|Home | About | Journals | Submit | Contact Us | Français|
We studied the induction of antigen-specific IgA memory B cells (BM) in volunteers who received live attenuated Shigella flexneri 2a vaccines. Subjects ingested a single oral dose of 107, 108 or 109 CFU of S. flexneri 2a with deletions in guaBA (CVD 1204) or in guaBA, set and sen (CVD 1208). Antigen-specific serum and stool antibody responses to LPS and Ipa B were measured on days 0, 7, 14, 28 and 42. IgA BM cells specific to LPS, Ipa B and total IgA were assessed on days 0 and 28. We show the induction of significant LPS-specific IgA BM cells in anti-LPS IgA seroresponders. Positive correlations were found between anti-LPS IgA BM cells and anti-LPS IgA in serum and stool; IgA BM cell responses to IpaB were also observed. These BM cell responses are likely play an important role in modulating the magnitude and longevity of the humoral response.
Shigella infections continue to be a major cause of morbidity and mortality among children under 5 years old living in the developing world. Every year, there are 165 million cases of shigellosis worldwide and 14,000 cases reported in the United States; it is estimated that because of underreporting, the number of actual cases may be twenty times higher [1,2]. The increasing prevalence of resistance to multiple antimicrobials is of concern  and Shigella is considered a Category B bioterror agent by the CDC . Shigella flexneri is endemic throughout the developing world, and causes more mortality than any other species of Shigella . There is a high demand for a safe and effective oral vaccine, and the WHO has prioritized the development of a well-tolerated vaccine that induces durable immunity against shigellosis [1,6].
By engineering rational deletions in the wild-type Shigella flexneri 2a strain 2457T, two vaccine candidates, designated CVD 1204 and CVD 1208, were constructed at the Center for Vaccine Development (CVD). CVD 1204 contains deletions in guaA (encoding a guanosine monophosphate synthase) and guaB (encoding an inositol monophosphate dehydrogenase), which impair the biosynthesis of guanine nucleotides; CVD 1208 has additional deletions of set and sen genes that encode Shigella enterotoxins 1 and 2, respectively. In a Phase 1 trial CVD 1204 was shown to be clearly attenuated compared to its wild type parent (based on comparison with data from multiple previous challenge studies), while CVD 1208 appeared fully attenuated yet immunogenic . Clinical adverse reactions (diarrhea, dysentery and/or fever) occurred in 8 of 23 recipients of CVD 1204 but in only 1 of 21 recipients of CVD 1208 .
Putative correlates of protection against shigellosis reported in the literature include serum IgG antibodies against lipopolysaccharide (LPS) O antigen and serotype specific O antigen peripheral blood IgA antibody secreting cells (ASC) [2,8,9]. Other antibody and cell-mediated immune responses (CMI) against conserved antigens such as invasion plasmid antigens (Ipa) may also play a role in protective immunity [2,10–13]. An optimal vaccine should not only induce enduring systemic and mucosal antibody responses but also allow the host to mount an anamnestic immune response upon subsequent re-exposure to antigen. This response is faster, stronger, and qualitatively better than primary responses and depends on the presence of BM cells . Following natural Shigella infection, as well as after ingestion of some live attenuated Shigella vaccines, relatively long-term humoral and secondary secretory IgA immune responses to LPS in stool have been described . We have previously demonstrated the induction of IgG BM responses by live attenuated Shigella vaccines in human volunteers . However, the presence of IgA BM responses has not been reported. In this study we examined the hypothesis that volunteers who display mucosal and serum antibody responses to CVD 1204 and CVD 1208 live-attenuated oral Shigella vaccines also exhibit IgA BM cell responses specific to LPS, IpaB and other Shigella antigens.
46 healthy adult volunteers 18–45 years of age from the Baltimore–Washington area received a single oral dose of S. flexneri 2a ΔguaBA (CVD 1204) or S. flexneri 2a ΔguaBA Δsen Δset (CVD 1208) as previously described . Volunteers received 107, 108, or 109 CFU of each vaccine strain or placebo, and sera and stools were collected on days 0, 7, 14, 28, and 42. In addition, peripheral blood mononuclear cells (PBMC) were obtained on days 0 and 28 after oral vaccination. PBMC specimens were cryopreserved and stored in liquid nitrogen until use as previously described . Seroresponse, measured by ELISA , was defined as ≥ 4-fold rise of antigen-specific IgA antibody in serum (seroresponders) and a ≥ 4-fold rise of antigen-specific IgA/total IgA in stool (mucosal responders) after oral vaccination as compared to pre-vaccination. Adequate specimens were available to assay 13 seroresponders and 11 non-seroresponders; these included subjects immunized with placebo or 107, 108, or 109 CFU of the Shigella strains. For BM assays, subjects from all three dosage level cohorts were analyzed. Prior to enrollment, the purpose of the study was explained to the subjects and they passed a written test containing questions regarding the rationale for the study, risks and procedures. Informed consent was obtained from all participants and the study was approved by the UMD Institutional Review Board.
LPS was purified by the hot aqueous phenol method of Westphal . IpaB, IpaC, IpaD, MxiH, VirG and Yersinia pestis LcrV antigens were purified as recombinant proteins from E. coli. A PCR fragment of virG encompassing amino acids 68–774 was copied from the vir plasmid of S. flexneri 2457T by standard PCR. The fragment was ligated into pET22b (Novagen, Madison, WI) and the ligation product used to transform E. coli NovaBlue. The resulting plasmid was sequenced and used in the protein expression system. IpaD, VirG and LcrV were purified by standard His tag chromatography and dialyzed into PBS as described previously [19,20]. IpaC and MxiH were solubilized from inclusion bodies with 6M urea, purified by standard His tag chromatography, and refolded by step dialysis into PBS as previously described [21,22]. IpaB was co-expressed with its cognant chaperone, IpgC, as described previously . The complex was purified by standard His tag chromatography via the His tag fused to the IpgC. IpaB was released from IpgC with 1% n-octyl-polyoxyethylene.
Serum antibodies specific for S. flexneri LPS and Ipa B were measured by ELISA as we previously described . Briefly, plates were coated with LPS (5 µg/mL) or IpaB (0.1 µg/mL) and blocked with 10% dried milk in PBS. Samples were evaluated in serial 2-fold dilutions. HRP–labeled goat anti–human Fc α chain (ICN) was used as conjugate and TMB microwell Peroxidase (Kirkegaard & Perry Laboratories, KPL) as substrate. Titers were calculated from linear regression curves as the reciprocal serum dilution that produced an OD of 0.2 above the blank (EU/ml). Total and LPS-specific fecal IgA were also measured by ELISA as previously described . Plates were coated with either α-chain specific anti–human IgA (1µg/mL; Jackson ImmunoResearch Laboratories) or LPS (10 µg/mL). Stool supernatants were tested in serial 2-fold dilutions. HRP-labeled goat anti–human IgA (Jackson) was used as conjugate and TMB (KPL) as substrate. IgA concentrations were calculated by interpolation into a standard curve of human IgA (Calbiochem). Data are reported as the ratio of LPS-specific/ total IgA levels.
IgA and IgG ASCs specific for S. flexneri LPS and IpaB were detected by ELISpot as we previously described [7,24]. A positive ASC response was defined as a post-vaccination count at least 3 SD above the mean prevaccination count and at least 8 cells/106 PBMC.
Expansion of PBMC to measure BM cells was performed as described by Crotty et al. . Briefly, PBMC specimens were thawed, washed with complete cRPMI 1640 containing 100 IU/mL penicillin, 100 µg /mL streptomycin (CellGro, Manassas, VA), 2 mM L-glutamine (BioWhittaker, Walkersville, MD), and 10% heat-inactivated fetal bovine serum (HyClone, Logan, UT), and expanded for 5 days in 6-well plates (1 × 106 cells/well) in the presence of 1/100,000 pokeweed mitogen (PWM, kindly provided by Dr. S. Crotty), 6 µg /mL CpG-2006 (Qiagen/Operon, Huntsville, AL), 50 uM β-Mercapto-ethanol, and 1/10,000 Staphylococcus aureus Cowan (SAC, Sigma–Aldrich, St. Louis, MO) in cRPMI (expansion media) in a total volume of 2 mL/well. After 2 days of incubation, cells were fed by replacing 2 mL of cRPMI in each well.
96-Well ELISPOT MAHA (Millipore, Billerica, MA) plates were coated with 5 µg/mL LPS, 0.5 µg/mL of IpaB, IpaC, IpaD, VirG, MxiH or Yersinia pestis LcrV (as a negative control) or 5 µg/mL of mouse anti-human IgA in PBS and incubated overnight at 4°C. The plates were then blocked with 1% bovine serum albumin (Sigma) in RPMI for 2 h at 37° C and 5% CO2, and 105 expanded PBMC added per well coated with LPS, IpaB, IpaC, IpaD, VirG, MxiH or LcrV. For total IgA measurements, serial 2-fold dilutions were performed starting at 7,500 expanded cells/well down to 234 cells/well. Cells were incubated for 5 h at 37° C and 5% CO2, washed with PBST followed by PBS, and incubated with mouse anti-human pan IgA Biotin Conjugated Antibody (Hybridoma Reagent Laboratory, Baltimore, MD) overnight at 4° C. Subsequently plates were washed with PBST and PBS and labeled with horseradish peroxidase-conjugated Avidin D (Vector Laboratories, Burlingame, CA) for 1 h at room temperature. The substrate 3-Amino-9 etheylcarbazole C (Calbiochem, La Jolla, CA, USA) was added at 100 µL per well for 20 minutes at room temperature and the reaction was stopped with ddH2O. Final enumeration of specific and total SFC was performed using the Immunospot Series 3B Analyzer ELISPOT reader (Cellular Technologies Ltd, Shaker Heights, OH) with aid of the Immunospot software version 4.0 (Cellular Technologies Ltd).
Adequate expansion of BM cells, assessed by the frequency of total IgA detected by ELISPOT, is critical to the sensitivity and consistency of this method. Thus, specimens which did not reach a minimum cut-off level following expansion (arbitrarily defined as the 10th percentile of the levels reached by all volunteers at any time point: i.e., > 8,300/106 total IgA SFC/106 expanded cells) were excluded from further analysis. Statistical analysis was performed on the mean number of SFC in antigen-coated wells minus the mean number of SFC in the negative control wells. The limit of detection of antigen-specific/total expanded cells in our assays was 1 in 100,000 (0.001%). The limit of detection of antigen-specific/total IgA-secreting cells in each volunteer was determined by taking into account the maximum number of total IgA SFC for that subject in anti-IgA coated ELISPOT wells. The latter had an acceptable minimum of 0.005% (i.e., 1 specific SFC in the 8,300 total IgA SFC/106 expanded cells cut-off).
Eight volunteers had sufficient PBMC available after expansion pre and post vaccination to enable flow cytometric measurements. Of these subjects, 3 volunteers were responders by LPS IgA in serum and stool by ELISA and 5 were non responders. Of the 3 LPS responders, 2 were also IpaB responders. Expanded PBMC were washed with 1% FBS in PBS and labeled with fluorochrome-labeled mAbs against the following antigens: (1) CD19-ECD (clone J3.119, Beckman-Coulter, Fullerton, CA), (2) CD20-APC-Cy7 (clone L27, BD Biosciences, San Jose, CA), (3) CD27-APC-A700 (clone 1A4CD27, Beckman-Coulter), (4) IgA-Biotin (clone G20-359, BD Biosciences - subsequently labeled with streptavidin-Pacific Orange (InVitrogen, Carlsbad, CA), (5) IgG-PE-Cy5 (clone G18-145, BD Biosciences), (6) integrin α4/β7-Alexa 647 (the anti-integrin α4/β7 ACT-1 mAb was kindly provided by Dr. W. Newman, PaxVax Inc., San Diego, CA and conjugated to Alexa 647 using an Alexa 647-labeling kit (Molecular probes, Eugene, OR), and (7) CD14-PacBlue (clone TÜK4 InVitrogen), CD3-PacBlue (clone UCHT1 BD Biosciences) and Vivid (InVitrogen), which were used to exclude cells staining positively with these mAbs using a “dump” channel gating strategy. Incubation with the mAbs was performed in volumes of 50 µl per tube for 20–30 minutes at 4°C, washed with 1% FBS in PBS, and fixed in 300 µl of 1% formaldehyde until run. Events were acquired on a MoFlow flow cytometer/cell sorter system (Beckman-Coulter) and analyzed using WinList 6.0 (Verity Software House, Topsham, ME) software.
Microsoft® Office Excel 2007, GraphPad Prism 5.0, and STATA 9.0 were used for statistical analysis. Our hypotheses were evaluated using non-parametric two-sided tests. Pre- and post-vaccination results were paired. Antigen-specific SFC/106 expanded cells were divided by total IgA SFC. The Wilcoxon signed rank test was used to assess continuous pre- to post-vaccination antigen-specific BM responses. Wilcoxon Rank Sum was used to compare responders to non-responders. Correlations between seroresponse and BM responses were assessed using Spearman rho test for continuous variables and Fisher’s exact test for dichotomous variables. Two sided p values <0.05 were considered significant.
Twenty four volunteers who received vaccine had sufficient PBMC pre- and post-vaccination to be included in these studies. Thirteen of the 24 subjects were anti-LPS IgA seroresponders. All 13 seroresponders were also mucosal responders by sIgA anti-LPS measured in stool. Three individuals were mucosal responders only with ≥ 4 fold increases in anti-LPS/total stool sIgA without a serum response. Cells from one anti-LPS seroresponder had to be excluded from analysis due to inadequate expansion. LPS-specific BM cells increased from a median of 7 SFC/106 expanded cells pre-vaccination to a median of 53 SFC/106 expanded cells 28 days post-vaccination (p=0.005) (Figure 1). Increases were observed in LPS-specific IgA BM in 10 out of 12 IgA anti-LPS (83%) seroresponders with adequate expansion. Of 24 vaccinated volunteers who had sufficient PBMC to perform these studies, 7 also displayed an IgA anti-IpaB seroresponse. Mean IgA anti-IpaB BM cells increased from 4 to 20 SFC/106 expanded cells pre- to post-vaccination (p= 0.062); 4 out of 7 (57%) IgA IpaB seroresponders manifested an increase in IgA anti-IpaB BM cells. The median percentages of antigen-specific SFC as a proportion of total IgA SFC showed increases from 0.03% pre-vaccination to 0.20% post-vaccination for LPS (p value = 0.005) and 0.03% pre-vaccination to 0.10% post-vaccination for IpaB (p=0.2) among seroresponders (Figure 1). Individuals who were not seroresponders did not exhibit a statistically significant increase in antigen-specific BM responses pre- to post-vaccination.
Because of the importance of other proteins in Shigella pathogenesis, e.g., VirG (IcsA), MxiH, IpaB, IpaC and IpaD [26,27], we studied whether immunization with a live oral Shigella vaccine also elicited specific BM to these antigens. Only sporadic BM responses were observed to IpaC, IpaD, MxiH, and VirG. No responses were detected against Y. pestis LcrV (negative control).
The frequency of BM can be quantified as the number of anti-LPS or as anti-IpaB specific IgA per 106 expanded PBMC or the % of anti-LPS or anti-IpaB specific IgA divided by the total number of IgA producing cells in expanded PBMC. As can be observed in Figure 2, highly significant correlations were observed between the two methods of quantification.
Mucosal responses to vaccination (i.e., the production of secretory IgA in stool/total ≥4 fold pre to post oral vaccination) was observed among 16 out of 24 subjects in this study. Among these 16 subjects, strong correlations were observed when comparing their LPS-specific BM (SFC/106 expanded cells) frequencies measured on day 28 with their respective post-vaccination peak seroresponse (Figure 3A), peak mucosal (stool) IgA response (Figure 3B), and peak peripheral blood IgA ASC responses (Figure 3C). Strong correlations were also observed among anti-LPS IgA BM and both mucosal and peripheral anti-LPS IgA responses among the 12 individuals who were seroresponders (data not shown). Of importance, the correlations remain significant when results were analyzed as percentages of specific anti-LPS IgA BM divided by total IgA secreting cells in each volunteer. No significant correlations were observed among antibody levels in serum and stool and peripheral BM cells on day 0 (pre-vaccination, data not shown). Calculations taking into account individual spot sizes (a relative measurement of the amount of antibody produced by each cell) did not change the results. Interestingly, no correlations were observed between the levels of IpaB-specific IgA BM cells and the serum levels of anti-IpaB IgA antibodies or the levels of anti-IpaB ASC in circulation (stool anti-IpaB antibody levels were not measured)(data not shown). Among the vaccine recipients who exhibited a mucosal response, the frequencies of IgA BM cells were better correlated with serum LPS IgA titers (Figure 3) than with serum LPS IgG (Spearman Rho 0.5 p=0.05).
To evaluate whether immunization with CVD 1204 or CVD 1208 resulted in changes in the proportion of B cell subsets of defined phenotypes, PBMC obtained before and after oral vaccination were expanded and stained to examine B cell subsets and expression of the gut homing integrin α4/β7 receptor. Cells were sequentially gated based on forward versus side scatter (“lymph region”), followed by the electronic elimination of doublets, dead cells, CD3+ T cells and CD14+ macrophages. The CD19+ cells co-expressing integrin α4/β7 (i.e., with the potential to home to the gut) or not were then gated and analyzed for their expression of IgG or IgA; each of these cell subsets was further gated based on their expression of CD27 and CD20 to define the CD19+ integrin α4/β7+ CD27+ CD20+ [largely BM] and CD19+ integrin α4/β7+ CD27+ CD20−/low [largely plasmablasts/plasmocytes] subsets [28,29].
The results from these studies demonstrated an increase in IgA secreting cells post-oral vaccination compared to pre-vaccination among the LPS seroresponders. This increase was statistically significant when comparing responders to non-responders by Wilcoxon Rank Sum in the CD19+ integrin α4/β7+ IgA+ CD27+ CD20+ as well as CD19+ integrin α4/β7+ IgA+ CD27+ CD20−/low subsets (Figure 4). No significant differences among responders and non-responders were observed in the percentages of any of the other subsets evaluated (i.e., IgG+ subsets, CD19+ integrin α4/β7− subsets, CD19+ integrin α4/β7+ CD27− CD20+ and CD19+ integrin α4/β7+ CD27− CD20− subsets).
Immune responses induced locally provide the first line of defense against the many pathogens that invade the human host via mucosal surfaces. Humans make more IgA than IgG, IgM and IgE combined and allows for active transport of sIgA across mucosal epithelia to facilitate antigen exclusion and neutralization . The normal human colon has a higher proportion of IgA than IgG-producing plasma cells as evidenced by immunohistochemistry . The induction of BM responses is widely accepted to be a major factor in the ability of vaccines to elicit long lasting, effective immunity. However, the precise role of IgA ASC and BM cells in primary and secondary [anamnestic] immune responses to infection remains ill defined.
A BM response can be demonstrated by documenting: (1) an anamnestic secondary antibody response, both stronger and faster than the initial immune response, (2) avidity maturation, and (3) the presence of BM cells . BM cells have been described in humans against vaccines known to induce a T-cell dependent response, including live viral vaccines administered parenterally (e.g. smallpox vaccine ) and orally (e.g., rotavirus vaccines ) and parenteral conjugate vaccines consisting of bacterial polysaccharides (such as pneumococcal  and meningococcal  capsular polysaccharides) covalently linked to carrier proteins. BM cells have also been reported following natural infection with bacterial enteropathogens and after the administration of live oral bacterial enteric vaccines [16,37]. We here report the observation that oral immunization with attenuated S. flexneri 2a vaccines elicits the generation specific IgA BM cells.
Although the immunological correlates of protection following Shigella infection have not yet been fully elucidated, high numbers of antigen-specific IgA ASC post oral vaccination have been found to be associated with protection from shigellosis after experimental challenge . Interestingly, the volunteers who have high numbers of antigen-specific ASC after vaccination and are asymptomatic after challenge have low levels of antigen-specific IgA ASC in peripheral blood after secondary antigen exposure . In the present study we observed that a single oral immunization with live attenuated Shigella vaccine CVD 1204 or CVD 1208 elicited significant increases in antigen-specific IgA anti-LPS BM responses among subjects who mounted ≥ fourfold specific sIgA or IgA antibody responses as measured by ELISA, respectively, in stool and serum. IpaB seroresponders also exhibited antigen-specific IgA BM responses to IpaB. Strong correlations were observed between the magnitude of IgA anti-LPS BM cells and increases in anti-LPS IgA responses, both among individuals who displayed a strong mucosal sIgA stool response and among IgA seroresponders. BM cells also correlated highly with anti-LPS IgA ASC, both among seroresponders and mucosal responders. It will be important to determine in future experimental challenge studies with wild-type Shigella whether antigen-specific IgA and IgG BM cells are elevated in subjects who had been (or not) previously immunized with oral live attenuated Shigella strains and whether BM cells are associated with protection.
We were surprised that no differences were detected in the magnitude of BM cell responses among recipients of the slightly reactogenic CVD 1204 or the very well tolerated CVD 1208, or related to the various dosage levels evaluated (107, 108, or 109 CFU). An optimistic interpretation of this observation is that the loss of reactogenicity resulting in a well tolerated further attenuated Shigella vaccine strains such as strain CVD 1208 is not accompanied by a diminution of the ability to elicit strong immune responses, including the generation of antigen-specific IgA BM cells.
In previous studies we determined by flow cytometry the proportions of phenotypically defined total B, BM and IgG+ BM cell populations before and after expansion . This information helped validate the BM cell ELISPOT assay as well as provide the rationale for limiting the analyses to specimens that showed evidence of appropriate in vitro expansion in the presence of mitogens. By eliminating specimens that exhibited suboptimal expansion of functional cells (defined as ≤8,300 total IgA SFC/106 expanded cells) in the current dataset, we observed increased levels of circulating antigen-specific IgA BM responses elicited by oral vaccination of up to 1.1% antigen-specific BM SFC/total IgA SFC post-vaccination for T-independent antigen LPS and up to 1.0% for the T-dependent antigen IpaB. These proportions of antigen-specific BM cells are similar to those previously reported in subjects who received efficacious parenteral vaccines. For example, responses to diphtheria and tetanus toxoids elicited specific BM cells over total IgG secreting cells in the range of 0.01–1% . Recombinant hepatitis B vaccine showed the induction of 0.07% hepatitis B surface antigen-specific Ig secreting cells over total Ig secreting cells; 88% of vaccinees had detectable levels of IgG BM cells and 76% had detectable levels of IgA and IgM BM cells . In natural cholera infection, anti-LPS IgA BM cells have been reported to be 0.6% of total IgA BM cells and proposed to play an important role in the anamnestic mucosal immune response . To our knowledge, the data included in the present manuscript is the first description of the induction of specific IgA BM cell response to LPS in recipients of an oral live-attenuated bacterial vaccine.
Since Shigella enters the host via the gut, it is important to define the local immune responses elicited by immunization. Due to regulatory constraints and other factors, intestinal mucosal biopsies are very difficult to obtain following vaccine administration or challenge with wild-type organisms. Alternative ways to investigate whether antigen-specific IgA and IgG ASC and BM found to be elevated in the periphery after oral vaccination have the potential to migrate to the gut mucosa involve indirect measurements, such as sIgA in stool and the measurement of gut homing molecule expression, e.g., integrin α4/β7, on IgA+ and IgG+ ASC and BM cells. Although in our experience flow cytometry is not as sensitive as ELISPOT in assessing antigen-specific BM responses, it nevertheless allowed us to characterize the phenotype of the B cell subsets that increased following oral vaccination.
In the present study we found significant increases in the proportion of vaccine-induced CD19+ CD27+ CD20+ integrin α4/β7+ IgA+ BM cells with a gut homing pattern in seroresponders as compared to non-seroresponders. Smaller, yet significant, increases have also been observed in CD19+ CD27+ CD20− integrin α4/β7+ IgA+ cells (a phenotype typically associated with plasmablasts) . While BM only express surface Ig, expression of Ig in plasmablasts is largely cytoplasmic with lower levels of surface Ig expression . Because this manuscript is focused on BM and sufficient cells were not available for independent staining panels (i.e., extracellular and intracellular IgA and/or IgG staining), we stained extracellularly for Ig. Thus, it is likely that the smaller differences observed in integrin α4/β7+ IgA+ plasmablasts in responders as compared to non-responders are the result of a lower efficiency in the detection of IgG+ and IgA+ plasmablasts. Future studies using intracellular Ig staining will allow this issue to be addressed directly. In contrast to the increases in IgA+ cells, this phenomenon was not observed in CD19+ α4/β7+ IgG BM cells (irrespective of expression of CD27 and/or CD20). In sum, we found that IgA+, but not IgG+, CD19+ B cells expressing the gut homing receptor integrin α4/β7 increased 28 days after oral vaccination, and that gating on subsets based on the expression of CD27 and CD20 enhanced the ability to identify this population. This observation is consistent with Crotty’s original description of the BM cell ELISPOT where sorting experiments were used to identify that BM have a CD27+ CD20+ phenotype . The present results suggest an increase in circulating Shigella specific IgA BM and plasmablasts with gut homing potential in individuals in whom immunization elicited anti-Shigella humoral responses.
A limitation of this study is the relatively small number of volunteers who could be evaluated, which is largely due to the fact that it employed “convenience” specimens based on availability. In spite of this limitation, we observed strong, statistically significant, specific anti-LPS IgA BM responses and associations with anti-LPS IgA antibody and ASC levels. However, it is very likely that the small sample size provided insufficient power to adequately estimate the presence and association of anti-IpaB IgA BM cells and seroresponses. Future studies will address this issue by evaluating larger numbers of subjects and vaccine candidates.
Virulent Shigella target the M cells that overlie gut-associated lymphoid tissue and rapidly attain an intracellular niche within epithelial cells. SIgA anti-Shigella antibodies can prevent mucosal invasion. Moreover, wild type organisms coated with such antibody elicit less profound inflammatory responses than Shigella not coated with antibody . It is not known how long sIgA intestinal and serum IgG anti-Shigella antibodies stimulated by clinical infection or oral immunization with Shigella vaccines remain elevated at levels adequate to prevent invasion and clinical illness. If antibody levels decline to non-protective levels, it falls upon BM cells to mount a sufficiently rapid anamnestic response to limit the clinical consequence of exposure to virulent Shigella. Experimental challenge studies with immunologically-naïve healthy adult volunteers and epidemiologic investigations of point source outbreaks suggest that for most Shigella serotypes the usual incubation is ~ 1–3 days from the point of ingestion of virulent organisms until the onset of clinical illness. Thus, an effective anamnestic response mediated by BM cells must occur within a mere 1–2 days. Few reports have directly addressed the duration of immunity that follows an initial clinical Shigella infection in the absence of repetitive boosting (as would occur in an endemic situation). Similarly, to our knowledge the only controlled studies that describe the duration of protection conferred by immunization with live oral Shigella vaccines were those of Mel et al who immunized children living in an endemic area where Shigella was highly seasonal [43,44]. Oral immunization with four doses over two weeks of streptomycin-dependent live oral Shigella vaccines led to significant serotype-homologous protection that endured for one year . However, the administration of just a single oral booster dose was able to prolong the protection for an additional year .
In summary, oral vaccination with a single oral dose of a live-attenuated Shigella vaccine elicited antigen-specific IgA BM responses among IgA seroresponders and sIgA mucosal responders with a high degree of correlation between antigen-specific BM cells and both serum and mucosal titers as well as antigen-specific ASC in peripheral circulation. Mucosal sIgA responses are more difficult to measure than serum antibody, as is the measurement of small number of IgA BM cells. Nevertheless, we observed positive correlations between antigen-specific anti-LPS BM cells and antigen-specific anti-LPS sIgA/total IgA in stool. As Shigella invades via the gastrointestinal mucosa, fecal sIgA responses likely play an important role as the first line of defense in protection. Antigen-specific IgA BM cells should be further studied in natural infection and challenge studies to determine their role in the anamnestic responses and protection against this important and devastating pathogen. For expedited identification of future candidate vaccines, it will be necessary to identify reliable surrogates of protection against symptomatic infection. A strong correlation between anti-LPS IgA BM cells and peak anti-LPS IgA antibody and ASC responses further reinforces our contention that BM cells may be an important indicator for long-term humoral immunity and a possible surrogate of protection against shigellosis.
Oral vaccination with live-attenuated S. flexneri 2a elicited detectable IgA BM cells to LPS and, to a lesser extent, to IpaB. Positive correlations were observed between anti-LPS IgA BM and both anti-LPS secretory IgA /total IgA in stool and anti-LPS IgA in serum as well as circulating ASC. In addition, flow cytometric analyses revealed significant differences in IgA+ BM subsets expressing integrin α4/β7 pre to post vaccination when comparing seroresponders to non-responders. Our results support the contention that BM cells may be an important indicator for long-term humoral immunity and a possible surrogate of protection against shigellosis
We thank the volunteers for participating in the clinical trial, the clinical and regulatory staff at the CVD, Drs. William Blackwelder and Yhukun Wu for helpful statistical discussions, and Dr. S. Crotty for providing PWM. Support for this research was provided by NIH R01-AI057927 (to M.B.S.), U19-AI-082655 (CCHI; to M.B.S.), K23-AI065759 (to J.S.) and N01-AI25461 (VTEU, to M.M.L). The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institute of Allergy and Infectious Diseases or the NIH.
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.