Conceived and designed the experiments: VR MC. Performed the experiments: VR AC MN EL DT. Analyzed the data: VR KU FV. Contributed reagents/materials/analysis tools: GD VT KU FV. Wrote the paper: VR FV MC.
The authors have declared that no competing interests exist.
Mucosae-associated epithelial chemokine (MEC or CCL28) binds to CCR3 and CCR10 and recruits IgA-secreting plasma cells (IgA-ASCs) in the mucosal
The development of prophylactic and therapeutic vaccines against mucosal infections still represents a challenge
We recently reported a positive correlation between titers of mucosal anti-HIV-1 IgA and the CCL28–CCR3/CCR10 system both in HIV-1 infected and HIV-1-exposed but sero-negative (HESN) individuals
Since mucosal vaccines aim at inducing antigen-specific secretory IgA antibodies in addition to effective cell-mediated responses, the use of chemokines as adjuvants could significantly improve vaccine immunogenicity and efficacy
In the present study, we further investigate the immunomodulatory effects of MEC/CCL28 in a mouse model using a prime-boost strategy based on HIV-1IIIB-VLPs in the presence/absence of the murine chemokine gene inserted into a CpG-free expression vector. Results confirm the pivotal role of CCL28 in the modulation of mucosal immunity and suggest that CCL28 could be useful in the design of mucosal vaccines finalized at the prevention of HIV infection.
Inbred female Balb/c mice were immunized with a prime-boost regime based on HIV-1IIIB VLP in the presence or in the absence of CCL28-expressing vector. The CCL19-expressing plasmid was used as a negative control. This chemokine was chosen because, similarly to CCL28, it plays a crucial role in lymphoid cell trafficking but nevertheless it does not have an effect on IgA+ plasma cells; additionally, CCL19 does not bind CCR3 or CCR10 but rather uses CCR7. To verify if pCCL28 and pCCL19 were effective expression vectors and whether their use resulted in systemic increases of the corresponding chemokines, CCL28 an CCL19 were measured in immune sera. Results showed a significant increase in serum CCL28 levels of HIV-VLPIIIB-CCL28 and CCL28 alone mice compared to HIV-VLPIIIB-CCL19, HIV-VLPIIIB, CCL19 alone and saline mice (p<0.01 in all cases). Similar results were observed when serum CCL19 levels from both HIV-VLPIIIB-CCL19 and CCL19 alone mice were compared to those obtained in HIV-VLPIIIB-CCL28, HIV-VLPIIIB, CCL28 alone or saline animals (p<0.03 in all cases) (
Panels
To analyze the expression of chemokine receptors on circulating lymphocytes, CCR3- and CCR10-expressing CD3+, CD14+ and CD19+ splenocytes were analyzed in all mice. Results showed that CD19+/CCR3+ and CD19+/CCR10+ splenocytes were significantly augmented in HIV-VLPIIIB-CCL28 treated mice as compared to HIV-VLPIIIB-CCL19 (p = 0.014 and 0.043, respectively), HIV-VLPIIIB alone (p = 0.005 and 0.006, respectively), CCL28 alone (p = 0.005 and 0.008, respectively), CCL19 alone (p = 0.002 and 0.005, respectively), and saline control mice (p = 0.001 and 0.004, respectively) (
Percentage of CCR3-expressing CD19+ splenocytes (
Evaluation of mean fluorescence intensity (MFI) showed that the CCR3 MFI on CD19+ splenocytes was significantly augmented in HIV-VLPIIIB-CCL28 treated mice compared to HIV-VLPIIIB-CCL19 (p = 0.029), HIV-VLPIIIB alone (p = 0.015), CCL28 alone (p = 0.012), CCL19 alone (p = 0.009), or saline control mice (p = 0.005) (
Splenic and colonic cells were restimulated with recombinant HIV-1IIIB/Env-gp120 to quantify Th1-type (IFN-γ) and Th2-type (IL-4 and IL-5) cytokines production. Results showed a significant increase of IFN-γ production by both spleen and PP/colonic cells from HIV-VLPIIIB-CCL28-treated mice as compared to mice receiving either HIV-VLPIIIB-CCL19 (spleen p = 0.034; colon p = 0.041) or VLP alone (spleen p = 0.026; colon p = 0.033) (
HIV-specific cytokine production in HIV-1-VLPIIIB-CCL28, HIV-1-VLPIIIB-CCL19, HIV-1-VLPIIIB, CCL28 alone, CCL19 alone and control (saline)-treated mice is shown. Panels
Serum was collected two weeks after the final immunization to measure Env-specific antibodies; ELISA were performed on microwell plates coated with HIV-1IIIB Env gp120. Results showed that serum Env-specific IgG and IgA were increased in HIV-VLPIIIB-CCL28 mice compared to HIV-VLPIIIB-CCL19 (p = 0.05 and 0.046) or HIV-VLPIIIB mice (p = 0.042 and 0.04) (
The anti-HIV-1 antibody production in mouse sera is shown at days 0 (T0), 14 (T1 and 28 (T2) from each of the following immunized mice groups: HIV-1-VLPIIIB-CCL28, HIV-1-VLPIIIB-CCL19, HIV-1-VLPIIIB, CCL28 alone, CCL19 alone and control (saline)-treated mice. Levels of anti-HIV-1/Env-specific IgG (
To assess CCL28-induced migration of Ig-ASCs at mucosal sites, both total and Env-specific antibody responses were quantified in vaginal secretions of immunized and control mice. Total IgA levels were significantly augmented in vaginal secretions of HIV-VLPIIIB-CCL28-treated mice compared to all other animals (vs. VLPIIIB-CCL19 p = 0.04, vs. VLPIIIB alone p = 0.008, vs. CCL28 alone, CCL19 alone, and saline p<0.001,) (
Mucosal anti-HIV-1 antibody production at days 0 (T0), 14 (T1) and 28 (T2) is represented for the following group of immunized mice: HIV-1-VLPIIIB-CCL28, HIV-1-VLPIIIB-CCL19, HIV-1-VLPIIIB, CCL28 alone, CCL19 alone and control (saline)-treated mice. Total (
To verify whether the use of CCL28 would result in an enhanced neutralizing activity of Env-specific antibodies, neutralization experiments using pooled sera against HIV-1IIIB and HIV-1DU174 were performed. Two different virus strains were used: HIV-1IIIB, a subtype B CXCR4-tropic strain, and HIV-1DU174, a subtype C CCR5-tropic strain. A marginal (<15%) neutralizing activity was detected in the pre-immune and immune sera of mice treated with either CCL28 alone, CCL19 alone, or saline (data not shown). Immune sera from HIV-VLPIIIB-CCL28 treated mice showed a neutralization activity titer of 220 and 160 (50% neutralization activity) against, respectively, HIV-1IIIB at a TCID50 of 40 and HIV-1DU174 at a TCID50 of 20 (
Experiments were run on pooled sera from HIV-1-VLPIIIB-CCL28, HIV-1-VLPIIIB-CCL19, HIV-1-VLPIIIB, or control (saline)-treated mice. Upper panels show
Immune vaginal secretions from HIV-VLPIIIB-CCL28-receiving mice showed a neutralization titer of 80 and 40 against, respectively, HIV-1IIIB and HIV-1DU174 (
The neutralization observed against both a clade B and a clade C virus with vaginal secretions is apparently puzzling considering that antibody concentration in vaginal secretions is lower than in serum. A technical issue with VLP-based vaccines is their tendency to elicit anti-cell antibodies against non-Env membrane proteins that could interfere with the neutralization assays
Furthermore, to verify whether the results could be a consequence of changes in cell viability, HIV-1IIIB- and HIV-1DU174-infected PBMCs that were either untreated or mixed with matched serum and vaginal secretions were stained with 7-AAD. As shown in
Experiments were run on pooled sera of HIV-1-VLPIIIB-CCL28, HIV-VLPIIIB-CCL19 and HIV-VLPIIIB-treated mice. Upper panels show the percentage of cell death in HIV-1IIIB-infected PBMCs mixed with 2-fold serial dilutions of murine sera (
Finally, in order to determine the relative contribution of the main antibody isotypes to the neutralizing activity of vaginal secretions, IgA, IgG or IgA + IgG immune absorption experiments were performed. Results clearly showed that IgA antibodies from VLP-immunized mice played a major role in the neutralization of both HIV-1IIIB and HIV-1DU174 and that IgA-mediated neutralization activity was enhanced in a CCL28-dependent manner. In the absence of CCL28, but in the presence of VLP alone or in combination with CCL19, the neutralization level was lower and mostly IgG-dependent. Nevertheless, this was confirmed by the observation that, along with depletion of both IgA and IgG antibodies, the neutralization activity was almost abolished (
Experiments were run on pooled IgG- or IgA- or IgG/IgA-depleted vaginal secretions from HIV-1-VLPIIIB-CCL28, HIV-1-VLPIIIB-CCL19, HIV-1-VLPIIIB mice. Upper panel show
Since the GI MLP is the principal site of HIV replication during primary infection the induction of virus-specific- mucosal immunity would be beneficial as it could obstacle the establishment of infection. As the results herein demonstrated that the up-regulation of CCL28-CCR3/CCR10 circuit is correlated with increased concentrations of both systemic and mucosal HIV-specific IgA, we analyzed IgA-ASC distribution in the GI mucosal
IgA+ plasma cell distribution at the mucosal level in BALB/c mice immunized with HIV-1-VLPIIIB-CCL28 (
The vast majority of newly acquired HIV infections are sexually transmitted. In this modality of infection the virus initially targets the vaginal or rectal mucosa; preventative vaccines or microbicides should thus be designed to protect such mucosae
Results presented herein demonstrate that immunization of mice with Env-expressing VLPs in the presence of CCL28 results in the modulation of the whole CCL28-CCR3/CCR10 circuit and optimizes immunization-induced HIV-specific immune response. Thus, cytokine secretion by spleen and colon cells, as well as total and HIV-specific IgA and IgA+ plasma cells were significantly increased in the presence of CCL28. Notably, HIV-specific IgA and IgG titers were significantly increased in serum as well, suggesting that, systemic as well as mucosal immune responses are efficiently modulated by CCL28. Finally, the observation that the neutralization ability of serum and vaginal washes of immunized mice in the presence of CCL28 was increased indicates that the immune modulation induced by this chemokine is associated with an augmented capacity to down-modulate HIV infectivity.
This study shows that the objective of up-regulating potentially beneficial mucosal immune responses is achievable, at least in the mouse model, by CCL28. Our results demonstrate that CCL28 indeed results in the recruitment of IgA-ASCs in mucosal epithelial tissues and this effect is associated with an increased neutralization activity. The increased neutralization ability toward at least two different HIV clades exhibiting the two major X4 and R5 tropisms, detected in immune sera and vaginal secretions, indicates that the reduction of HIV infectivity
The tissue-targeted lymphocyte migration is strongly regulated by a complex network of chemokines
IgA are mostly mucosal antibodies involved in the first line of defense of adaptive immunity facing pathogens. The anti-HIV-1 mucosal IgA are not only observed in HIV-1 infected individuals
The gut-associated lymphoid tissue (GALT) contains the majority of T-cells in the body; recent data showed that GALT is the preferential target for HIV replication during the acute phase of HIV infection
In this study we report that intramuscular administration of HIV-1IIIB-VLPs in the presence of the CCL28-expressing plasmid correlates with a robust up-regulation of chemokine receptor expression on circulating B-cells and an increase in both systemic and mucosal immunity. This could apparently be surprising, as parenteral immunization is generally thought not to induce significant immune responses at mucosal surfaces. However, recent studies have revealed the possibility of cross-talk between systemic compartments and mucosal tissues
As far as the CCL28-CCR3/CCR10 circuit is concerned, the percentage of CD19+/CCR3+ cells was unexpectedly high as compared with previous results observed in humans
It is noteworthy that results showed significant differences in the plasmid-induced serum levels between CCL28 and CCL19
In conclusion, we show that CCL28 used as an adjuvant has a robust immunomodulatory effect on potentially beneficial mucosal and systemic immune responses. Notably, the beneficial immunomodulatory effects of CCL28 is not likely limited to HIV infection as this chemokine was recently shown to be effective in eliciting long-lived antibody responses that neutralized influenza A/PR8/34 and protected mice from morbidity and mortality associated with a lethal intranasal viral challenge
HIV-1 virus like particles (VLP) were produced by transient transfection of HEK/293T-cells (human embryo kidney carcinoma cell-line, T clone) (Sigma-Aldrich, Germany) with the plasmids pCD-Hgpsyn and pConBgp160-opt. pCD-Hgpsyn containing the codon-optimized gag-pol sequence of HIV-1IIIB kindly provided by Geneart (Regensburg, Germany). pConBgp160-opt encodes a consensus full-length HIV-1/subtype B (HIV-1/B) envelope obtained from the AIDS Research and Reference Reagent Program, Division of AIDS, NIH, (Dr. Beatrice Hahn). Confluence of 80–90%, HEK/293T cells were transfected with 40 µg of each plasmid in DMEM medium supplemented with PEI [10 mg/ml], 5% FCS and penicillin/streptomycin. The transfection medium was changed 18 hr after transfection with DMEM medium supplemented with 2% FCS and penicillin/streptomycin and harvested 48 hr later. The harvested medium was centrifuged at 300×g for 10 min and filtered through a 0.45 µm filter to remove cell debris. VLP were further purified and concentrated from the conditioned medium by ultracentrifugation through a 20% sucrose gradient and a subsequent ultracentrifugation at 28,000 rpm for 2 hr a SW28 rotor (Beckman, CA). Supernatants were discarded and the pellets containing viral particles were resuspended in PBS. The endotoxin level was measured by QCL-1000® Chromogenic LAL Endpoint Assay (Cambrex, Germany) according to the manufacturer's instructions. The concentration of viral proteins in the final VLP preparation was determined by an in-house ELISA and corresponded to 1.5 µg/ml of HIV-1 Env and 0.55 µg/ml of HIV-1 Gag. The calibration curve was set up by using recombinant HIV-1IIIB Env gp120 (Baculovirus) (EVA607 supplied by the NIBSC centralized facility for AIDS Reagents supported by EVA EU Program) and Gagpr55 (AIDS Research and Reference Reagent Program, Division of AIDS, NIAID, NIH).
The MEC/CCL28 murine chemokine gene was digested with AgeI from the pORF-mCCL28 plasmid (InvivoGen, San Diego, CA) and a blunt end was produced with T4-DNA polymerase. Following digestion with NheI, the resulting 410 bp fragment was inserted into a pCpG expression vector, previously digested with ScaI/NheI. The CCL19/MIP-3β murine gene from the pORF5-mMIP3β inserted into a pCpG expression vector (InvivoGen, San Diego, CA) as described above was used as a negative control. The endotoxin level was measured by QCL-1000® Chromogenic LAL Endpoint Assay (Cambrex, Germany) according to the manufacturer's instructions.
The study was approved by the Italian Ministry of Health (Permit Number: 02/010) and animals were managed according to the principles of the "Guide for the Care and Use of Laboratory Animals" and in accordance with the Italian national law (Legislative Decree. 116/1992) and the recommendations of the European Community (86/609/CEE) for the care and use of laboratory animals. Adult inbred female Balb/c mice, 6–8 weeks old, were purchased from Charles River Laboratories (Calco, Italy). Mouse colonies were maintained on a 12-h light-dark cycle in cages of 5 animals with water and food provided
Fifty microliters of blood were collected by tail vein bleeding and added to 100 µl of phosphate-buffered saline (PBS, Organon Teknika Corp.) containing 100 IU of heparin (Hip-PBS, Sigma). Serum was obtained after centrifugation at 800 g for 10 min and stored at -80°C. Vaginal secretions were collected with 50 µl of pre-warmed PBS supplemented with 100X-concentrated protease inhibitor cocktail (1%, v/v) (O-complete, Roche Applied Science, Germany) by repeated aspiration until a discrete quantity of mucus was recovered. Material was collected and centrifuged at 12,000 g for 10 min, the supernatant was transferred to sterile microcentrifuge tubes and frozen at −20°C until use. The endotoxin level was measured by QCL-1000® Chromogenic LAL Endpoint Assay (Cambrex, Germany) in both sera and vaginal secretions according to the manufacturer's instructions, resulting in a an endotoxin concentration of 0.8 ng/ml and 0.4 ng/ml, respectively.
Mice were anesthetized for 2 min with ether gas and sacrificed by neck dislocation. Spleens were excised under sterile conditions in a laminar flow hood and put through a 100 µm plastic strainer (BD Falcon 2350, BD Biosciences, Bedford, MA) for cell recovery. Splenocytes were layered on a continuous 40–100% Percoll gradient (Sigma) and washed twice in PBS to obtain lymphocyte-rich cells. Cell viability was determined using Trypan blue staining. Splenocytes were resuspended in cell culture medium (RPMI 1640) (Organon Teknika Corp.) and used in cell culture assays. Peyer's patches (PP)/colonic mononuclear cells were recovered from freshly obtained specimens. Colon specimens were first washed in HBSS-with Phenol Red (BioWhittaker Inc., Walkersville, MD) cut into 0.5-cm pieces. They were then incubated twice, each time for 15 min in 0.37 mg/ml EDTA-HBSS (Sigma) and 0.145 mg/ml dithiothreitol (Thermo Fisher Scientific, Rockford, IL) at 37°C. Tissues were further digested at 37°C for 10 min with 400 U/ml of Collagenase D (Roche) and 0.01 mg/ml of DNase I (Sigma). Tissue-released cells were suspended in a continuous 100%–40% Percoll gradient to obtain the lymphocyte-enriched population. Mice recta were also analyzed by immune-histochemistry.
Cell count was performed with the automated cell counter ADAM-MC (Digital Bio, NanoEnTek Inc, Korea). ADAM-MC automatic cell counter measures total cell numbers and cell viabilities by cutting-edge detection technologies. In addition to Trypan blue staining, ADAM-MC procedure was carried out using two sensitive fluorescence dye staining solutions, AccuStain Solution T (Propidium Iodide/lysis solution) and AccuStain Solution N (Propidium IodideI/PBS). AccuStain Solution T permeabilizes plasma membrane stain nucleus that allow measurements of total cell enumeration, while AccuStain Solution N exclusively stains non-viable cells. A 532 nm optic laser is automatically focused onto the cell suspension contained into a disposable microchip where cell analysis is made with a CDD camera.
CCR3 or CCR10 receptor expression was evaluated on splenocytes. Cells were resuspended in PBS and their surface stained with mAbs CD3e PE- Cy5 (Armenian Hamster IgG isotype, eBioscience, San Diego, CA), anti-mouse CD14 PE-C-conjugated Cy5 (rat IgG2a isotype, eBioscience), PE-Cy5-conjugated anti-mouse CD19 (rat IgG2a isotype, eBioscience), a rat FITC-conjugated anti-mouse CCR3 IgG2a isotype (R&D Systems, Minneapolis, MN), and a rat anti-mouse CCR10 PE IgG2b isotype (R&D Systems). Following incubation 15 min at room temperature in the dark, cells were washed 3 times in PBS and fixed in 1% formaldehyde. All the cytometric analyses were performed using an FC500 flow cytometer (Beckman-Coulter, Miami, FL) equipped with a double 15-mW argon ion laser operating at 456 and 488 interfaced with an Intercorp (Venice, Italy) computer. For each analysis 20,000 events were acquired and gated on CD3 or CD14 or CD19 expression and side scatter properties. Green fluorescence from FITC (FL1) was collected through a 525-nm band-pass filter, orangered fluorescence from R-PE (FL2) was collected through a 575-nm bandpass filter and red fluorescence from Cy5PE (FL4) was collected through a 670-nm band-pass filter. Data were collected using linear amplifiers for forward and side scatter and logarithmic amplifiers for FL1, FL2, FL4, and FL5.
IFN-γ, IL-4 and IL-5 production were measured in the supernatants from splenocytes as well as in PP/colonic T cells after
Anti-HIV-IgG and IgA were measured in sera and vaginal secretions by an ELISA method based on a recombinant HIV-1 envelope protein. 96 well ELISA plates were coated overnight at 4°C with 100 µl of 4 µg/ml recombinant HIV-1IIIB Env gp120 (Baculo) (NIBSC). Plates were washed three times with PBS-Tween-20 0.05% buffer, and incubated for 2 hr at 37°C, 5% CO2 with PBS containing 3% of Bovine Serum Albumin (BSA) (Sigma, St Louis, MO) to block non-specific protein binding sites. Serum dilutions 1/10 to 1/10000 and vaginal wash dilutions 1/2 to 1/1000 were incubated for 2 hr at 37°C, 5% CO2. Plates were then washed three times and a goat HRP-conjugated anti- α-chain IgA antibody mouse (Sigma) or a goat HRP-conjugated anti-mouse IgG antibody (Jackson Immuno-Research, West Grove, PA) diluted 1∶1000 or 1∶30000 in PBS/BSA respectively was added to the plates. Following 1 h of incubation at 37°C, 5% CO2, the plates were washed and incubated with tetramethylbenzidine (TMB, R&D Systems) substrate solution for 30 min at RT. The addition H2SO4 1.8 M stopped the colour reaction. IgG or S-IgA concentrations were measured at an Optical density absorbance λ = OD490. Total IgA from vaginal secretions were measured by ELISA as described above, excepting the coating that was carried out at 4°C overnight with 100 µl of a goat anti-mouse IgA antibody (Kamiya Biomedical Company, Seattle, WA). Sample concentrations were determined from standard curves, using purified Ig standard mouse IgA and IgG (Sigma) assayed in parallel; values were expressed in nanograms per milliliter.
Streptococcal protein G and Staphilococcal protein A were used to specifically remove IgG and/or IgA from mucosal samples. Mouse vaginal secretions were diluted 1∶2 with RPMI 1640 medium (Organon Teknika Corp.) supplemented with 1% penicillin/streptomycin and incubated with an equal volume of Sepharose-immobilized recombinant protein G (PGS) and/or Sepharose-immobilized recombinant protein A (PAS) (Sigma Chemical Co., St. Louis, MO). The incubation was carried out for 2 h at 37°C with constant gentle rotation to prevent sedimentation of the PGS and/or the PAS. After centrifugation for 2 min at 1000 g, the supernatant was withdrawn and incubated for 12 h on the rotator at 4°C with a second aliquot of PGS and/or PAS equal to the first. The specimens were then returned to the 37°C rotating incubator for an additional 2 h. After centrifugation at 1000g for 2 min, the 1∶2 diluted vaginal washes, either IgG-depleted or IgA-depleted or IgG/IgA-depleted, were harvested.
HIV-1 neutralization assay was performed in a pooled donor PBMC-based assay. Briefly, pooled sera from pre-immunized and immunized mice were complement-depleted by heat inactivation at 56°C for 60 min and serially diluted at 2-fold dilutions, starting from 1/20. Pooled vaginal secretions were serially diluted at 2-fold dilutions, starting from 1/10. Vaginal secretions were assayed before and after Ig depletion. Sera and vaginal secretions were incubated in a humidified 5% CO2 incubator at 37°C for 60 min in duplicate in 96-well plates with the laboratory strain HIV-1IIIB and the primary isolate HIV-1DU174 (NIBSC) of 40 and 20 TCID50, respectively. One hundred thousand PBMCs pooled from HIV-negative donors were stimulated with PHA for 48 hr, added to each well, and incubated overnight (18 hr) at 37°C. Cells were extensively washed with RPMI 1640 (Organon Teknika Corp.), centrifuged at 400 g for 10 min and fresh medium (RPMI 1640, 200 mM L-glutamine, 20% fetal calf serum, 10 U IL-2, 1% penicillin and 2% streptomycin) was replaced on day 1 and day 3. After 7 days, supernatants were harvested and p24 antigen levels were determined by ELISA according to the manufacturer's instructions (Perkin-Elmer, Waltham, MA). Five hundred thousand uninfected and infected PBMCs either not treated or mixed with matched serum and vaginal secretion dilutions from immunized mice were also stained with 7-AAD viability dye (Beckman Coulter) to test cell toxicity of murine samples. After 20 min incubation at room temperature, cells were centrifuged at 800 g for 10 min in PBS and analyzed by flow cytometry. HIV-2CLB-20 (NIBSC) neutralization assay was performed in a PBMC-based assay to identify non-specific activity of murine samples and the protein p27 was detected by ELISA according to the manufacturer's instructions (ZeptoMetrix Corp., NY).
Tissues obtained from the colon of mice (2-cm specimens from the anus toward the left colon) were fixed in 10% buffered-formalin for 24 hr at room temperature and embedded in paraffin. Haematoxylin-eosin stained sections were used for histological evaluation. The evaluation of IgA+ plasma cells was made on 3 µm paraffin embedded slides, the sections were dewaxed in xylene, rehydrated in an ascendent ethanol scale and pre-treated in a microwave oven (two cycles for 5 minutes each at 780 W, in EDTA buffer, pH 8). Endogen biotin and non-specific signals were blocked with appropriated reagents. For immuno-histochemistry, a goat anti-mouse IgA (dilution 1∶400, AbD Serotec) was used, slides were incubated for 2 hr at room temperature in a humid chamber, washed in PBS, and revealed by biotinylated anti-goat IgG (dilution 1∶50, 1 hr incubation, R&D Systems, London, UK) followed by HRP-Conjugated Streptavidin (30 min incubation, R&D Systems, London, UK). The chromogen was 3, 3′-diaminobenzidine free base (DAB).
Comparisons between groups were analyzed to evaluate immunological differences. Kruskaal-Wallis analysis of variance was performed for each variable; Bonferroni correction was applied to the results. Two-sided p-values were considered. Data analysis was performed using the SPSS statistical package (SPSS Inc. Chicago, IL).