present study, we (i) determined the differential distribution of IgA subclasses in various body fluids using identical assays and reagents, (ii) evaluated the factors influencing IgA subclass distribution of a specific immune response, and (iii) explored whether the differential distribution of IgA subclasses in the various body fluids is associated with differential homing profiles of IgA1 and IgA2 plasmablasts in the circulation. MATERIALS AND METHODS In the present study, we investigated two types of cells in the circulation: all immunoglobulin-secreting cells (ISC) and antigen-specific antibody-secreting cells (ASC). All ISC represent only a small fraction of all circulating B cells, the fraction which has developed into the effector-cell stage, actively secretes immunoglobulins, and is known as plasmablasts. These ISC represent the sum of numerous ASC populations, each specific to different antigens encountered recently by the individual. In the present study, antigen-specific ASC are a population of plasmablasts producing specific antibodies against the specific vaccine antigen with which the volunteer was vaccinated 7 days earlier. Study design. A total of 73 healthy volunteers were enrolled in the study; 61 were females and 12 males, aged 21 to 48 years. None of the participants was IgA deficient. Fifty-five volunteers provided blood samples, and 16 provided both secretions and blood samples. Fifty-three volunteers with no history of the vaccination or the respective disease were vaccinated and assessed for vaccine antigen-specific responses in their secretions, sera, or whole blood (B cells). The vaccinations and sample collection were performed after written informed consent. The study protocol was approved by the Ethics Committees of the University of Alabama at Birmingham and of the Helsinki University Central Hospital. Sixteen women were enrolled for collection of external secretions (tears, parotid saliva, nasal wash fluid, intestinal lavage fluid, and cervicovaginal secretions), serum, and heparinized whole blood, all collected during one day. They all had regular menstrual cycles; two were on birth control pills. Twelve volunteers were a subpopulation of a previous study (20) who had been vaccinated orally (7) or rectally (5) with Salmonella enterica serovar Typhi Ty21a and had provided samples of external secretions (tears, parotid saliva, nasal wash fluid, and intestinal lavage fluid) after immunization. Samples for assessment of total concentrations of IgA1 and IgA2 were collected at a time point with no history of recent immunizations or recent infections, and for those providing samples for analysis of Salmonella serovar Typhi-specific IgA1 and IgA2 responses, both total and specific IgA1 and IgA2 levels were assessed 28 days after vaccination. Total and specific IgA1 and IgA2 were assessed with an enzyme-linked immunosorbent assay (ELISA). Peripheral blood mononuclear cells (PBMC) were isolated from heparinized peripheral blood samples (18) and assessed for two kinds of cell populations by enzyme-linked immunospot (ELISPOT) assay: ISC were enumerated in a total of 48 volunteers, and antigen-specific ASC in 53 vaccinees. Peripheral blood samples for assessment of total ISC were collected either at a time point with no history of recent immunizations or recent infections or, for analysis of vaccine antigen-specific ASC, on day 7 after vaccination. Both ISC and ASC of IgA, IgG, and IgM isotype-secreting cells were enumerated individually. They were also assessed for the frequency of different homing-associated (ISC and ASC) and maturational markers (ISC) on their surface. This was performed by separating PBMC into marker-positive and -negative populations and assaying the resulting cell populations for ISC and ASC by ELISPOT assay. Because of practical restrictions, all receptors could not be studied simultaneously on PBMC from all participants. Vaccinations. Twelve volunteers were given typhoid vaccine orally, 11 rectally, and seven parenterally. Ten volunteers were given the Haemophilus influenzae type b (Hib-D) conjugate vaccine parenterally, and 15 the unconjugated pneumococcal polysaccharide vaccine parenterally. The oral vaccine was the live, attenuated Salmonella serovar Typhi Ty21a vaccine (Vivotif; Crucell, Berne, Switzerland) administered as three doses two days apart perorally, according to the manufacturer s instructions, or rectally, as described earlier (20). The parenteral Ty21a vaccine was prepared by the Swiss Serum and Vaccine Institute (currently Crucell) according to a previously described protocol (19). The vaccine contained formalin-killed bacteria per dose. The Hib-D vaccine was a conjugate vaccine (HibTITER; Lederle-Praxis Biologicals, Inc., United States) with a polysaccharide component (polyribosyl phosphate [PRP]; 10 g) conjugated to diphtheria toxoid (25 g CRM 197 protein). Pneumovax is a 23- valent commercial pneumococcal polysaccharide vaccine (Pasteur Mérieux, Lyon, France) which contains, per dose, mg of capsular polysaccharide of each of the following serotypes: 1, 2, 3, 4, 5, 6B, 7F, 8, 9N, 9V, 10A, 12F, 14, 15B, 17F, 18C, 19A, 19F, 20, 22F, 23F, and 33F. All three parenteral vaccines were administered as an intramuscular injection of 0.5 ml in the left arm. Collection of external secretions and blood samples. The collection of all secretions has been described in detail earlier (20). Tears were collected with a capillary tube from the medial corner of the eye after stimulation with mist expressed from an orange peel. Nasal wash fluid was obtained by instillation of prewarmed saline in the nostrils and expelling after 30 s. Unstimulated parotid saliva was collected using a Schaefer cup placed over the parotid duct and centrifuged to remove debris. Intestinal secretions were obtained either with rectal wick (Polyfiltronics, Inc., Rockland, MA) or by the oral administration of Golytely (buffered polyethylene glycol solution; Braintree Laboratories, Braintree, MA) as 2 liters taken within 1 h. The effluent was collected and filtered through cheesecloth, protease inhibitors were added, and the fluid was clarified by centrifugation (20). The elution of proteins from the rectal wick samples was performed as described earlier (20). Cervicovaginal secretions were collected by instillation of saline into the vagina, rinsing, and aspiration with a plastic pipette, and samples were further centrifuged to remove particulate material. After the

3 VOL. 17, 2010 DIFFERENTIAL DISTRIBUTION OF IgA1 AND IgA2 SUBCLASSES 395 addition of protease inhibitors, samples were divided into multiple aliquots. Sera and the external secretions were frozen immediately after collection and kept at 70 C until assay. Assay of IgA1 and IgA2 concentrations in the body fluids. Samples were assayed for concentrations of total IgA1 and IgA2 by ELISA as described previously (17, 42). Briefly, ELISA plates (Nunc Maxisorp; Nunc, Roskilde, Denmark) were coated with human Ig-absorbed goat anti-mouse IgG (Southern Biotechnology Associates, Birmingham, AL), followed by overnight incubation with either anti-human IgA1 (Sigma) or anti-human IgA2 (Recognition Sciences Ltd., Birmingham, United Kingdom). Duplicate serial dilutions of the sample or standard were incubated overnight. Bound Igs were detected with peroxidaseconjugated goat anti-human IgA (Jackson ImmunoResearch Laboratories, Inc., West Grove, PA). As substrate, o-phenylenediamine-h 2 O 2 in citrate-phosphate buffer (ph 5.0) was used; development was stopped with 1 M sulfuric acid after 15 min, and absorbance was read at 490 nm in a V max microplate reader (Molecular Devices Corp., Menlo Park, CA). Assay of the concentrations of Salmonella serovar Typhi-specific IgA1 and IgA2 in the body fluids. The volunteers whose Salmonella serovar Typhi-specific IgA1 and IgA2 concentrations in body fluids were determined were a subgroup from another study (20). These samples were assayed for concentrations of total and Salmonella serovar Typhi-specific IgA1 and IgA2 by ELISA as described previously (42). Briefly, a whole-cell preparation of Salmonella serovar Typhi Ty21a was used for capture, the plates were blocked with 5% fetal bovine serum in phosphate-buffered saline (PBS)-Tween, and serial 2-fold dilutions of samples and standards were added. Human sera with known antibody titers against Salmonella serovar Typhi Ty21a and lipopolysaccharide (LPS) were included on each plate as positive control. Bound antibodies were detected with biotinlabeled mouse monoclonal IgG anti-human IgA1 or IgA2 (Nordic) followed by ExtrAvidin peroxidase conjugate (Sigma-Aldrich). The wells were developed with o-phenylenediamine-h 2 O 2 substrate (Sigma-Aldrich), the color reaction stopped with 1 M sulfuric acid, and the absorbance at 490 nm read. Isolation of mononuclear cells. The mononuclear cells were isolated from the heparinized venous blood by Ficoll-Hypaque density gradient centrifugation as previously described (18). Separation of the receptor-negative and -positive cell populations. The separation of the cells into receptor-negative and -positive populations has been described in detail previously (21, 24, 27). Briefly, cells were incubated with one of the following first-stage monoclonal antibodies: anti- 4 7 (ACT-1; Millennium Pharmaceuticals, Cambridge, MA), anti-l-selectin (anti-cd62l; Becton-Dickinson, San Jose, CA), anti- 4 (anti-cd49d; Immunotech, Marseille, France), anti- E 7 (anti-cd103 and anti-hml-1; Immunotech), anti- HLA-DR (Dako, Glostrup, Denmark), anti-cd28 (Becton-Dickinson), and anti-cla (HECA-452; a generous gift from Sirpa Jalkanen, originating from Eugene Butcher, Stanford, CA). After washing, the cells were incubated with Dynal M-450 magnetic beads coated with sheep anti-mouse IgG (Dynal, Oslo, Norway). Receptor-positive and -negative cells were separated with magnetic cell sorting. Both the receptor-positive and -negative cell populations were immediately analyzed with the ELISPOT assay for numbers of ISC or vaccine antigenspecific ASC. The efficiency of the cell separations was checked with flow cytometry in pilot experiments as described earlier (21). Assay of immunoglobulin-secreting cells and antigen-specific antibody-secreting cells (ELISPOT assay). The assays of IgA, IgG, and IgM ISC/ASC (18) and IgA1 and IgA2 ISC (25) have been described in detail previously. Briefly, microtiter plates (Maxisorp; Nunc) were coated with antisera to human IgA or IgM (Dako) or IgG (Sigma) for the ISC assay or with a vaccine antigen for the ASC assay (2 h at 37 C). For samples from those immunized with Salmonella serovar Typhi Ty21a (oral, rectal, or intramuscular administration), the coating antigen was a preparation of formalin-killed strain SL2404 (18); for immunization with Hib-D conjugate vaccine, the antigens were PRP (1.5 g/ml; a gift from Helena Käyhty, National Institute for Health and Welfare, Helsinki, Finland) and diphtheria toxoid (5.5 Lf/ml [Lf represents flocculation units]; a gift from Rose-Marie Ölander, National Institute for Health and Welfare, Helsinki, Finland); and for immunization with pneumococcal vaccines, the coating antigen was purified pneumococcal capsule polysaccharide 14 (2.5 g/ml; ATCC, Bethesda, Maryland). After coating, the plates were blocked with 1% bovine serum albumin (Sigma) in PBS, and the cells were incubated in the wells at four different dilutions ( to cells/ml) for 2 to 3 h. Antibodies secreted during this time were detected with alkaline phosphatase-conjugated anti-human IgA (Sigma-Aldrich), IgG (Sigma-Aldrich), and IgM (SouthernBiotech, Birmingham, England) antisera or, for the IgA subclass assays, with monoclonal antibodies to IgA1 and IgA2 (Nordic Immunological Laboratories, Tilburg, The Netherlands) (1:100 in 1% bovine serum albumin [BSA] PBS, overnight at 4 C) followed by alkaline phosphatase-conjugated anti-mouse IgG (Jackson ImmunoResearch Laboratories, West Grove, PA) diluted 1:500 in 1% BSA PBS for 2 h at 37 C. Substrate was added to all plates in hot agarose as described in detail earlier (18). Statistics. The paired t test was used to compare the concentrations of IgA1 and IgA2 in the same fluid samples. As the concentrations of Salmonella serovar Typhi-specific IgA1 and IgA2 were determined from body fluids only after a long storage, the results of these determinations are given only as proportions. The percentages of cells expressing the different receptors were determined as the arithmetic means of the percentages of ISC expressing the given cell surface marker. The proportions of the receptor-positive ISC were calculated as follows: % of receptor-positive cells among ISC (100 the number of ASC or ISC in receptor-positive population)/(the sum of the number of ASC or ISC in receptorpositive and receptor-negative populations). In order to get reliable statistics for the proportions of cells expressing a given marker, we set an inclusion limit of 20 ASC that needed to be identified among the cells studied. Statistical comparisons between IgA1 and IgA2 ISC/specific IgA1 and IgA2 ASC of the same volunteers were carried out using the paired Student s t test. The results of all statistical analyses were considered significant only when the P value was RESULTS Distribution of IgA subclasses in various body fluids. As shown in Fig. 1, IgA1 was found to be the predominant isotype in serum (P 0.001), tears (P 0.001), nasal wash fluid (P 0.001), and saliva (P 0.01). In the vaginal wash fluid, no significant difference was found between IgA1 and IgA2 concentrations. In the intestinal lavage fluid, IgA2 predominated (P 0.01). As seen by the results in Fig. 1, there was a marked variability in levels of IgA1 and IgA2 among individuals. Distribution of Salmonella serovar Typhi-specific IgA subclasses in various body fluids. The total and Salmonella serovar Typhi-specific IgA1 and IgA2 subclass distributions in body fluids collected from 12 volunteers were determined (20) (Table 1). Only in the intestinal lavage fluid was the total proportion of IgA2 higher than that of IgA1 (Table 1). Salmonella serovar Typhi-specific IgA1 and IgA2 were found in all secretions, consistent with our previous results for Salmonella serovar Typhi-specific total IgA (20). The concentration of Salmonella serovar Typhi-specific IgA1 exceeded that of Salmonella serovar Typhi-specific IgA2, consistent with the findings of IgA subclass distribution of Salmonella serovar Typhi-specific IgA1 and IgA2 ASC in the blood (see below). Numbers of total Ig, IgA1, and IgA2 ISC and vaccine antigen-specific IgA1 and IgA2 ASC. On average, 0.2% of PBMC were ISC (plasmablasts). An amount of 908 cells per 10 6 PBMC secreted IgA, which exceeded the numbers of IgMsecreting cells (Table 2); 73% of IgA ISC were found to secrete IgA1 and 27% IgA2. Before the immunizations, no vaccine antigen-specific ASC were found in the circulation of the vaccinees. Table 3 shows the numbers of IgA1- and IgA2-secreting cells specific for vaccine antigens in the blood on day 7 as a response to oral, rectal, or parenteral typhoid immunization or parenteral vaccination with the H. influenzae conjugate vaccine or pneumococcal polysaccharide vaccine. The expression of homing receptors on IgA1 and IgA2 ISC. The frequencies of cells expressing 4 7, L-selectin, and CLA among circulating IgA1 and IgA2 ISC are shown in Fig. 2. IgA2 ISC were found to express 4 7 significantly more frequently than IgA1 ISC, whereas no difference was found in the levels of expression of L-selectin or CLA between the cells of the two subclasses. The levels of expression of 4 -chain, 7 - chain, E 7 -integrin, HLA-DR, and CD28 on IgA1 and IgA2 ISC are shown in Fig. 3. The statistical analysis did not reveal

4 396 PAKKANEN ET AL. CLIN. VACCINE IMMUNOL. Downloaded from FIG. 1. Differential distribution of IgA1 and IgA2 in serum and external secretions of 16 unimmunized healthy adult volunteers. The data are given as values ( g/ml) for individual volunteers (n 10) in serum (A), tears (B), saliva (C), nasal wash fluid (D), vaginal wash fluid (E), and intestinal lavage fluid (F), and the geometric mean is indicated by a longer line. Note that serum, tears, and saliva were collected undiluted and these values represent actual concentrations, while nasal, vaginal and intestinal lavage fluids provide only the distribution of the two subclasses, as the collection of lavaged samples implies an undetermined dilution. Statistically significant differences between concentrations of IgA1 and IgA2 in paired Student s t test are indicated by asterisks ( ***, P 0.001; **, P 0.01). on August 23, 2018 by guest TABLE 1. Salmonella serovar Typhi-specific and total IgA1/IgA2 ratios in secretions 28 days after oral or rectal Salmonella serovar Typhi Ty21a vaccination of healthy volunteers Type of secretion Ratio (arithmetic mean SD) a of: Salmonella serovar Typhi-specific IgA1/IgA2 Total IgA1/IgA2 Saliva Nasal wash fluid Rectal-wick-sampled mucus Tears a Data for orally and rectally immunized volunteers (n 12) are pooled together. any significant differences between the two subclasses with respect to the expression of any of these markers. However, a significant difference between IgA1 and IgA2 ISC was seen when the proportion of E 7 integrin-expressing cells was subtracted from the proportion of 7-chain-expressing cells (Fig. 3). As 7 -chain is known to pair either with E -or 4-chain, this approach has been used previously to describe the expression of 4 7 ( 7- E 7 4 7) (24, 27) when the monoclonal antibody to 4 7 was not yet available. The expression of homing receptors on vaccine antigenspecific IgA1 and IgA2 ASC after oral or parenteral vaccination. The total populations of IgA1 and IgA2 ISC represent the sum of smaller, antigen-specific populations of IgA1 and IgA2 ASC. Since there was a significant difference in the frequency

5 VOL. 17, 2010 DIFFERENTIAL DISTRIBUTION OF IgA1 AND IgA2 SUBCLASSES 397 TABLE 2. Numbers of ISC isotypes in the peripheral blood of unimmunized healthy volunteers a ISC isotype No. of ISC/10 6 PBMC (geometric mean SEM) secreting indicated Ig IgG IgM IgA IgA1 b IgA IgA1 IgA a Data from 24 volunteers were pooled. b The percentage of ISC secreting IgA1 among all ISC was (mean standard deviation). of 4 7-expressing cells among the total populations of circulating IgA1 and IgA2 ISC (see above), it was of interest to examine the causes of this difference by focusing on small antigen-specific populations of ASC. The expression levels of HR were compared between antigen-specific IgA1 and IgA2 ASC in the peripheral blood of seven volunteers vaccinated orally with Salmonella serovar Typhi Ty21a and in five volunteers after parenteral pneumococcal polysaccharide vaccination. The HR data are only given for those volunteers whose ASC numbers reached the inclusion limit of 20 ASC detected (see Materials and Methods). Consistent with our earlier results (23), the frequencies of cells expressing the different HR differed between the two vaccination groups (Fig. 4): after oral Ty21a vaccination, all antigen-specific ASC expressed 4 7 and only a small amount expressed L-selectin, indicating an intestinal homing profile of these cells, while after pneumococcal vaccination, the frequency of 4 7 ASC was low and that of L-selectin ASC high, consistent with the systemic profile of HR expression. Notably, a comparison of HR expression on antigen-specific IgA1 and IgA2 ASC after an oral Ty21a or intramuscular pneumococcal vaccination revealed no differences between their frequencies on IgA1 and IgA2 ASC, indicating that both IgA1 and IgA2 cells antigen sensitized on a given site (mucosal versus systemic) have a similar homing potential. FIG. 2. Arithmetic means ( standard deviations) of the percentages of IgA1- and IgA2-secreting cells in the peripheral blood of healthy, unimmunized volunteers expressing the gut homing receptor ( 4 7), peripheral lymph node HR (L-selectin), and skin HR (CLA). At least PBMC from each volunteer were assayed. The numbers of volunteers from whom the data were pooled are given under the bars. Statistically significant differences between IgA1 and IgA2 ISC in paired Student s t test are indicated by asterisks ( **, P 0.01). DISCUSSION The present study confirms and extends the earlier notions of the differential distribution of IgA subclass antibodies in various body fluids (13). A significantly higher proportion of IgA1 than of IgA2 was found in serum, tears, nasal wash fluid, and saliva, while no difference was seen in the distribution in the vaginal wash fluid, and in the intestinal lavage fluid, IgA2 predominated. The proportions of IgA1 and IgA2 cells in a given mucosal tissue approximate the IgA1 to IgA2 ratio in a corresponding secretion (11, 30). Unfortunately, precise comparative studies of the proportions of IgA1 and IgA2 ISC and the proportions of IgA1 and IgA2 in the corresponding secretion have not been performed with the use of identical subclass-specific reagents. In the present study, the only mucosal site with actual IgA2 predominance was the intestine. This is consistent with previous reports on the distribution of IgA1 and IgA2 cells at different mucosal sites: the mucosa of the TABLE 3. Numbers of vaccine antigen-specific IgA1- and IgA2-secreting cells in the peripheral blood of healthy volunteers immunized with different vaccine regimens Vaccine regimen a Antigen specificity of ASC No. (geometric mean SEM) of ASC/ 10 6 PBMC secreting: IgA1 IgA2 IgA1 IgA2 % (arithmetic mean) of IgA-secreting cells that secreted IgA1 No. of volunteers P value (paired t test) for IgA1 vs IgA2 b Oral Ty21a Salmonella serovar Typhispecific cells Rectal Ty21a Parenteral Ty21a Parenteral Hib-D PRP-specific cells , NS Diphtheria toxoid-specific cells Parenteral Pnc Pnc polysaccharide type 14-specific cells NS a Vaccine regimens included Salmonella serovar Typhi Ty21a given orally, rectally, or parenterally and Haemophilus influenzae-diphtheria toxoid conjugate vaccine or unconjugated pneumococcal polysaccharide (Pnc) vaccine given parenterally. b NS, difference was not significant.

6 398 PAKKANEN ET AL. CLIN. VACCINE IMMUNOL. FIG. 3. Arithmetic means ( standard deviations) of the percentages of IgA1- and IgA2-secreting cells in peripheral blood expressing various lymphocyte markers. The numbers of unimmunized healthy volunteers from whom the data were pooled are given under the bars. Statistically significant differences between IgA1 and IgA2 ISC in paired Student s t test are indicated by an asterisk (, 0.01 P 0.05). colon has been shown to be the only mucosa with a pronounced dominance of IgA2 ISC (11, 29, 30). In previous studies, IgA1 and IgA2 cells were found in equal proportions in the female genital tract tissues (35) and the two subclasses were represented equally in the cervical mucus (34). The present data, with equal IgA1 and IgA2 concentrations in the vaginal wash fluid, confirm the data of those reports. The results of the present study concerning the distribution of IgA1 (60%) and IgA2 (40%) in tears are consistent with those of previous works by Delacroix et al. (13), Crago et al. (11), and Allansmith et al. (1) but differ from the results of Kett et al. (29). In the lacrimal glands, Crago et al. (11) found IgA1 and IgA2 cells in a ratio of 53:48 and Allansmith et al. (1) found them in a ratio of 56:44, consistent with the distribution of IgA1 and IgA2 in tears in the present study. In contrast, Kett et al. (30) found a more pronounced dominance of IgA1 cells (81%) in the lacrimal gland. The mechanisms underlying the selective distribution of IgA1 and IgA2 and, particularly, the preponderance of IgA2 in the large intestine are not fully understood. Tarkowski et al. (50) and Lue et al. (37) have shown that polysaccharide vaccines elicit a dominant IgA2 response both in serum and in ASC. These results are consistent with those of the present study: an IgA2-dominated ASC response to a polysaccharide antigen from pneumococci was found, while an IgA1-dominated response was elicited to a protein antigen, diphtheria toxoid. Interestingly, a clear IgA2 dominance was not found in the response to the PRP antigen of Haemophilus influenzae, which could be associated with the type of vaccine: this vaccine is a protein conjugate vaccine, where the polysaccharide antigen is linked to a protein carrier, which alters the responses to T-cell-dependent type (39). Responses to Ty21a were measured against a whole-cell preparation of bacteria. Even if the O-9,12 polysaccharide antigen is probably the major antigen against which the response is mounted, some protein antigens on the surface of the bacteria used for coating may be a target of minor responses as well (19) and shift the response toward IgA1, as suggested by the data in Table 3. Findings of the distribution of IgA subclass response as related to the type of antigen suggest that the IgA2 dominance in the large intestine could be a consequence of a significant local stimulation with predominantly polysaccharide antigens and LPS (5, 7, 38), which would lead to the expansion of cells of the IgA2 subclass at this site. Consistently, the large intestine, with the most intensive local stimulation by endotoxin, displays more IgA2 than IgA1 cells (11, 28, 30). In the present study, the only other mucosal site where IgA2 reached the concentration of IgA1 was the vaginal fluid; the female genital tract is colonized by a significant Gram-negative bacterial population. In other mucosal sites, with less significant supplies of LPS, IgA1 cells have been shown to be more numerous than IgA2 cells (11, 30) and, in the present study also, IgA1 predominated in those secretions. The high proportion of IgA2 in the lower digestive tract and the female genital tract may be of functional importance: IgA2 is resistant to IgA1 proteases produced by some pathogenic bacteria (31), and moreover, IgA2 is more efficient than IgA1 in inhibiting bacterial adherence to the mucosal epithelium (52). In order to search for the basis of differential distribution of IgA1 and IgA2 cells in the body, it appeared especially interesting to examine the HR expression on peripheral blood IgA1 and IgA2 plasmablasts before their potential homing to the systemic or mucosal tissues (4, 33, 48). It is probable that the differential distribution of IgA1 and IgA2 cells in various tissues would be a consequence of differential expression of HR and CCR on these cells in the circulation. Consistent with the highest proportions of IgA2 being found in intestinal lavage fluid, the intestinal HR, 4 7, was expressed more frequently on IgA2 than on IgA1 plasmablasts. The differential expression of HR on circulating IgA1 and IgA2 plasmablasts suggests that this is a key factor directing the cells of the two IgA subclasses to be differentially distributed in various tissues. In addition to ISC, we examined the expression of HR on antigen-specific ASC after vaccination through mucosal and systemic routes with T-cell-dependent and -independent vaccines. The results confirmed our previous findings (21) that the expression of HR depends on the administration route: all orally induced ASC exhibited an intestinal HR profile, with all cells expressing 4 7 and a smaller proportion L-selectin, but after parenteral antigen administration, a systemic profile was seen, with only a small proportion of 4 7 cells and a higher proportion of L-selectin cells. Importantly,

7 VOL. 17, 2010 DIFFERENTIAL DISTRIBUTION OF IgA1 AND IgA2 SUBCLASSES 399 Downloaded from FIG. 4. Arithmetic means ( standard deviations) of the percentages of antigen-specific IgA1- and IgA2-secreting cells expressing the indicated HR in healthy volunteers after oral vaccination with live attenuated Salmonella serovar Typhi Ty21a (A) or parenteral vaccination with an unconjugated pneumococcal polysaccharide vaccine (B). The numbers of volunteers from whom the data were pooled are given under the bars. There were no statistically significant differences between IgA1 and IgA2 ASC in the paired Student s t test. the present study was the first to analyze HR on antigenspecific IgA1 and IgA2 ASC: IgA1 and IgA2 cells activated at the same site had similar HR profiles, indicating that the site of antigen encounter determines the HR expression of the two cell types. It is of interest to consider why the frequency of 4 7- expressing cells was higher among IgA2 than IgA1 ISC. Because ISC in the blood represent the sum of numerous populations of ASC, each specific to some recently encountered antigen, and the intestine is known to be the only site of antigen encounter that results in 100% expression of 4 7, the high level of 4 7 expression on IgA2 ISC implies a high content of populations activated in the intestine. It seems reasonable to think that the high intestinal polysaccharide and LPS content would result in high numbers of IgA2-dominated responses and, thus, shift the 4 7 ISC among circulating IgA2 to a larger proportion than in IgA1 ISC. This also appears reasonable from the perspective that the majority of all plasma cells are known to be located at the intestinal sites (14, 40). In conclusion, the nature of the antigen at a given site directs the IgA subclass distribution of the immune response to these antigens, with polysaccharide antigens shifting the distribution toward IgA2 and proteins toward IgA1. The locally activated lymphocytes express certain sets of HR and CCR which enable these cells to home back to mucosal sites after their circulation through lymphatics and blood. Consistent with the high intestinal polysaccharide and LPS loads, the concentration of IgA2 is higher in the intestine and the proportion of gut HR-expressing cells more frequent among IgA2 plasmablasts than among IgA1 plasmablasts. Hence, together with the nature of the antigen, the site-specific homing of IgA1 and IgA2 plasmablasts is a key factor responsible for the differential distribution of IgA subclasses in the body fluids. on August 23, 2018 by guest

8 400 PAKKANEN ET AL. CLIN. VACCINE IMMUNOL. ACKNOWLEDGMENTS This study was supported by the Finnish Medical Association, the special Finnish governmental subsidy for health sciences research, and the Paulo Foundation. The principal funding recipient was A.K. It was also supported by NIH grants AI and AI This work was performed in the Department of Microbiology and Medicine, University of Alabama at Birmingham, in the Department of Bacteriology and Immunology, Haartman Institute, University of Helsinki, and in the Division of Infectious Diseases, Department of Medicine, Helsinki University Central Hospital, Helsinki, Finland. The authors do not have a financial or other association that might pose a conflict of interest. REFERENCES 1. Allansmith, M. R., J. Radl, J. J. Haaijman, and J. Mestecky Molecular forms of tear IgA and distribution of IgA subclasses in human lacrimal glands. J. Allergy Clin. Immunol. 76: Berlin, C., E. L. Berg, M. J. Briskin, D. P. Andrew, P. J. Kilshaw, B. Holzmann, I. L. Weissman, A. Hamann, and E. C. Butcher Alpha 4 beta 7 integrin mediates lymphocyte binding to the mucosal vascular addressin MAdCAM-1. Cell 74: Brandtzaeg, P Induction of secretory immunity and memory at mucosal surfaces. Vaccine 25: Brandtzaeg, P., and F. E. Johansen Mucosal B cells: phenotypic characteristics, transcriptional regulation, and homing properties. Immunol. Rev. 206: Brown, T. A., and J. Mestecky Immunoglobulin A subclass distribution of naturally occurring salivary antibodies to microbial antigens. Infect. Immun. 49: Camerini, D., S. P. James, I. Stamenkovic, and B. Seed Leu-8/TQ1 is the human equivalent of the mel-14 lymph node homing receptor. Nature 342: Cebra, J., H. Jiang, N. Boiko, and H. Tlasklova-Hogenova The role of mucosal microbiota in the development, maintenance, and pathologies of the mucosal immune system, p In J. Mestecky, J. Bienenstock, M. E. Lamm, L. Mayer, J. R. McGhee, and W. Strober (ed.), Mucosal immunology, 3rd ed. Elsevier Academic Press, San Diego, CA. 8. Cerutti, A The regulation of IgA class switching. Nat. Rev. Immunol. 8: Childers, N. K., C. Greenleaf, F. Li, A. P. Dasanayake, W. D. Powell, and S. M. Michalek Effect of age on immunoglobulin A subclass distribution in human parotid saliva. Oral. Microbiol. Immunol. 18: Conley, M. E., and D. L. Delacroix Intravascular and mucosal immunoglobulin A: two separate but related systems of immune defense? Ann. Intern. Med. 106: Crago, S. S., W. H. Kutteh, I. Moro, M. R. Allansmith, J. Radl, J. J. Haaijman, and J. Mestecky Distribution of IgA1-, IgA2-, and J chaincontaining cells in human tissues. J. Immunol. 132: Craig, S. W., and J. J. Cebra Peyer s patches: an enriched source of precursors for IgA-producing immunocytes in the rabbit. J. Exp. Med. 134: Delacroix, D. L., C. Dive, J. C. Rambaud, and J. P. Vaerman IgA subclasses in various secretions and in serum. Immunology 47: Fagarasan, S., and T. Honjo Intestinal IgA synthesis: regulation of front-line body defences. Nat. Rev. Immunol. 3: Gowans, J. L., and E. J. Knight The route of re-circulation of lymphocytes in the rat. Proc. R. Soc. Lond. B. Biol. Sci. 14: Griscelli, C., P. Vassalli, and R. T. McCluskey The distribution of large dividing lymph node cells in syngeneic recipient rats after intravenous injection. J. Exp. Med. 130: Hedges, S. R., M. S. Mayo, J. Mestecky, E. W. Hook III, and M. W. Russell Limited local and systemic antibody responses to Neisseria gonorrhoeae during uncomplicated genital infections. Infect. Immun. 67: Kantele, A Antibody-secreting cells in the evaluation of the immunogenicity of an oral vaccine. Vaccine 8: Kantele, A., H. Arvilommi, J. M. Kantele, L. Rintala, and P. H. Mäkelä Comparison of the human immune response to live oral, killed oral or killed parenteral Salmonella typhi TY21A vaccines. Microb. Pathog. 10: Kantele, A., M. Häkkinen, Z. Moldoveanu, A. Lu, E. Savilahti, R. D. Alvarez, S. Michalek, and J. Mestecky Differences in immune responses induced by oral and rectal immunizations with Salmonella typhi Ty21a: evidence for compartmentalization within the common mucosal immune system in humans. Infect. Immun. 66: Kantele, A., J. M. Kantele, E. Savilahti, M. Westerholm, H. Arvilommi, A. Lazarovits, E. C. Butcher, and P. H. Mäkelä Homing potentials of circulating lymphocytes in humans depend on the site of activation: oral, but not parenteral, typhoid vaccination induces circulating antibody-secreting cells that all bear homing receptors directing them to the gut. J. Immunol. 158: Kantele, A., E. Savilahti, H. Tiimonen, K. Iikkanen, S. Autio, and J. M. Kantele Cutaneous lymphocyte antigen expression on human effector B cells depends on the site and on the nature of antigen encounter. Eur. J. Immunol. 33: Kantele, A., M. Westerholm, J. M. Kantele, P. H. Mäkelä, and E. Savilahti Homing potentials of circulating antibody-secreting cells after administration of oral or parenteral protein or polysaccharide vaccine in humans. Vaccine 17: Kantele, J. M., A. Kantele, and H. Arvilommi Circulating immunoglobulin-secreting cells are heterogeneous in their expression of maturation markers and homing receptors. Clin. Exp. Immunol. 104: Kantele, J. M., E. Savilahti, M. Westerholm-Ormio, S. Pakkanen, H. S. Arvilommi, T. Reunala, and A. M. Kantele Decreased numbers of circulating plasmablasts and differences in IgA1 plasmablast homing to skin in coeliac disease and dermatitis herpetiformis. Clin. Exp. Immunol. 156: Kantele, A. M., N. V. Palkola, H. S. Arvilommi, and J. M. Kantele Distinctive homing profile of pathogen-specific activated lymphocytes in human urinary tract infection. Clin. Immunol. 128: Kantele, J. M., H. Arvilommi, S. Kontiainen, M. Salmi, S. Jalkanen, E. Savilahti, M. Westerholm, and A. Kantele Mucosally activated circulating human B cells in diarrhea express homing receptors directing them back to the gut. Gastroenterology 110: Kett, K., K. Baklien, A. Bakken, J. G. Kral, O. Fausa, and P. Brandtzaeg Intestinal B-cell isotype response in relation to local bacterial load: evidence for immunoglobulin A subclass adaptation. Gastroenterology 109: Kett, K., P. Brandtzaeg, and O. Fausa J-chain expression is more prominent in immunoglobulin A2 than in immunoglobulin A1 colonic immunocytes and is decreased in both subclasses associated with inflammatory bowel disease. Gastroenterology 94: Kett, K., P. Brandtzaeg, J. Radl, and J. J. Haaijman Different subclass distribution of IgA-producing cells in human lymphoid organs and various secretory tissues. J. Immunol. 136: Kilian, M., and M. W. Russell Microbial evasion of IgA functions, p In J. Mestecky, J. Bienenstock, M. E. Lamm, L. Mayer, J. R. McGhee, and W. Strober (ed.), Mucosal immunology, 3rd ed. Elsevier Academic Press, San Diego, CA. 32. Kishimoto, T. K., M. A. Jutila, and E. C. Butcher Identification of a human peripheral lymph node homing receptor: a rapidly down-regulated adhesion molecule. Proc. Natl. Acad. Sci. U. S. A. 87: Kunkel, E. J., and E. C. Butcher Plasma-cell homing. Nat. Rev. Immunol. 3: Kutteh, W. H., S. J. Prince, K. R. Hammond, C. C. Kutteh, and J. Mestecky Variations in immunoglobulins and IgA subclasses of human uterine cervical secretions around the time of ovulation. Clin. Exp. Immunol. 104: Kutteh, W. H., K. D. Hatch, R. E. Blackwell, and J. Mestecky Secretory immune system of the female reproductive tract. I. Immunoglobulin and secretory component-containing cells. Obstet. Gynecol. 71: Ladjeva, I., J. H. Peterman, and J. Mestecky IgA subclasses of human colostral antibodies specific for microbial and food antigens. Clin. Exp. Immunol. 78: Lue, C., A. Tarkowski, and J. Mestecky Systemic immunization with pneumococcal polysaccharide vaccine induces a predominant IgA2 response of peripheral blood lymphocytes and increases of both serum and secretory anti-pneumococcal antibodies. J. Immunol. 140: Macpherson, A. J., and E. Slack The functional interactions of commensal bacteria with intestinal secretory IgA. Curr. Opin. Gastroenterol. 23: McVernon, J., N. A. Mitchison, and E. R. Moxon T helper cells and efficacy of Haemophilus influenzae type b conjugate vaccination. Lancet Infect. Dis. 4: Mestecky, J., I. Moro, M. Kerr, and J. Woof Mucosal immunoglobulins, p In J. Mestecky, J. Bienenstock, M. E. Lamm, L. Mayer, J. R. McGhee, and W. Strober (ed.), Mucosal immunology, 3rd ed. Elsevier Academic Press, San Diego, CA. 41. Mestecky, J., and M. W. Russell IgA subclasses. Monogr. Allergy 19: Moldoveanu, Z., W. Q. Huang, R. Kulhavy, M. S. Pate, and J. Mestecky Human male genital tract secretions: both mucosal and systemic immune compartments contribute to the humoral immunity. J. Immunol. 175: Moro, J. R., M. Iwata, and U. H. von Andrian Vitamin effects on the immune system: vitamins A and D take centre stage. Nat. Rev. Immunol. 8: Picker, L. J., T. K. Kishimoto, C. W. Smith, R. A. Warnock, and E. C. Butcher ELAM-1 is an adhesion molecule for skin-homing T cells. Nature 349: Quiding-Järbrink, M., M. Lakew, I. Nordstrom, J. Banchereau, E. Butcher, J. Holmgren, and C. Czerkinsky Human circulating specific antibodyforming cells after systemic and mucosal immunizations: differential homing

9 VOL. 17, 2010 DIFFERENTIAL DISTRIBUTION OF IgA1 AND IgA2 SUBCLASSES 401 commitments and cell surface differentiation markers. Eur. J. Immunol. 25: Renegar, K. B., and P. A. J. Small Passive transfer of local immunity to influenza virus infection by IgA antibody. J. Immunol. 146: Russell, M. W., and M. Kilian Biological activities of IgA, p In J. Mestecky, J. Bienenstock, M. E. Lamm, L. Mayer, J. R. McGhee, and W. Strober (ed.), Mucosal immunology, 3rd ed. Elsevier Academic Press, San Diego, CA. 48. Salmi, M., and S. Jalkanen Lymphocyte homing to the gut: attraction, adhesion, and commitment. Immunol. Rev. 206: Tamura, S., H. Funato, Y. Hirabayashi, K. Kikuta, Y. Suzuki, T. Nagamine, C. Aizawa, M. Nakagawa, and T. Kurata Functional role of respiratory tract haemagglutinin-specific IgA antibodies in protection against influenza. Vaccine 8: Tarkowski, A., C. Lue, Z. Moldoveanu, H. Kiyono, J. R. McGhee, and J. Mestecky Immunization of humans with polysaccharide vaccines induces systemic, predominantly polymeric IgA2-subclass antibody responses. J. Immunol. 144: Winner, L., III, J. Mack, R. Weltzin, J. J. Mekalanos, J. P. Kraehenbuhl, and M. R. Neutra New model for analysis of mucosal immunity: intestinal secretion of specific monoclonal immunoglobulin A from hybridoma tumors protects against Vibrio cholerae infection. Infect. Immun. 59: Wold, A. E., J. Mestecky, M. Tomana, A. Kobata, H. Ohbayashi, T. Endo, and C. Svanborg Eden Secretory immunoglobulin A carries oligosaccharide receptors for Escherichia coli type 1 fimbrial lectin. Infect. Immun. 58: Woof, J. M., and J. Mestecky Mucosal immunoglobulins. Immunol. Rev. 206: Xu, B., N. Wagner, L. N. Pham, V. Magno, Z. Shan, E. C. Butcher, and S. A. Michie Lymphocyte homing to bronchus-associated lymphoid tissue (BALT) is mediated by L-selectin/PNAd, alpha4beta1 integrin/vcam-1, and LFA-1 adhesion pathways. J. Exp. Med. 197: