The route of vaccination is important in influencing immune responses at the initial site of pathogen invasion where protection is most effective. Immune responses required for mucosal protection can differ vastly depending on the individual pathogen. For some mucosal pathogens, including acute self-limiting infections, high-titer neutralizing Abs that enter tissue parenchyma or transude into the mucosal lumen are sufficient for clearing cell-free virus. However, for pathogens causing chronic infections such as HIV, hepatitis C virus, herpes viruses, mycobacteria, and fungal and parasitic infections, a single arm of the immune response generated by systemic vaccination may be insufficient for protection. Induction of the mucosal innate and adaptive immune systems, including CD4+ T help, Th17, high avidity CD8+ CTL, and secretory IgA and IgG1 neutralizing Abs, at the site of pathogen entry may be required for effective protection against highly invasive pathogens that lead to chronic infection and may be generated predominantly by mucosal vaccination.
The major entry point for many human pathogens occurs at gastrointestinal (e.g., polio virus, Escherichia coli, Salmonella, Shigella, Vibrio cholerae, Helicobacter pylori, and HIV-1), respiratory (e.g., influenza virus, Mycobacterium tuberculosis, adenovirus, coronavirus, rhinovirus, respiratory syncytial virus), or genital (HSV, human papillomavirus (HPV),2 HIV-1, Chlamydia, and Neisseria gonorrhoeae) mucosal surfaces (1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11). The main function of the innate mucosal immune system, to discriminate between dangerous and innocuous organisms, is determined by the recognition of specific pathogen-associated molecular patterns via activation of TLRs, NOD-like receptors, retinoic acid (RA)-inducible gene I-like helicases, and C-type lectins (12). Mucosal pathogens may disseminate to distant systemic sites through entry into the blood stream, although some microorganisms are limited to the development of disease only at the site of initial mucosal invasion (8, 13, 14). For HIV, numerous studies have demonstrated that the gut mucosa is the predominant site of viral replication and amplification (15).
Protective immunity against mucosal pathogens will require novel vaccine strategies to induce mucosal immune responses tailored to the anatomic location and the threat of the invading pathogen (16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32). However, the requirement for mucosal immunization to generate protective “frontline immunity” against mucosal pathogens is highly controversial. On the one hand, numerous studies in the literature have demonstrated that immune responses are readily detectable at mucosal sites following systemic delivery of vaccines, and complete or partial protection from mucosal challenge is attainable (2, 33, 34, 35, 36, 37, 38, 39, 40). Systemic immunization is adequate for successful vaccines for some mucosal pathogens, notably the polio virus and the influenza virus, where high titers of neutralizing Abs are capable of clearing cell-free virus and preventing disease (34, 41). On the other hand, mucosal pathogens such as HIV-1, HPV, herpes viruses, Mycobacterium species, and other intracellular pathogens may require mucosal vaccine strategies that activate multiple arms of the innate and adaptive immune systems (29, 30, 42, 43, 44, 45). We as well as others have shown that protective mucosal immune responses are most effectively induced by mucosal immunization through oral, intranasal (i.n.), intrarectal, or intravaginal routes, and an optimized mucosal vaccination strategy may have a much greater potential for generating local protective mucosal immune responses (30, 46, 47, 48, 49, 50, 51, 52). Studies evaluating mucosal infection and immunization in humans and animals have demonstrated the existence of a common mucosal immune system (CMIS) that consists of gastrointestinal, respiratory, and genital mucosa (3, 46, 53, 54). The CMIS implies the ability of Ag-specific lymphocytes to home to mucosal effector sites in addition to the site where initial Ag exposure occurred (53). Different mucosal routes of immunization such as oral, nasal, or rectal routes can induce generalized mucosal immune responses not only at the portals of entry of infectious agents but in distant mucosal effector sites as well, although additional constraints on mucosal compartmentalization are evidently dependent upon the adjuvant and the Ag delivery vehicle. The term mucosal compartmentalization refers to the process of Ag uptake, migration, and processing by local APCs and the magnitude and phenotype of T cells or predominant Ab isotype(s) induced in relationship to the route of immunization and mucosal site(s) analyzed (3, 54). Cytokine and costimulatory molecule expression by locally activated dendritic cells (DCs), epithelial cells, and stromal cells account for the differences in T cell function induced in proximal mucosal sites vs distal tissue or lymph nodes. Thus, the generation of mucosal immunity after systemic vaccination may not recapitulate the immune signatures generated after mucosal vaccination with the same Ag. In general, compartmentalized mucosal immune responses (CMIRs) that consist of innate responses, mucosal Abs (secretory Ig (sIg) A and sIgG) and CD8+ CTLs localized to tissues proximal to the mucosal site of immunization are necessary for protection from mucosal pathogens (Fig. 1). Optimum mucosal vaccination leading to compartmentalized mucosal immune responses might ensure that the appropriate cells are armed and ready to respond immediately to infection and to confer protection not achieved following natural infection (Fig. 1). Vaccine strategies that induce both Ag-specific CD4+ Th cells in draining lymph nodes to provide help during recall and effector/memory CD8+ cells in mucosal effector sites (lamina propria (LP)) will be crucial in containing initial viral replication and subsequent disease course following HIV infection. Induction of central memory CD8+ T cells rapidly recruited to sites of initial virus replication could provide a concerted second wave attack to limit spread of the virus systemically.
Only a few mucosal vaccines have been approved for human vaccination, including poliovirus, influenza, Salmonella typhi, and V. cholerae vaccines (16, 20, 21, 24). Improved mucosal vaccines are needed against a wide range of respiratory, gastrointestinal, and urogenital tract mucosal pathogens. The main challenge for mucosal vaccine development is to design vaccines that are sufficiently immunogenic and can be safely delivered by mucosal routes of vaccination to prevent (or reduce) transmission of the mucosal pathogen through the mucosal barrier and dissemination into the systemic circulation. This will require vaccine strategies that induce immune responses tailored to the individual pathogen. In this review we posit mechanisms of protection for existing mucosal vaccines currently in use and provide analysis of experimental studies comparing mucosal vs systemic vaccination strategies. This knowledge may help to provide unifying principles toward the design of better vaccines with mucosal delivery strategies that can be effective against diverse pathogens associated with chronic mucosal infection and disease.
A timeless debate: the oral poliovirus vaccine (OPV) vs inactivated poliovirus vaccine (IPV)
The first effective polio vaccine was developed in 1952 by Jonas Salk at the University of Pittsburgh (Pittsburgh, PA). The Salk or inactivated poliovirus vaccine is based on three wild-type reference strains (Mahoney, MEF-1, and Saukett) inactivated with formalin. The oral poliovirus vaccine was developed by Albert Sabin using attenuated live poliovirus that he had received from Hilary Koprowski (16, 55). Sabin’s vaccine was licensed in 1962 and the World Health Organization adopted a policy for the exclusive use of OPV in developing countries. The advantage attributed to the live attenuated vaccine is that it persisted in the gastrointestinal tract for 4–6 wk and induced both mucosal and serum-neutralizing anti-poliovirus Abs against poliovirus types 1, 2, and 3. In contrast, the IPV is also highly immunogenic, confers low titers of mucosal neutralizing Abs, and affords protection, all qualities of a good vaccine. However, parenteral vaccination with IPV does not appear to be highly effective in inducing secretory Ab response in the nasopharynx or intestinal mucosa during primary immunization (16). In some cases, IPV alone was shown to provide inadequate protection against poliovirus, especially type 1 (56). A recent study (1) characterized mucosal immunity in a polio-free community immunized with a combined enhanced inactivated/oral polio vaccine. The percentage of excreted virus was inversely correlated with prechallenged intestinal neutralizing Ab titers. The authors concluded that the main factor in the interruption of disease and virus transmission in the community was a memory Ab response induced by the vaccine. In the United States, <10 cases of paralytic poliomyelitis are reported each year, and all cases are associated with the vaccine (57). In developing countries, the incidence of poliomyelitis is due to poor seroconversion rates after vaccination with IPV (57). In the United States OPV is recommended for routine vaccination in part because of its ability to generate intestinal and pharyngeal immunity to enhance protection during reinfection (17). It was demonstrated that even enhanced potency IPV was less effective than OPV in preventing and limiting intestinal infection, even though the vaccine generated higher serum Ab titers (17). According to a recent publication, healthcare officials in Germany and Switzerland recommended vaccination with IPV instead of OPV (58) in an effort to achieve complete global eradication of the virus. However, travelers vaccinated with IPV only might become infected with vaccine-derived virus, shedding it in large quantities and increasing the probability of transmission to unvaccinated individuals. It is therefore recommended to include at least one dose of OPV in a polio vaccination schedule to generate strong intestinal immunity (58).
The prospect of mucosal vaccines for influenza
Systemic delivery of inactivated influenza vaccines has proven effective in providing protection against the specific influenza subtypes included in the vaccine. These vaccines are capable of eliciting high-titer, subtype-specific neutralizing Abs that prevent disease but are insufficient to prevent infection at mucosal surfaces. Mucosal delivery strategies that mimic the process of natural infection to elicit effective innate and adaptive immune responses are currently being explored in an effort to broaden the protective capabilities of the influenza vaccines in current use. The innate immune response to influenza virus involves activation of multiple pathways of pattern-recognition receptors in immune and nonimmune cells, including TLRs, RA-inducible gene I-like receptors (RLR), and the NALP3 inflammasome (59). Currently licensed inactivated influenza vaccines administered by the i.m. route induce systemic IgG Abs but fail to induce highly cross-reactive mucosal IgA or heterosubtypic CD8+ CTL. Therefore, yearly influenza vaccines are developed based upon prediction of the most common subtypes prevalent in the coming flu season. Licensed attenuated influenza vaccine in man can induce mucosal IgA and IgG Abs and CTL against virus subtypes, but their use is precluded due to adverse effects in immunocompromised individuals.
A recent study found that mice who received two i.n. immunizations of H1N1 influenza virus-like particles (VLPs) were protected against a lethal challenge with both the 1918 H1N1 virus and the heterotypic H5N1 virus (60). In addition, mucosal immunization was able to induce a higher level of cross-reactive IgG and IgA Abs compared with parenteral vaccination with the same vaccine (60). Mice immunized i.m. with two doses of 1918 VLPs were only protected against the homologous virus challenge. Mucosal vaccination with H1N1 VLPs in ferrets also provided complete protection against lethal challenge with the H5N1 virus, whereas only 50% of ferrets survived infection after i.m. vaccination (60). Pulmonary immunization with extremely low doses (375-fold reduction in Ag dose) of influenza vaccine with the ISCOMATRIX adjuvant induced significantly higher titer serum Abs, and neutralizing anti-hemagglutinin Abs were only present in the lung after vaccination via the respiratory tract (61). In the cotton rat model, partial protection against different influenza A subtypes has been achieved by i.n. immunization with a live virus or by i.m. vaccination with an inactivated virus (62). Another study found that parenteral immunization with influenza vaccine in humans could induce both systemic immunity (anti-influenza Ab in serum and Ag-specific CD4+T cells in PBMC) as well as IgA Abs in saliva and IFN-γ-secreting T cells in palatine tonsils (63). FluMist is the first live attenuated influenza vaccine and also the first nasally administered vaccine to be marketed in the United States for the prevention of flu in individuals 2–49 years of age (64). FluMist is administered as an i.n. spray and has been shown to provide high level of protection against both matched and mismatched strains in children and adults. In a direct comparison of FluMist and inactivated vaccine in children, individuals receiving FluMist expressed 35–53% fewer cases of flu illness caused by antigenically matched strains (64).
Live oral attenuated vaccine against typhoid fever (S. typhi Ty21a)
Salmonella infections transmitted by contaminated poultry and eggs represent a major global health burden. The Ty21a vaccine is available as enteric coated capsules or as a “liquid formulation” (lyophilized vaccine organisms reconstituted with buffer and water into a vaccine mixture). A recent study demonstrated that the inductive site of immunization rather than the isotype commitment of Ab-secreting cells (ASCs) was the determining factor in cytokine responsiveness and migration of human effector B cells (18). Oral immunization with a Salmonella vaccine induced a robust migration of specific IgA and IgM Ab-secreting cells that coexpressed CCR9 and CCR10 toward the mucosal chemokines CCL25 (TECK) in the small bowel and CCL28 (MEC) in the colon (18). After oral Ty21a vaccination, all circulating ASCs in peripheral blood expressed the α4β7 integrin and migrated to the mucosal compartment, whereas only a portion of ASCs after parenteral Ty21a vaccination expressed mucosal homing markers (65). Oral vaccination with Ty21a predominantly elicited immune response in saliva and vaginal secretion, whereas rectal immunization was most effective in inducing responses in nasal secretion, rectum, and tears (53). Mucosal application of Salmonella Dublin flagellin containing microparticles has been shown to enhance transepithelial transport by the follicle-associated epithelium (FAE) in mouse Peyer’s patches (PPs) and rapid migration of subepithelial DCs into the FAE (66). Thus, innate responses elicited by Salmonella flagellin in mucosal epithelial cells are important in determining an effective adaptive immune response (66).
In human clinical trials a two-dose regimen with the Ty21a vaccine has proven to provide statistically significant protection for the first 2 years, whereas a three dose-regimen provided protection for up to four years (67). The purified bacterial capsule (Vi vaccine) given parenterally appears to be less toxic and equally effective for up to 2 years compared with a single-dose parenteral immunization with the whole cell vaccine. Two dose regimens provided significant protection for 5 years, yet whole cell vaccines were associated with higher toxicity (67).
Live oral attenuated vaccine against cholera (V. cholerae CVD 103-HgR)
The CVD 103-HgR is constructed by deleting 94% of the enzymatically active A subunit of cholera toxin from the classical Inaba V. cholerae O1 569B. This vaccine, administered as a “liquid formulation,” is highly immunogenic following a single oral dose. In experimental challenge studies in volunteers, a single dose of CVD 103-HgR resulted in intestinal colonization and rapid immune responses that conferred protection for up to 6 mo against challenge with either the wild-type V. cholerae O1 or the El Tor biotype and either the Inaba or the Ogawa serotype. The best correlate of protection against cholera is the titer of serum vibriocidal Abs directed against the O Ag of V. cholerae O1. However, one study found no correlation between protection and secondary Ab responses following wild-type homologous challenge in individuals vaccinated with CVD 103-HgR (68). It is possible that pre-existing immunity may limit the efficacy of the live oral vaccine among partially immune individuals. A safety concern associated with CVD 103-HgR is the possibility of horizontal gene transfer and recombination, leading to reversion to virulence. In volunteer studies, CVD 103-HgR provides 62–100% protection against cholera caused by V. cholerae for at least 6 mo; however, results of a large-scale field trial in Indonesia demonstrated lower protective efficacy. A single dose of CVD 103-HgR was found to be immunogenic and without adverse effects in HIV-infected adults in Mali (24).
The live attenuated vaccine strains V. cholerae CVD 103-HgR and S. typhi Ty21a can be combined into an oral bivalent vaccine without compromising the immunogenicity of the individual vaccine strains. Seroconversion rates of 87–94% for Inaba vibriocidal Abs and 72–91% for anti-S. typhi LPS Abs were reported in healthy volunteers receiving the bivalent vaccine.
The live attenuated bacillus Calmette-Guérin (BCG) vaccine against tuberculosis (TB)
Since 1923, BCG vaccines have been used worldwide and, to date, are the only licensed TB vaccines administered in humans. Over 3 billion people have been vaccinated with BCG, and over 100 million newborn children are vaccinated with BCG every year. BCG induces protection against childhood TB but fails to protect against pulmonary TB in adults. Because pulmonary TB represents >85% of the total TB burden, new-generation TB vaccines are urgently needed. A mucosally administered vaccine could provide better protection against M. tuberculosis infection in the respiratory mucosa. A recent study demonstrated that i.m. immunization with a M. tuberculosis DNA vaccine induced robust T cell responses in the spleen but failed to generate Ag-specific T cells in the airway lumen (28). Pulmonary protection against M. tuberculosis after systemic vaccination is critically determined by T cell geography and depends on whether Ag-specific CD8+ T cells are located within or outside of the mucosal lumen. Airway boosting immunizations with soluble Ag were shown to effectively mobilize systemically activated T cells into the airway lumen. Ag-specific local airway T cells can provide significant protection against airway M. tuberculosis challenge. Adoptive transfer of either local airway T cells of i.n. vaccinated animals or splenic Ag-specific T cells of systemically immunized mice into the lungs of naive SCID mice could confer protection against pulmonary M. tuberculosis exposure (69). Protection against pulmonary M. tuberculosis challenge correlated with the generation of Ag 85A-specific, IFN-γ-producing T cells in draining lung lymph nodes. In vivo neutralization experiments demonstrated that the accumulation of T cells in the airway lumen was mediated by chemokines, including MIP-1α or IFN-inducible protein-10 (28). A recombinant vesicular stomatitis virus (VSVAg85A) was shown to be immunogenic in mice by either the i.n. or the i.m. route (70). However, only i.n. vaccination with VSVAg85A generated a mucosal T cell response that was protective against pulmonary M. tuberculosis exposure. A heterologous VSVAg85A vaccine boost following recombinant adenoviral prime immunization translated into significantly better protective efficacy against pulmonary challenge with M. tuberculosis than either vaccine used alone (70). For protection following a BCG prime immunization, an i.n. AdAg85A boost was protective whereas i.m. boosting with the same recombinant vector was not effective. A recent study showed that i.n. administration of the antigenic fusion protein Ag85B-EAST-6 mixed with the mucosal combined adjuvant vector CTA1-DD/ISCOM could significantly boost IFN-γ-secreting CD4+ T cells in the lung elicited by prior BCG immunization and improved protection against M. tuberculosis challenge (25).
Candidate mucosal vaccines against HIV/AIDS
Gastrointestinal and vaginal mucosa are the primary sites of natural HIV transmission and a major reservoir for virus replication. Approximately 30–40% of all new HIV-1 infections occur in women through the vaginal mucosa. It is clear today that an effective HIV vaccine must be capable of eliciting local mucosal immune responses (both mucosal Ab and CD8+ CTL) to impact initial virus replication and prevent CD4+ depletion, enteropathy, immune dysregulation, and the rapid dissemination of virus from the mucosa into the systemic lymphoid tissue. HIV-1 envelope-specific mucosal IgA Abs have been shown to block mucosal HIV entry in vitro as well as transcytosis of the virus through epithelial barriers. Highly exposed seronegative women who have unprotected sex with HIV-positive individuals have been shown to secrete β-defensins and have sIgA in their cervicovaginal secretions that could contribute to a temporary protective state (71). In addition, passive infusion of IgG1 neutralizing Abs has been shown to be effective in the protection of macaques against vaginal transmission of pathogenic simian HIV (SHIV). A recent study in macaques evaluated whether a plasma concentration of Ab corresponding to modest neutralization titers in vitro could protect animals against intravaginal challenge with low doses of SHIV that uses the CCR5 coreceptor (72). It was demonstrated that lower titers of neutralizing Ab could provide a benefit in the context of typical human exposure to HIV-1 (72). However, for protective T cell responses there is no consensus within the field as to whether parenteral immunization can generate mucosal immunity sufficient to protect against mucosal exposure to HIV-1, or whether a mucosal route of vaccination is capable of generating the quality of compartmentalized local immunity necessary for an effective HIV vaccine. A recent study by Kaufman et al., found that i.m. immunization with recombinant adenovirus generated CD8+ CTL responses at multiple mucosal sites in both mice and rhesus macaques (39). The authors concluded that i.m. immunization with a replication-defective recombinant adenovirus was sufficient for induction of both systemic and long-lived tissue resident mucosal CD8+ T cell memory, thereby bypassing the necessity of a mucosal route of administration for the establishment of Ag-specific T cells at mucosal effector sites (39). Another study by Pal et al., showed that systemic immunization with an ALVAC-HIV recombinant vaccine decreased the viral set point in blood as well as in mucosal sites and protected macaques from peripheral CD4+ T cell depletion (36). It is important to note that in these two studies a mucosal immunization regimen was not evaluated for comparison of the relative efficacy of systemic vs mucosal delivery. Although results from a number of studies have demonstrated that mucosal immune responses are detected after systemic vaccination, comparative studies in macaques have shown that a vaccine delivered to mucosal inductive sites can be more effective in eliciting high avidity protective mucosal immune responses than the same vaccine given systemically. Results from our studies showed that a mucosal vaccine can induce mucosal immune responses and protect macaques against mucosal infection with SHIV much more effectively than the same vaccine given s.c. (42). In intrarecatally immunized macaques, viral load was more effectively controlled in intestinal tissues, which are major reservoirs for virus replication and persistence (42). Although s.c. immunization could induce tetramer-positive CD8+ T cells in gut mucosa, functionally active CD8+ CTLs in mucosal compartments were best generated after intrarectal immunization (46). Subcutaneous vaccination generated higher avidity CD8+ T cells in systemic lymphoid tissue, whereas mucosal immunization induced a higher proportion of functionally active CD8+ CTLs in the GALT. Generation of high-avidity CD8+ CTLs by the mucosally delivered vaccine inversely correlated with viral load in the gut mucosa and directly correlated with the preservation of CD4+ T help in the colonic LP after intrarectal challenge of macaques with SHIV (46). Furthermore, we showed that a single systemic immunization is capable of inducing high avidity CTLs in regional lymph nodes, whereas a mucosal heterologous prime-boost strategy is critical for the generation of high-avidity CD8+ CTL responses in mucosal tissue, at least in a small animal model test system (73).
CMIS vs CMIR and “long-lived” mucosal memory T and B cells
Evidence for a CMIS is based on the concept that primed B cells from the gut subsequently enter the circulation and migrate to local (intestinal LP) and distant mucosal tissues (the respiratory and urogenital mucosae) and to glands where terminal differentiation occurs. For example, it is well known that intrarectal immunization can elicit specific Abs not only in local rectal secretions but can also lead to Ab production at distant sites via the CMIS. Intrarectal immunization with the Ty21a typhoid vaccine generated Ag-specific IgA in saliva. However, local exposure to Ag can result in much higher levels of specific IgA in the region of exposure than at distant sites. Intrarectal vaccination induces the most optimum mucosal immune responses in the rectum and yet was ineffective for eliciting cholera toxin B-specific Abs in the genital mucosa in women (45). The generation of vigorous and protective HIV immune responses in both the rectum and the genital mucosa will likely require a combination of rectal and vaginal immunization. It is the authors’ opinion that compartmentalized mucosal immune responses in the majority of cases play a much more significant role for mucosal vaccine protection compared with mucosal responses in distant sites based on the concept of a common mucosal immune system.
Anatomic location directly impacts memory T cell formation and functional programs. Mucosal memory T cell phenotypes differ substantially depending on the regimen of immunization, and secondary responses result in preferential accumulation of memory T cells in the LP following mucosal immunization. Thus, heterologous prime-boost immunization has been shown to have a major effect on CD8 T cells’ functional quality and surveillance within mucosal effector sites (74). Effector memory cells at mucosal sites appear to be “long-lived” and can proliferate and exercise a similar function as that of central memory in protection from disease (74). Furthermore, a direct correlation between the magnitude and quality of effector CD8+ T cells at mucosal sites and viral load was found during acute SIV infection (75). In our opinion, mucosal immunity and compartmentalized mucosal protection benefit the most from prime-boost strategies, and such strategies should be widely applied for future mucosal vaccine development against infectious diseases and cancer. However, to prevent viral transmission through a mucosal surface, immune cells should be substantially activatable because the presence of memory T and B cells in LP may not be sufficient for protection. Recently, a number of studies have reported on the immunogenicity and safety of plant-derived, cold chain-free oral vaccines that effectively activate T and B cell immunity at mucosal sites.
Role of mucosal DCs in the generation of protective immunity in mucosa
Mucosal DCs play a central role in both the induction and maintenance of tolerance toward the normal microflora and the foods we ingest as well as the induction of protective immunity against invasive pathogens. Unique DC subsets are responsible for Ag presentation following mucosal vaccination, and systemic immunization may have a limited effect on the delivery of Ag to mucosal DCs (46, 76, 77, 78, 79, 80, 81). Other nonhematopoietic cell types, such as intestinal epithelial cells, that encounter a mucosal pathogen (e.g., invasive Salmonella typhimurium) serve important roles in the induction of adaptive immune responses and can produce CXCL8, which rapidly attracts neutrophils to the site of infection followed by inflammatory monocytes and DC precursors, with eventual infiltration of lymphoid cells into the inflammatory site (82). DC precursors recruited to the mesenteric lymph nodes (MLNs) during inflammation can undergo activation in situ before exposure to stromal and epithelial conditioning; they are fully capable of secreting IL-12 and are potent inducers of Th1 IFN-γ responses. During infection with S. typhimurium, activation of a protective T cell response requires the recruitment of CCR6+ DCs from the subepithelial dome of PPs to the follicle-associated epithelium (83) (Fig. 2). CX3CR1+ DCs also reside in the follicle-associated epithelium, where they can extend dendrites into the intestinal lumen to sample Ags (Fig. 2). DCs normally resident in the MLNs (CD103-MLN DCs) are capable of producing higher levels of proinflammatory cytokines than intestine-derived CD103+ DCs and driving IFN-γ production by CD4+ T cells (76) (Fig. 2). Although the PPs and MLNs are hotbeds of regulatory T cells (Tregs) during normal homeostasis, innate signals delivered to DCs and activated T cells can temporarily suspend Treg suppressive function and shift the balance from an anti-inflammatory milieu characterized by cytokines IL-10 and TGF-β toward a proinflammatory milieu of Il-1β, IL-1α, IL-6, IL-12, IFN-γ, and IFN-α leading to the induction of adaptive immune responses calibrated to the pathogenic threat (Fig. 2). A recent study showed that IL-10-secreting Tr1 regulatory T cells lose their suppressive functions and convert to IFN-γ- and IL-4 secreting T cells following CpG ligation of TLR9 up-regulated on Tr1 cells by 1α,25-dihydroxyvitamin D3 (84). In addition, TLR8 ligation expressed on natural FoxP3+CD4+CD25+ cells abrogates Treg suppression, whereas TLR2 ligation results in temporary loss of suppressor activity (85, 86). A recent study linked the sphingosine 1-phosphate receptor type 1-Akt-mammalian targets of rapamycin kinase pathway (S1P1-Akt-mTOR) to suppression of the FoxP3 Treg function, suggesting one mechanism for the differential regulation of Treg function and the activation of naive T cells for induction of adaptive immune responses in MLNs (87). In addition, CD4+ T effector cells stimulated through TLR5 are refractory to the suppressive effects of Tregs. Understanding the mechanisms that influence the balance between mucosal homeostasis and the induction of adaptive immune responses has important implications for mucosal vaccine design and delivery strategies. Diverse populations of respiratory DCs have been characterized in both normal and inflamed lungs (88). A recent study demonstrated that in response to influenza infection, two distinct subsets of migrant respiratory DCs expressing CD103+ and CD11b+ were involved in presenting Ag to naive CD4+ and CD8+ T cells in the draining lymph nodes (88). However, airway-derived migrant CD103+ DCs in the inflamed MLNs were superior at cross-presenting influenza virus Ags to naive CD8+ T lymphocytes (88). Similarly, only the migratory subtype of CD103+ skin-derived DCs efficiently presented herpes simplex virus type 1 Ags for the induction of naive CD8+ T cells (89). DCs can be recruited to the intestinal LP as a result of local inflammation.
A recent study identified a population of CD11chighCD11bhigh LP DCs that express TLR5 and produce proinflammatory cytokines such as IL-6 and IL-12, but not IL-23 or IL-10, in response to flagellin (90) (Fig. 2). Different DC populations are also responsible for priming T cells dependent upon the route of HSV-1 infection (78). Following vaginal infection, HSV-1 Ags were largely presented by tissue-derived migratory DCs, whereas after needle injection HSV-1 was rapidly presented by lymph node-resident DCs to rare Ag-specific T cells (78). Understanding the unique properties of mucosal DCs and the mucosal milieu in regulating local T and B cell responses and “mucosal memory” will be critical in determining how best to deliver vaccines for protection from mucosal infection. Studies in diverse infection models involving different mucosal compartments have clearly shown that targeting local mucosal DCs and the induction of local immune responses by mucosal vaccination is more effective in containing mucosal infections, as our understanding of the magnitude and quality of responses and the tissue residency of cells migrating to mucosal sites following systemic immunization needs further investigation. Developing mucosal immunization strategies will depend upon further understanding of how best to induce a compartmentalized immune response tailored to the specific pathogen and calibrated to the potential pathogenic threat.
Th17 cells and mucosal protection
Th17 cells have emerged as a unique mucosal CD4+ effector cell population responsible for protection from extracellular bacteria and fungi. Effector function is attributed to the interaction of its signature cytokine IL-17 with IL-17R expressed on fibroblasts and epithelial cells to induce chemokines such as MCP-2 and G-CSF and CXC chemokines. Enhanced recruitment of neutrophils and granulopoiesis leads to pathogen clearance, whereas CXCR-3 ligating chemokines can activate macrophages and recruit additional immune cells, including CD4+ Th1 cells. IL-22 is another important effector cytokine produced by Th17 cells for inducing the secretion of antimicrobial peptides and β-defensin-2 by epithelial cells and for contributing to barrier function and tissue repair (Fig. 2). In the gut, Th17 cells are the main sources of IL-17, whereas in the lung γδT cells, NKT, NK, and RORγt NKp46+ cells are the major producers of this cytokine. The rules governing Th17 differentiation in vivo reflect tissue localization and the balance of inflammatory and anti-inflammatory cytokines induced by pathogen-activated DCs. In the PPs and MLNs, DC production of TGF-β and RA plus IL-2 leads to the preferential induction of induced Treg (iTreg). TGF-β and IL-10 are pivotal cytokines in the control of inflammation induced by TLR activation, suppressing the MyD88 signaling pathway and type I IFN production. Pathogens that induce robust tissue inflammation and proinflammatory cytokines can shift this balance toward differentiation of Th17 cells (Fig. 1). Human CD4 cells most efficiently differentiate into Th17 cells by a combination of TGF-β and IL-6 plus IL-1β or IL-β plus IL-23 or IL-21. IL-6-STAT3 signaling is critical for Th17 differentiation, inducing expression of two transcription factors, RORγτ and RORα, and the IL-23R. IL-21 and IL-23 also signal through STAT3, stabilizing the Th17 phenotype and amplifying effector function by decreasing IL-10 production and increasing trafficking. Patients with an autosomal dominant STAT3 mutation (hyper-IgE syndrome/Jobs syndrome) are deficient in Th17 cell differentiation and suffer mucocutaneous Candida and recurrent pyogenic infections.
Although Th17 inflammatory responses have been shown to play a key role in both protection and pathology to Gram-negative extracellular bacteria and fungi as well as autoimmune diseases, their role in enforcing mucosal barrier function and protective immunity to intracellular bacteria and viral infections is beginning to be appreciated. A recent study showed that strong influenza-specific Th17 responses develop in IL-10-deficient mice and can provide protection from lethal challenge by mechanisms independent of Th1 helper functions (perforin-mediated cytotoxicity and IFN-γ) or IL-17A (91). Another study suggests that CCR6 α4β7high Th17 central memory cells in gut mucosa are the initial population of cells infected following SIV infection. This result is not surprising considering that tissue-resident Th17 cells in the small intestine LP comprise ∼80% of the total CD4 cells and express CCR6 and high levels of CCR5. Furthermore, depletion of this cell population in the gut mucosa is predictive of subsequent disease pathology (92). It will be important to determine whether mucosal vaccine strategies that induce HIV specific-Th17/Th1 effector memory responses can reverse the outcome of CD4+ Th17 α4β7highCCR5high cells as targets of HIV infection into aggressive effector cells capable of contributing to a protective response.
Similarly as in the reciprocal transcriptional regulation of iTregs and Th17 differentiation by TGFβ, Th17 cells subsequently share a common branch point with T follicular helper cells (Fig. 2). Although both subsets rely on autocrine IL-21 for their differentiation, lineage commitment might be dependent on additional signals delivered by DCs such as IL-1β and IL-23 or ICOSL to determine CCR6 Th17 or CXCR5 T follicular helper cell fate. Thus, Th17 and Treg cells are the main modulators of the balance between tolerance and inflammation in mucosal tissue and are involved both in primary infection and the recall responses seen in vaccine studies. These cells might be exploited to develop the most effective mucosal vaccines against pathogens causing chronic infections such as HIV, HCV, herpes viruses, mycobacteria, and fungi.
New design and delivery strategies for eliciting local mucosal immunity are urgently needed for developing effective prophylactic and therapeutic vaccines against mucosal pathogens such as HIV, HPV, influenza, herpes viruses, and M. tuberculosis, as well as pathogenic enteric and fungal infections. Many vaccinologists feel that parenteral immunization is sufficient, given the history of successful vaccination in humans. However, numerous studies in human clinical trial and experimental settings have demonstrated that mucosal vaccine strategies are more effective in the prevention of pathogen transmission and the resolution of infection at the site of initial pathogen entry. An optimum mucosal vaccine strategy can activate local innate immune responses that translate into the generation of neutralizing sIgA Abs, polyfunctional CD8+ CTLs in effector target sites, and CD4+ T cell effector memory responses, whereas systemic vaccine strategies are much less effective in eliciting the compartmentalized mucosal immunity necessary for these protective responses. Targeting Ags to mucosal DC that imprint gut-homing receptors on pathogen-specific T and B cells and the induction of Th17/Th1 responses will require novel mucosal vaccine strategies, including effective adjuvants and immunomodulatory molecules. Compared with adjuvants for parenteral vaccination, mucosal adjuvants today are much less effective. Advances in new design and delivery strategies and novel mucosal adjuvants will greatly improve the efficacy of mucosal vaccination and contribute to the broad implementation of effective mucosal vaccines in the future.
We thank Drs. David Franz, Niveen Mulholland, Kenton Lohman, and John Lednicky (Midwest Research Institute, Frederick, MD) for critical reading of the manuscript and helpful suggestions. We apologize to the authors whose work we were unable to include in our citations due to space limitations.
The authors have no financial conflict of interest.
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
Abbreviations used in this paper: HPV, human papillomavirus; ASC, Ab-secreting cell; BCG, bacillus Calmette-Guérin; CMIR, compartmentalized mucosal immune response; CMIS, common mucosal immune system; DC, dendritic cell; i.n., intranasal; IPV, inactive poliovirus vaccine; iTreg, induced Treg; LP, lamina propria; MLN, mesenteric lymph node; OPV, oral poliovirus vaccine; RA, retinoic acid; SHIV, simian/human immunodeficiency virus; sIg, secretory Ig; TB, tuberculosis; Treg, regulatory T cell; VLP, virus-like particle.