Novel vaccination strategies for the control of mucosal infection

Novel vaccination strategies for the control of mucosal infection A l a n J. H u s b a n d

Enteric' disease remains one of the 9reatest causes of mortality and morbidity in both human and reterinar)' species. There has been a remarkable lack of success in vaccination to control mucosal disease and it is therefore apparent that novel strateqies are required to achieve effective mucosal immunity. Basic studies described in this paper have addressed problems associated with antiqen handlin9 and the induction of an immune res'ponse in the intestine, and the subsequent dissemination o f effector cells and molecules to intestinal and extra-intestinal submucosal re(lions. Effective induction o f IqA responses is dependent on T-cell help and requires coqnate interactions between T cells" and B cells" within orqanized 9ut-associated lymphoid tissue (GALT). The deliver)' oj" an IrA antibody response to mucosal sites is also a T cell dependent but antigen driven proeess. The normal route by which antiqen is taken up by G A L T is via the epithelial surface but antiqen presented in this" way via villus epithelial cells 9enerates predominantly a suppressor response. Strateqies desiqned to overcome this e~ect include the use ¢?f powerful adjuvants (such as cholera toxin, muramyldipeptide and phorbol esters), the use o f immunoqenic carriers, or delivery via liposomes, microspheres or 9enetically enqineered viral or bacterial vectors. Alternatively, the feasibili O, ~?~accessin 9 GAL T via the serosal surface by administration o f in traperitoneal antiqen in oil emulsion has been explored and a vaccine formulation (Auspharm (patent pendinq)) has been developed which is' suitable for IP delivery in commercial applications. Keywords: Mucosal vaccination: IgA antibody; adjuvants; intraperitoneal immunization INTRODUCTION Enteric disease remains one of the greatest causes of mortality and morbidity in both human and veterinary species but diseases of other mucosal sites (particularly the respiratory tract, urogenital tract and breast) also create significant disease burden. The continual ingestion and inhalation of potentially pathogenic material represents a barrage of antigens against which the intestinal immune defence network of the newborn must mount rapid and effective defence. This is of particular relevance in infants where passive protection may be provided via placental transfer of antibody or by ingestion of colostrum after birth, but this provides only transient defence and the neonate remains at risk of infection via these surfaces particularly during the postweaning period unless active immunity can be established. Although immunization strategies have been available for many years and have achieved significant success in reduction of disease incidence and mortality when applied to diseases affecting systemic organs, there has been a remarkable lack of success in vaccination to control mucosal disease. It is therefore apparent that novel strategies are required to achieve effective defence at these sites. Mucosal surfaces, particularly the intestine, Department of Veterinary Pathology, University of Sydney, NSW 2006, Australia 0264410X/93/020107q36 ~ 1993 Butterworth-HeinemannLtd

represent a physiological dilemma whereby selective absorption of essential nutrients must occur with concomitant exclusion of potential pathogens. For this reason the mucosal defence system has evolved as a specialized branch of immunity with a peculiar set of effector elements. In particular, the IgA class of immunoglobulins characterizes the antibody response at mucosal surfaces, and a unique transport mechanism has evolved to ensure their selective secretion into mucosal surfaces. However, attempts to achieve IgA responses in the intestine by oral immunization with non-replicating antigens have been characterized by ineffective responses of short duration unless long-term dosages are administered 1.2 This paper will review mechanisms of antigen absorption and presentation by the intestinal immune system, the factors determining the response to antigen encountered at this site and novel strategies which may be employed to optimize these responses in the face of inherent downregulatory mechanisms.

INDUCTION RESPONSES

OF MUCOSAL

IMMUNE

The normal route by which antigen is taken up by the gut-associated lymphoid tissue ( G A L T ) is via the epithelial surface. The predominant site for antigen uptake in immunogenic form is through the modified

Vaccine, Vol. 11, Issue 2, 1993 107

Vaccination in mucosal infection control." A.J. Husband

epithelium overlying the Peyer's patches 3 transported through M cells to underlying dendritic and lymphoid cells 4. It is now well established that the effector cells produced in response to mucosal antigenic stimulation arise from Peyer's patches although it is also possible that minor aggregations of lymphoid tissue in the lamina propria contribute to this response. Antigen taken up via M cells is processed and presented by subepithelial dendritic ceils in association with MHC Class II to T-helper cells which provide isotype-specific help for IgA production 5 ~. It has been recognized for some time that antigen-specific T-cell help occurs in Peyer's patches s and that IgA responses are highly T-cell dependent 9 ~1 Thus in addition to providing antigen in appropriate site, form and with appropriate adjuvants, it is necessary to provide both B- and T-cell activation signals. In this regard we have recently described the requirement for cognate T-cell help in induction of lgA responses using a hapten carrier immunization system. By administration of hapten and carrier to Peyer's patches in different regions of the intestine, we have demonstrated the requirement for both carrier primed T cells and hapten primed B cells in the induction of an IgA antibody response to hapten ~2. However, it has also been demonstrated more recently that antigen presented via normal villus epithelial cells in the intestine triggers a predominantly suppressor response. Bland and Warren ~3 described the production of antigen-specific suppression by T cells when activated by presentation of antigen in association with normal intestinal epithelial cells in rats, which was associated with an increase in expression of the suppressor phenotype. This was confirmed in humans by Mayer and Shlien ~4 who demonstrated that, in addition to the generation of T cells of suppressor phenotype after antigen presentation by epithelial cells, non-antigenspecific suppression was also observed. Subsequent work by Bland and Whiting ~5 suggests that part of the explanation for this predisposition towards suppression by epithelial processing by epithelial cells compared with macrophages. Taken together these findings indicate that antigen presentation via the mucosal surface is an inefficient means of response stimulation. Consequently, to generate effective and protective immune responses by this route, novel strategies are required. One such strategy has been to incorporate powerful adjuvants such as cholera toxin, muramyldipeptide and phorbol esters into vaccine formulations. For example Nedrud and Sigmund 16 have shown that cholera toxin promotes the activation of IgG and IgA antigen-specific responses providing the antigens administered in this context are novel to the host. This appears to operate through the combined ability of the cholera toxin B subunit to bind to the GM-1 ganglioside 17 and the ability of the A subunit to inhibit IL-2-dependent T-cell proliferation by blockingT-cell receptor-mediated transmembrane signalling through a cyclic AMP independent pathway ~8. Other strategies have included the incorporation of muramyldipeptide ( M D P ) in antigen preparations TM. M D P is a purified peptidoglycan moiety of the cell wall of mycobacterium which is the immunostimulating component of mycobacterial cells and which has been shown to provide enhanced T-cell help 2°. Protein kinase C activating phorbol esters have also been shown to provide selective induction of high levels of IgA synthesis in Peyer's patch

108

Vaccine, Vol. 11, Issue 2, 1993

cells 21 and this may provide a new mechanism by which adjuvant effects may be acl,ieved in novel formulations. Presentation of antigen via genetically engineered mucosal binding viral or bacterial vectors with reconstructed genetic potential to produce antigens of desired specificity are also a promising avenue of investigation for mucosal immunostimulation. Salmonella mutants with reduced virulence and gene insertions which code for protective antigens from other pathogens have received considerable attention as potential vectors 22'23. Vaccine vectors can also be constructed to express selected cytokines in addition to antigenic peptides. Recombinant vaccinia virus co-expressing IL-5 and influenza virus haemagglutinin (HA) promoted enhanced IgA-specific HA responses in the respiratory tract 24. This suggests the exciting possibility that micro-environments conducive to IgA B-cell proliferation can be created by providing a source of interleukins additional to those occurring by T-cell activation. An alternative strategy is to modify the physical form of antigen to encourage activation of mucosal associated lymphoid tissue. This can be achieved using microsphere technology whereby antigens are incorporated into microcapsules (constructed for instance with a poly(D,Llactide-coglycolide) microsphere) which protects antigen from gastric degradation allowing selective absorption by M cells in the Peyer's patches 25. Linking peptide molecules to high molecular weight protein carriers has also been successful in promoting a mucosal response to otherwise poorly immunogenic molecules 2°. Combinations of these strategies are likely to yield promising results in future applications 2~ 29.

ROUTE OF PRESENTATION ANTIGEN

OF MUCOSAL

The above discussion has focused on problems of stimulation of an IgA response via the mucosal surface, which is complicated by the inherent production of suppressor factors via epithelial processing. It is apparent, however, that Peyer's patch sites may also be accessed via the serosal surface if antigen is administered via the IP route. Early studies by Pierce and Gowans 3° suggested the potential for IP immunization in stimulating an intestinal IgA antibody response. These experiments were conducted using Freund's complete adjuvant which created antigen depots in the mesentery and on the serosa and presumably allowed antigen access to Peyer's patches by establishing peritoneal inflammation and increasing serosal permeability. Further studies by Husband and Gowans 31 verified that this response was dependent on the presence of Peyer's patches in immunized segments of intestine. In a series of subsequent experiments, also using Freund's complete adjuvant administered IP, effective intestinal IgA responses were achieved in sheep and pigs 3z-34. However, Freund's adjuvant is generally unacceptable for IP administration, causing intense peritonitis and persistent mesenteric lesions. An alternative formulation was sought to produce a similar effect and in experiments described below, several adjuvants and vaccine vehicle preparations have been assessed for IP use. Two adjuvants, saponin 3~ and MDP 19, chosen on the basis of their availability, ease of preparation and demonstrated adjuvanticity in systemic immunization

Vaccination in mucosal infection control: A.J. Husband

applications, were formulated in either vegetable oil emulsion or liposome vehicles, representing biodegradable alternatives to the mineral oil component used in Freund's adjuvant. These formulations were used to deliver ovalbumin (OVA) IP inrats and the subsequent anti-OVA-containing cell (AOCC) response in the intestine was assessed. Lyophilized OVA was dissolved in phosphate buffered saline (PBS; pH 7.3 ) and, with the exception of liposome preparations, the concentration was adjusted such that all vaccine formulations contained 500/~g OVA/IP dose. For liposome preparations the aqueous phase contained 10mg/ml OVA. The adjuvants tested were added to OVA solutions as follows. Saponin was dissolved in dimethylsulphoxide and added to aqueous phase OVA to provide a final concentration of 2 mg/ml saponin and 10% v/v dimethylsulphoxide; MDP (N-acetylmuramylL-alanyl-D-isoglutamine, Sigma Chemical Co., St Louis, MO) was added to OVA solution to a final concentration of 1 mg/ml. Stable oil-in-water emulsions were prepared Table 1 Intestinal AOCC response in rats following IP administration of OVA with various adjuvant and vehicle preparations followed by ID challenge with OVA

IP OVA OVA OVA OVA OVA OVA OVA

+ CFA + + + + +

OIL SAP MDP SAP + OIL MDP + OIL

ID

AOCC/cm

%lgA

OVA OVA OVA OVA OVA OVA OVA

102.3 4.1 6.4 28.7 43.3 53.6 80.2

79.7 -+ 2.1

_+ 5.4 -+ 1.5 -+ 3.6 _+ 9.1 _+ 5.9 _+ 14.4 _+ 3.9

60.0 57.0 30.8 82.7

_+ 5.4 _+ 14.5 _+ 9.54 _+ 1.59

Number of rats 13 13 9 9 4 4 12

Protocol: IP 14days ID 5days Kill Values are mean _+ standard error of data from the number of animals indicated. CFA = complete Freund's adjuvant; O I L - vegetable oil; SAP = saponin; MDP - muramyldipeptide

Table 2 Intestinal AOCC response in rats following IP administration of OVA in vegetable oil emulsion with SAP or MDP adjuvants followed by ID challenge with OVA + MDP

IP

ID

AOCC/cm

% IgA

Number of rats

OVA+SAP+OIL OVA+MDP+OIL

OVA+MDP OVA+MDP

110.0_+10.5 123.4_+ 8.9

86.9_+3.6 65.6___1.3

6 9

Protocol: IP 14days ID 5days Kill Values are mean ± standard error of data from the number of animals indicated. OIL - vegetable oil; SAP = saponin; MDP muramyldipeptide

by addition of equal volumes of vegetable oil and adjuvanted protein solutions with emulsifier. Immunization protocols were based on those previously established using OVA in Freund's adjuvant by injection of 0.5 ml of vaccine into the peritoneal cavity of rats. Intraduodenal (ID) challenge was performed 14 days after IP immunization by injection of 0.5 ml OVA (10mg/ml) directly into the lumen of the duodenum after its exposure by laparotomy. Rats were killed 5 days later and tissues collected for immunofluorescent histology. Liposomes were prepared incorporating both MDP and saponin with OVA in the aqueous phase using the method of Shek and Sabiston 36. Lipid phase, consisting of dipalmitoylphosphatidylcholine,cholesterol and phosphatidic acid in molar ratios of 2: 1.5:0.2, was dried on to flasks using a rotary evaporator. Aqueous phase (containing OVA 10 mg/ml, MDP 1 mg/ml and saponin 2 mg/ml in PBS) was entrapped in lipid vesicles by vigorous shaking for 5 min after addition to the flask. Liposomes were washed twice in PBS by centrifugation before incorporation into the vegetable oil emulsion. The results in Table 1 show that both saponin and MDP have adjuvant activity for IgA responses to OVA when given IP, especially if delivered in an oil emulsion, but that MDP is marginally superior in this regard. Further, whereas rats given OVA in Freund's complete adjuvant had extensive mesenteric granulomatous lesions with multiple adhesions, none of the animals given any of the other vaccine formulations had any evidence of peritonitis or lesions and their mesenteries appeared normal in all respects. Since MDP has also been used effectively as an oral adjuvant for IgA responses 3v, further experiments were undertaken to determine whether the responses obtained with MDP or saponin delivered by the IP route in oil emulsions could be improved if MDP was also added to the ID challenge. The data in Table 2 indicate that greatly enhanced AOCC responses were achieved by these treatments which exceeded those obtained in rats immunized IP with OVA in Freund's adjuvant and subsequently challenged with OVA in PBS. Again, the rats immunized IP with vaccine containing MDP responded marginally better than those given the saponin IP vaccine, although in this case the IgA component was reduced. While these results show that both MDP and saponin given in vegetable oil emulsion provide an alternative to Freund's adjuvant for IP stimulation of IgA responses, the success of liposomes in systemic antigen delivery29'3s'39 suggested their potential as an alternative vehicle for IP use. Animals were immunized IP with OVA-containing liposomes and challenged ID with OVA, OVA + MDP

Table 3 Intestinal AOCC response in rats following IP administration of OVA in liposomes or vegetable oil emulsion followed by ID challenge with OVA, OVA + MDP or OVA liposomes

IP

ID

AOCC/cm

%lgA

OVA/Liposomes OVA/Liposomes OVA/Liposomes OVA + MDP + OIL

OVA OVA + MDP OVA/Liposomes OVA/Liposomes

17.8 22.4 18.4 124.9

36.7 33.9 43.6 79.3

4- 5.7 4- 6.3 _+ 5.4 4- 13.9

± -+ -+ 4-

Number of rats 5.7 4.6 3.2 0.9

5 4 5 4

Protocol: IP 14days ID 5days Kill Values are means _+ standard error of data from the number of animals indicated. OVA liposomes were constructed containing both MDP and SAP adjuvants. OIL - vegetable oil; SAP = saponin; MDP = muramyldipeptide

V a c c i n e , Vol. 11, I s s u e 2, 1993

109

Vaccination in mucosal infection control: A.J. Husband

in aqueous solution, or OVA + M D P and saponin in liposomes. The results in Table3 indicate that the IgA-specific AOCC response in the intestine was small in all animals primed IP with liposome formulations regardless of the nature of the ID challenge. Liposomes did, however, appear to have a role in mucosal antigen delivery since animals immunized IP with OVA + M D P in oil emulsion but challenged ID with OVA-containing liposomes produced an outstanding response, greater than that achieved by OVA + Freund's adjuvant and equivalent to that obtained in animals primed and challenged with OVA + MDP. STUDIES WITH LARGE ANIMALS These studies demonstrated that delivery of antigen via the IP route could be achieved using a formulation based on vegetable oil which was tolerated well and caused no mesenteric lesions but retained adjuvanticity for an IgA response. This encouraged experiments to be undertaken in large animals to determine the practicability of IP delivery for stimulation of mucosal responses. A formulation was developed based on the above findings incorporating antigen plus bacterial-derived adjuvant in a vegetable oil emulsion. This was developed in collaboration with Auspharm International Ltd. who hold the patent to its formulation. OVA was incorporated in the Auspharm (patent pending) adjuvant and the response following IP immunization was compared with OVA in Freund's adjuvant with respect to the stimulation of an IgA-specific AOCC response in the intestine of sheep. For the purpose of these scaling-up experiments sheep were immunized with 10 ml of vaccine consisting of 5 ml OVA solution emulsified in an equal volume of Auspharm (patent pending) adjuvant. In these experiments it was observed that a single IP dose without I D challenge as sufficient to produce a substantial lgA-specific AOCC response in the intestine (Table 4). The Auspharm (patent pending) formulation produced an AOCC response equivalent in magnitude to that observed with OVA + Freund's adjuvant but with an elevated proportion of AOCC of the lgA isotype. In further scaling-up studies pigs were immunized with OVA in Freund's adjuvant or OVA in Auspharm (patent pending) adjuvant by the IP route and 14 days later given OVA with or without DEAE-dextran given ID. The results in Table 5 indicate that the Auspharm (patent pending) adjuvant delivered IP in pigs established a substantial IgA-specific AOCC response in the intestine although not as great as that observed with Freund's adjuvant. Based on the understanding that the intestine forms the key to mucosal defence at remote sites such as the Table 4 Intestinal AOCC response in sheep following IP administration of OVA in CFA compared with OVA in Auspharm' adjuvant followed by ID challenge with OVA

IP

ID

AOCC,'cm

%lgA

Number of sheep

OVA 4- CFA OVA + Auspharm'

OVA OVA

102.3 ÷ 5.4 77.5 -- 5.7

79.7 + 2.1 82.7 + 1.59

13 4

Protocol: IP 14days ID 5 d a y s Kill Values are mean + standard error of data from the number of animals indicated. CFA - complete Freund's adjuvant

110

V a c c i n e , Vol. 11, I s s u e 2, 1993

Table 5 Intestinal AOCC response in pigs following IP administration of OVA in CFA compared with OVA in Auspharm' adjuvant followed by oral challenge with OVA with or without DEAE-dextran IP Day 0

Oral Day 14

AOCC/cm

% IgA

No. pigs

OVA+FCA O V A + FCA OVA+Auspharm' OVA+Auspharm'

OVA OVA + DEX ~ OVA OVA+DEX t

61.7_+ 9.8 145.7±25.1 25.7_+_ 7.4 74.9_+ 8.7

35.1_+4.6 66.8±3.7 24.6_+3.2 64.8_+5.1

5 5 4 4

+Daily oral dose for 7 days Values are means ± standard error of data from the number of animals indicated. CFA - complete Freund's adjuvant; DEX = DEAE-dextran

respiratory tract, additional studies have been undertaken to assess the capacity of the Auspharm (patent pending) formulation, containing antigens derived from Mycoplasnza hyopneumoniae, to protect against mycoplasma in pigs after immunization by the IP route. These studies, undertaken by Sheldrake et al. 4°, have established a commercial potential for this vaccine formulation which is presently under investigation. FACTORS AFFECTING THE DELIVERY OF A N IgA A N T I B O D Y R E S P O N S E T O MUCOSAL SITES This paper has focused on the induction of IgA responses and modes of antigen presentation to maximize this event. Since inductive events occur within organized lymphoid structures such as Peyer's patches, yet the effector cells derived from this response are required at remote subepithelial lamina propria sites, the factors which determine the localization of effector cells at these sites also require attention. Previous studies from this laboratory 41"42 have examined the role of antigen in determining the localization patterns of IgA-producing plasma cells. These experiments prompted the conclusion that antigen is responsible for retention and proliferation of ACC which extravasate in the gut lamina propria. However, if antigen is the only factor required for lgA ACC localization, then adoptive transfer of AOCC to unprimed recipients given only lumenal antigen should produce the same result. However, experiments reported recently by Dunkley and Husband 43 indicate that additional factors are required. Whereas adoptive transfer of AOCC to animals previously primed IP and challenged ID localized well in the intestinal lamina propria, animals given only ID challenge failed to localize cells to any substantial extent. However, animals given Peyer's patch immunization without ID challenge were able to localize cells in the lamina propria. This suggested that cellular factors operate in addition to antigen in localizing an IgA response in the lamina propria. To test this hypothesis non-immunized rats were adoptively transferred with immune T cells for 3 days prior to AOCC adoptive transfer. This resulted in substantial numbers of adoptively transferred AOCC appearing in the gut lamina propria 24h after transfer. These studies demonstrate a role for T cells in IgA plasma cell distribution in non-organized lymphoid tissue in addition to any role they have in Peyer's patch induction sites. Cell labelling studies have revealed that T-helper cells which migrate to the lamina propria are predominantly of large lymphocyte morphology and that recently activated T-helper blasts migrate to the lamina propria

V a c c i n a t i o n in m u c o s a l infection control: A.J. H u s b a n d

more readily than small resting T-helper cells 44. This, coupled with evidence we have obtained for the requirement of cognate help in the Peyer's patch induction site lz, suggests a functional heterogeneity among T-helper phenotypes involved in IgA responses. The division ofT-helper cells into Th 1 and Th2, suggested by Mosmann and Coffman 45, may provide a framework to explain the heterogeneity of helper function in Peyer's patch and lamina propria sites in the gut. Cells capable of providing cognate help for the early induction of B cell responses probably predominate in Peyer's patches. These correlate with Thl cells, characterized by IFN-~, and IL-2 production. On the other hand the Th2 subset, which derive from short-lived non-recirculating T cells and which are characterized by IL-4, 5 and 6 production, are more likely to predominate in the lamina propria effector sites where they promote expansion of IgA-producing B-cell clones. These possibilities are currently under investigation.

10

CONCLUSIONS

19

The failure of traditional immunization strategies to control diseases of mucosal surfaces has stimulated the investigation of a range of novel approaches. Adjuvants with demonstrated potential to stimulate IgA responses by mucosal delivery are being developed. An alternative strategy is proposed in this paper whereby antigen delivered to gut-associated lymphoid tissue without traversing the epithelium (i.e. by serosal delivery) has a distinct advantage over oral delivery systems. While these approaches have been developed for veterinary applications in which they have proved to be both effective and practicable, for human mucosal vaccines the development of improved oral delivery systems is of paramount importance. In this regard the development ofmicroparticle formulations and bacterial or viral vector systems, coupled with appropriate adjuvants, is likely to yield the most effective outcome.

11 12 13

14 15 16

17 18

20

21

22

23

24

REFERENCES 25 1

Rowley, D. The problems of oral immunisation. Aust. J. Exp. Biol. Med. Sci. 1977, 55, 1 2 Newby, T.J. Protective immune responses in the intestinal tract. In: Local Immune Responses of the Gut (Eds Newby, T.J. and Stokes, C.R.) CRC Press, Boca Raton, Florida, 1984, p. 143 3 Owen, R.L. Sequential uptake of horseradish peroxidase by lymphoid epithelium of Peyer's patches in the normal unobstructed mouse intestine: an ultrastructural study. Gastroenterology 1977, 72, 440 4 Owen, R.L. and Jones, A.L. Epithelial cell specialisation within human Peyer's patches: an ultrastructural study of intestinal lymphoid follicles. Gastroenterology 1974, 68, 189 5 Kawanishi, H., Saltzman, L.E. and Strober, W. Mechanisms regulating IgA class-specific immunoglobulin production in murine gut-associated lymphoid tissues. I. T cells derived from Peyer's patches that switch slgM B cells to slgA B cells in vitro. J. Exp. Med. 1983, 157, 433 6 Dunkley, M.L. and Husband, A.J. The induction and migration of antigen-specific helper cells for IgA responses in the intestine. Immunology 1986, 57, 379 7 Dunkley, M.L. and Husband, A.J. Distribution and functional characteristics of antigen-specific helper T cells arising after Peyer's patch immunization. Immunology 1987, 61, 475 8 Benson, E.B. and Strober, W. Regulation of IgA secretion by T cell clones derived from the human gastrointestinal tract. J. Immunol. 1988, 146, 1874 9 Clough, JD., Mires, L.H. and Strober, S. Deficient IgA antibody responses to arsanilic acid bovine serum albumin (BSA) in neonatally thymectomised rabbits. J. Immunol. 1971, 106, 1624

26

27

28

29

30 31

32 33

Crewther, P. and Warner, N.L. Serum immunoglobins and antibodies in congenitally athymic (nude) mice. Aust. J. Exp. Biol. Med. Sci. 1972, 58, 625 Pritchard, H., Riddaway, J. and Micklem, H.S. Immune responses in congenitally thymus-less mice. Clin. Exp. Immunol. 1973,13, 125 Dunkley, M.L., Husband, A.J. and Underdown, B.J. Cognate T cell help in the induction of IgA responses in vivo. Immunology 1990, 71, 16 Bland, P.W. and Warren, L.G. Antigen presentation by epithelial cells of the rat small intestine. I1. Selective induction of suppressor T cells. Immunology 1986, 58, 9 Mayer, L. and Shlien, R. Evidence for function of la molecules on gut epithelal cells in man. J. Exp. Med. 1987, 166, 1471 Bland, P.W. and Whiting, C.V. Antigen processing by isolated rat intestinal villus enterocytes. Immunology 1989, 68, 497 Nedrud, J.G. and Sigmund, N. Cholera toxin as a mucosal adjuvant: III. Antibody responses to nontarget dietary antigens are not increased. Regul. Immunol. 1991, 3, 217 Holmgren, J. Actions of cholera toxin and the prevention and treatment of cholera. Nature 1981, 292, 413 Sommermeyer, H., Schwinzer, R., Kaever, V., Wessel, K., Schmidt, R.E., Behl, B., Wonigeit, K., Szamel, M. and Resch, K. Cholera toxin modulates the T cell antigen receptor/CD3 complex but not the CD2 molecule and inhibits signalling via both receptor structures in the human T cell lymphoma Jurkat. Eur. J. Immunol. 1989, 19, 2387 Kiyono, H., McGhee, J.R., Kearney, J.F. and Michalek, S.M Enhancement of in vitro immune responses of murine Peyer's patch cultures by concanavalin A, muramyl dipeptide and lipopolysaccharide. Scand. J. Immunol. 1982, 15, 329 Sugimoto, M., Germain, RN., Chedid, L. and Benacerraf, B. Enhancement of carrier-specific helper T cell function by the synthetic adjuvant N-acetylmuramyI-L-alanyl-g-isoglutamine (MDP). J. Immunol. 1978, 120, 980 Li, Y.S., Dearden-Badet, M-T. and Revillard, J.-P. Selective induction of high levels of IgA synthesis in Peyer's patch B cells by protein kinase C-activating phorbol esters. J. Immunol. 1991, 147, 1752 Hone, D., Morona, R., Attridge, S. and Hackett, J. Construction of defined gale mutants of Salmonella for use as vaccines. J. Infect. Dis. 1987, 156, 167 Clements, J.D., Lyon, F.L., Lowe, K.L., Farrand, A.L. and EL-Morshidy, S. Oral immunization of mice with attenuated Salmonella enteritidis containing a recombinant plasmid which codes for production of the B subunit of heat-labile Escherichia coil enterotoxin. Infect. Immun. 1986, 53, 685 Ramsay, A.J., Kohonen-Corish, M.R.J. and Turner, R. Interleukins 5 and 6 encoded by recombinant vaccinia virus enhance IgA reactivity to co-expressed influenza haemagglutinin in vivo (Abstract). In: Proceedings of the 20th Annual Meeting Australian Society for Immunology (Ed. Roland, J.) Australian Society for Immunology, 1990, p. 98 Eldridge, J.H., Staas, JK., Meulbroek, J.A., McGhee, J.R., Tice, T.TR. and Gilley, R.M. Disseminated mucosal anti-toxin antibody responses induced through oral or intrathecal immunisation with toxoid-containing biodegradable microspheres. In: Advances in Mucosal Immunology (Eds MacDonald, T.T., Challacombe, S.J., Bland, P.W., Stokes, C.R., Heatley, R.V. and Mowat, A.McI.) Kluwer Academic Publishers, Lancaster, 1990, p. 375 Nervig, M., Gough P.M., Kaeberle, ML. and Whetstone, C.A. Advances in Carriers and Adjuvants for Veterinary Biologics. Iowa State University Press, Ames, 1986 Klein, J.R., Lefrancois, L. and Kagnoff, M. A murine cytotoxic T lymphocyte clone from the intestinal mucosa that is antigen specific for proliferation and displays broadly reactive inducibiy cytotoxic activity. J. Immunol. 1985, 135, 3697 McKenzie, S.J. and Halsey, J.F. Cholera toxin B subunit as a carrier protein to stimulate a mucosal immune response. J. Immunol. 1984, 133, 1818 Garcon, N.M.J. and Six, H.R. Universal vaccine carrier. Liposomes that provide T-dependent help to weak antigens. J. Immunol. 1991, 146, 3697 Pierce, N.F. and Gowans, J.L. Cellular kinetics of the intestinal response to cholera toxoid in rats. J. Exp. Med. 1975, 142, 1550 Husband, A.J. and Gowans, J.L. The origin and antigen-dependent distribution of IgA-containing cells in the intestine. J. Exp. Med. 1978, 148, 1146 Husband, A.J. An immunisation model for the control of infectious enteritis. Res. Vet. Sci. 1978, 25, 173 Husband, A.J., Beh, K.J. and Lascelles, A.K. IgA-containing cells in the ruminant intestine following intraperitoneal and local immunization. Immunology 1979, 37, 597

Vacci ne, Vol. 11, I ssue 2, 1993

111

Vaccination in m u c o s a l infection control: A.J. H u s b a n d 34

Husband, A.J. and Seaman, J.T. Vaccination of piglets against Escherichia coli enteritis. Aust. Vet. J. 1979, 55, 435 35 McColm, A.A., Bomford, R. and Dalton, L. A comparison of saponin with other adjuvants for the potentiation of protective immunity by a killed Plasmodium yoelii vaccine in the mouse. Parasite Immunol. 1982, 4, 337 36 Shek, PN. and Sabiston, B.H. Immune response mediated by liposome-associated protein antigens. I. Potentiation of the plaque-forming cell response. Immunology 1981, 45, 349 37 Kiyono, H., Michalek, S.M., Mosteller, L.M., Torii, M., Hamada, S. and McGhee, J.R. Enhancement of murine immune responses to orally administered haptenated Streptococcus mutans, Scand. J. Immunol. 1982, 16, 455 38 Tamauchi, H., Tadakuma, T., Yasuda, T., Tsumita, T. and Saito, K. Enhancement of immunogenicity by incorporation of lipid A into liposomal model membranes and its application to membraneassociated antigens. Immunology 1983, 58, 605 39 Kramp, W.J., Six, H.R., Drake, S. and Kasel, J.A. Liposomal enhancement of the immunogenicity of adenovirus type 5 hexon

112

Vaccine, Vol. 11, Issue 2, 1993

40

41

42

43

44

45

and fiber vaccine. Infect. Immun. 1979, 25, 771 Sheldrake, R.F., Gardner, I.A., Saunders, M.M. and Romalis, L.F. Intraperitoneal vaccination of pigs for the control of Mycoplasma hyopneumoniae. Res. Vet. Sci. 1992 in press Husband, A.J. Kinetics of extravasation and redistribution of IgA-specific antibody-containing cells in the intestine. J. Immunol. 1982, 128, 1355 Husband, A.J. and Dunkley, M.L. Lack of site of origin effects on distribution of IgA antibody-containing cells. Immunology 1985, 54, 215 Dunkley, M.L. and Husband, A.J. The role of non-B cells in localizing an IgA plasma cell response in the intestine. Regul. Immunol. 1991, 3, 336 Dunkley, M.L. and Husband, A.J. Role of antigen in migration patterns of T cell subsets arising from gut-associated lymphoid tissue. Regul. Immunol. 1989, 2, 213 Mosmann, T.R. and Coffman, R.L. TH1 and TH2 cells: different patterns of lymphokine secretion lead to different functional properties. Annu. Rev. Immunol. 1989, 7, 145