John F. Timoney, MVB, PhD, DSc, Keeneland Professor of Infectious Diseases,
Gluck Equine Research Center, University of Kentucky, Lexington, KY 40546-0099
Intranasal vaccination has now emerged as a practical approach to the prevention of equine respiratory infections in N. America; modified live intranasal vaccines against strangles and equine influenza have become available since 1998 and owe their existence to evidence that stimulation of protective mucosal immune responses does not result from parenterally inoculated vaccines but rather requires local induction. Moreover, mucosal immunization often induces a combination of systemic and local responses associated with production of a greater variety of immunoglobulin subisotypes and specificities that closely mimic those induced by natural infection. Finally, vaccines administered intranasally are likely to produce fewer adverse reactions than when given parenterally and with a lower probability of generating a non-functional isotype or high affinity antibody to non-protective epitopes.
Potential mucosal immune induction sites in the equine nasopharynx
Nasal and oropharyngeal tonsillar tissue has been described as the gatekeeper to mucosal immunity because of its location at the entrance to the respiratory and alimentary tracts. These tissues in the horse consist mainly of the nasopharyngeal tonsils located in the dorsal nasopharyngeal recesses and the palatine and lingual tonsils. Less well defined aggregates of follicular lymphoid tissue are also distributed throughout the nasopharyngeal mucosa. Since the nasopharyngeal tonsil is potentially an important immune induction target for intranasally administered vaccines it has been more closely studied (1,2). In brief, its significant features are a ciliated epithelium that merges with patches of follicle associated antigen sampling epithelium with M cells that overlie lymphoid tissue consisting of primary or secondary follicles populated with immunoglobulin producing lymphocytes and dome and parafollicular areas with CD4 and CD8 positive lymphocytes (Figure 1). CD4 cells also form densely packed clusters in the subepithelial lamina propria and interfollicular area and probably constitute a population of naive and memory T cells which provide help for B cells to differentiate and synthesize immunoglobulin. Intraepithelial lymphocytes are predominantly CD8 but include some B lymphocytes that are actively extruded into surface secretions where they release IgA, IgG and IgM. IgA followed by IgGb are quantitatively the most important subisotypes in nasopharyngeal secretions of adult horses; IgA is absent from those of the foal in the first months of life (3). The lingual and palatine tonsils differ from the nasopharyngeal tonsil in having parakeratinized or stratified squamous non-keratinized epithelia respectively (4). Deep crypts increase the available epithelial surface as much as sixfold. Studies on human nasopharyngeal and palatine tonsils have demonstrated direct transepithelial access of antigens from the crypt epithelium to underlying lymphoid tissue. In laboratory rodents some soluble antigens that enter the entire nasopharyngeal epithelium induce a systemic response or tolerance. Particulate antigens that enter the nasopharyngeal tonsil are more likely to elicit a secretory IgA response (1). It should be noted that the lingual and palatine tonsils are preferred sites of attachment of Streptococcus equi including the avirulent Pinnacle vaccine strain.
Figure 1. Schematic organization of the equine nasopharyngeal tonsil. LE = lymphoepithelium; FAE = follicle associated epithelium; ¤ = T lymphocytes (CD4); = T lymphocyte (CD8); ? = B lymphocyte; G = seromucous glands. Courtesy Dr. Pawan Kumar.
The mucosa of the equine nasopharynx has a complex secretory immune system that responds to microbial infection by synthesis of specific antibody. The best studied examples of these infections are strangles and equine influenza (5,6). Mucosal IgA responses to the antiphagocytic SeM protein of S. equi begin about 2 weeks after infection, peak about 3 weeks later and persist for 3 to 6 months. Some local SeM specific IgGa and IgGb are also produced but levels decline rapidly in 3 to 4 weeks. The epitope specificities of locally produced immunoglobulins may be different than those of serum immunoglobulins – a not unexpected finding given the well documented independence of mucosal and systemic antibody responses. Recovery from strangles is associated with immunity to reinfection that persists in approximately 70% of animals for 4 years or longer. Horses challenged shortly after recovery do not make anamnestic responses to SeM suggesting that convalescent immunity involves immune exclusion. Interestingly, mucosal responses to S. zooepidemicus in the tonsils do not include antibody protective against S. equi challenge although the DNA’s of the 2 organisms are almost identical.
Mucosal IgA responses to influenza virus peak on day 16 after infection and return to baseline by 60 days (5). Low levels of IgGa and IgGb are also present after primary infection for about 3 months. Horses recovered from natural infection were resistant to challenge 3 months later. In contrast, horses vaccinated with a parenterally administered vaccine made only IgG(T) and IgG(C) responses and were not protected. Nevertheless, these horses were apparently primed and made strong mucosal IgA responses subsequent to challenge.
Inductive sites in the nasopharynx are an attractive target for new generation vaccines against respiratory pathogens because of their accessibility, and the absence of chemical barriers such as acidity and hydrolytic enzymes which may degrade immunogens administered per os to intestinal mucosal induction sites. However, an intranasal vaccine must be presented in a manner that will circumvent mucociliary clearance in the nasopharynx and resist mechanical scouring of the lingual and palatine tonsillar surfaces that accompanies swallowing. The vaccine must not be affected by naturally occurring polyreactive IgA and must not induce tolerance. It should induce a memory B and T lymphocyte response. A number of different intranasal delivery systems have been evaluated in the horse including microparticle encapsulation, mucoadhesive carriers, cholera toxin chimeras and avirulent salmonellas.
Microparticle encapsulation: Polylactide microspheres have been shown in laboratory animals to be effective carriers for intranasal delivery of peptides to the mucosal and systemic immune compartments (7,8). Microencapsulation involves coating of antigen with a biodegradable polymer such as poly DL-lactide-co-glycolide so that particles less than 5 m are formed. Particles of this size are endocytosed by M cells and, after antigen release in associated lymphoid tissue, elicit mucosal and systemic antibodies. However, ponies immunized with 250 g of an immunogenic peptide (SeMF3) of the SeM protein of S. equi on days 0, 7 and 42 made no detectable serum antibody. SeM specific mucosal IgA responses were detected in 2 vaccinates on day 21 and in all 3 on day 49. The absence of a systemic response may have been due to failure of release by antigen from the mucosal compartment. Lack of potency has been frequently associated with microencapsulation as a mode of mucosal antigen delivery and so the technology has not been widely embraced (8). An important theoretical advantage of microparticles is that, unlike living vectors, they do not elicit immune responses to themselves and so repeated boostering should be possible.
Mucoadhesive compounds: Bioadhesive polymers such as sodium alginate have been used successfully for mucosal drug delivery and oral application of killed influenza vaccine (8). Strong mucosal and systemic antibody responses were obtained in this way. Non-polymeric (food grade) sucrose acetate isobutyrate (SAIB) is a highly viscous excipient which becomes much less viscous in appropriate solvents including ethanol in which it can be easily aerosolized. After deposition on the nasopharyngeal mucosa, the solvent evaporates and the SAIB and dissolved peptides form a sticky film which degrades very slowly. Thus there is prolonged exposure of the mucosa to antigen trapped in the film. We have used SAIB to stimulate mucosal and systemic antibody to SeMF3 (9). Adult Thoroughbred mares were inoculated intranasally with 200 g SeMF3 in 2 mls 75:25 SAIB-ethanol solution on day 0 and again 28 days later. Control mares were given 200 SeMF3 alone. Ninety-one percent of mares vaccinated with SeMF3-SAIB had made SeM specific serum and mucosal IgA by day 42. Control mares made only short-lived mucosal and no serum responses. SAIB was therefore highly effective in amplifying mucosal and systemic antibody responses to an intranasally administered immunogen. Interestingly, serum and mucosal antibodies of the same mares showed different patterns of recognition consistent with the conclusion that SAIB was effective in delivery of SeMF3 to both the mucosal and systemic immune compartments.
Choleratoxin-SeMF3 chimeras: Cholera toxin (CT) is a potent mucosal immunogen and adjuvant. The pentameric B (CTB) subunit binds to GM1 gangliosides on epithelial cells and M cells thereby enhancing its own uptake by mucosal associated lymphoid tissue. The 5 kDa A2 subunit is linked to the central hole formed by the five B subunits and so is potentially a convenient means of attaching a foreign peptide sequence to the B pentamer (8). Based on this concept and using recombinant DNA technology we have constructed a CT-SeMF3 chimera that contains the entire ctB gene and part (A2) of the A sequence encoding the non-toxic region of the A subunit. One group of ponies was immunized intranasally on days 0, 14, 42, 70 and 126 with purified CT-SeMF3 and a second group with CTB-A2 alone on days 0, 21 and 56. All ponies vaccinated with CT-SeMF3 made strong CTB and SeMF3 specific serum IgGb responses after the first immunization but did not respond to subsequent immunizations as antibody levels slowly declined (Figure 2). In contrast, SeMF3 specific mucosal IgA responses varied in time of onset for each pony and were boostable in most ponies. Mucosal responses were delayed in onset relative to those detected in serum and did not attain the levels seen following strangles. These unexpected limitations of the CT-chimera approach in the horse are as yet unexplained and are at variance with reports of experiments with other constructs in laboratory rodents.
Figure 2. SeMF3 specific serum (A) and nasal wash (B) immunoglobulin subisotype responses of 5 ponies immunized intranasally with CT SeMF3. The arrows on the abscissa indicate immunization dates. Courtesy Dr. Abhineet Sheoran.
Avirulent salmonella vectors: Avirulent mutants of salmonellas are attractive mucosal vaccine delivery systems because they stimulate both the mucosal and systemic immune compartments by penetrating M cells of the Peyer’s patches and then migrating to the spleen and liver (8). We have demonstrated that intranasal inoculation of ponies with a cya crp mutant (MGN 707) of Salmonella typhimurium safely elicits strong specific mucosal and systemic responses. Based on these results and using recombinant DNA technology we prepared a construct of MGN 707 containing a plasmid with temperature regulated expression of the SeMF3 peptide. Ponies were inoculated intranasally on days 0, 35, 56 and 161 with 6 x 109 cfu MGN 707 pSeMF3 in which expression of SeMF3 had been induced by raising the culture temperature prior to administration. These ponies made strong SeM specific serum responses and delayed but strong mucosal IgG responses. In contrast, ponies immunized with non-induced salmonellas made no detectable mucosal responses although their serum responses to SeM were similar to those of ponies vaccinated with preinduced salmonellas. This suggests that the vectored peptide must be expressed and displayed at the time of administration for induction of antigen specific antibody on the nasopharyngeal mucosa. Both serum and mucosal responses to salmonella antigens and SeMF3 were boostable allaying the concern raised by other studies in rodents that preexisting antibody would result in immune exclusion of salmonellas delivered in the booster inoculum (10).
1. Kuper, C.F., Koornstra, P.J., Hameleers, D.M.H., Biewenga, J., Spit, B.J., Dijvestyn, A.M., van Breda Vriesman, P.C. and Sminia, T. 1992. The role of nasopharyngeal lymphoid tissue. Immunology Today. 13:219-224.
2. Kumar, P., Timoney, J.F. and Sheoran, A.S. 2000. M cells and associated lymphoid tissue of the equine nasopharyngeal tonsil. Eq. Vet. J. In press.
3. Sheoran, A.S., Timoney, J.F., Holmes, M.A., Karzenski, S.S. and Crisman, M.V. 2000. Immunoglobulin isotypes in sera and nasal mucosal secretions and their neonatal transfer and distribution in horses. Am. J. Vet. Res. 61:1099-1105.
4. Perry, M. and Whyte, A. 1998. Immunology of the tonsils. Immunology Today. 19:414-421.
5. Sheoran, A.S., Sponseller, B.T., Holmes, M.A. and Timoney, J.F. 1997. Serum and mucosal antibody isotype responses to M-like protein (SeM) of Streptococcus equi in convalescent and vaccinated horses. Vet. Immunol. Immunopathol. 59:239-251.
6. Nelson, K.M., Schrun, B.R., McGregor, M.W., Olsen, C.W. and Lunn, D.P. 1998. Local and systemic isotype-specific antibody responses to equine influenza virus infection versus conventional vaccination. Vaccine. 16:1306-1313.
7. Almeida, A.J. and Alphar, H.O. 1996. Nasal delivery of vaccines. J. Drug Targeting. 3:455-467.
8. Mestecky, J., Moldoveanu, Z., Michalak, S.M., Morrow, C.D., Compans, R.W., Schafer, D.P. and Russell, M.W. 1997. Current options for vaccine delivery systems by mucosal routes. J. Controlled Release. 48:243-257.
9. Nally, J.E., Artiushin, S., Sheoran, A.S., Burns, P.J., Simon, B., Gilley, R.M., Gibson, J., Sullivan, S. and Timoney, J.F. 2000. Induction of mucosal and systemic antibody specific for SeMF3 of Streptococcus equi by intranasal vaccination using a sucrose acetate isobutyrate based delivery system. Vaccine 19:500-505.
10. Roberts, M., Bacon, A., Li, J. and Chatfield, S. 1999. Prior immunity to homologous and heterologous Salmonella serotypes suppresses local and systemic anti-fragment C antibody responses and protection from tetanus toxin in mice immunized with Salmonella strains expressing fragment C. Infect. Immun. 67:3810-3815.