Use of novel adjuvants and delivery systems to improve the humoral and cellular immune response to malaria vaccine candidate antigens

Use of novel adjuvants and delivery systems to improve the humoral and cellular immune response to malaria vaccine candidate antigens Daniel M. Gordon

The immune effector mechanisms responsible j or the solid protection against malaria, as demonstrated by immunization with radiation-attenuated sporozoites, are poorly understood. An effective malaria vaccine must induce a well orchestrated combination of humoral and cellular immune responses directed against critical parasite antigens / epitopes expressed during different stages o f the parasite's complicated life cycle. Currently licensed human vaccine adjuvants, such as alum, may improve antibody production but are poor stimulators of cellular effector mechanisms, while potent cellular stimulants such as Freund's adjuvant are too reactogenic for human use. Over the last 5 years we have systematically evaluated several methods o f antigen presentation to include chemical conjugation to bacterial carrier proteins, emulsification in 'Freund's-like' preparations, and incorporation into liposomes. This work has resulted in the production o f safe, potent vaccine delivery systems capable of targeting multiple antigenic determinants to the host's immune system. Further advances in malaria vaccine development now depend on the identification o['appropriate parasite epitopes for inclusion in a multicomponent-multistage vaccine. Keywords: Malaria vaccines; adjuvants; delivery system

Malaria is a protozoan infection caused, in human beings, by members of the subgenera Plasmodium (P. vivax, P. ovale, and P. malariae ) or Laverania ( P. falciparum ). The life cycle of this parasite is complex and includes an extrinsic, sexual phase which occurs in Anopheline mosquitoes, and an intrinsic, asexual phase that occurs in humans. The parasite is normally transmitted to humans when the sporozoite stage is introduced by the bite of infected female mosquitoes. These sporozoites travel through the blood stream and invade liver cells where they multiply asexually as exoerythrocytic stage parasites. In contrast to P. falciparum, P. malariae, and P. ovale, which have only one type of exoerythrocytic form, P. vivax has two types of exoerythrocytic forms, a primary type which matures in 6 to 9 days as well as a secondary type, referred to as hypnozoites, which can remain dormant for months before maturing and releasing tissue merozoites. Once released, tissue merozoires invade erythrocytes and initiate the erythrocytic phase of the infection. During the erythrocytic phase, merozoites invade erythrocytes, undergo asexual maturation, and ultimately rupture the erythrogite releasing new merozoites. It is this phase of the infection that results Department of Immunology, Walter Reed Army Institute of Research, Washington, DC 20307-5100, USA 0264-~410X/93/05/0591-03 1993 Butterworth-HeinemannLtd

in the clinical symptoms of the disease referred to as malaria. A portion of the merozoites which invade erythrocytes do not multiply but differentiate into sexual forms, gametocytes. When ingested by a mosquito, male and female gametocytes unite to form an ookinete which then develops into an oocyst on the gut wall of the mosquito. After undergoing further development, the oocyst ruptures releasing thousands of sporozoites which migrate to the mosquito's salivary glands where they undergo final maturation. The life cycle is complete when mature sporozoites are injected into a new vertebrate host. Immunity to malaria is even more complicated than the life cycle, involving both humoral and cellular mechanisms directed against critical parasite epitopes expressed during different stages of the parasite's complicated life cycle 1. Investigators around the world are working on identifying and characterizing useful immune targets from the various stages of the parasite's development for incorporation into a multi-stage, multicomponent, vaccine preparation 2 4. One problem that must be addressed, however, is how to present these numerous epitopes effectively to the immune system in order to induce the necessary humoral and cellular mechanisms responsible for solid protective immunity. The remainder of this presentation will summarize our clinical experience with vaccine adjuvants and delivery systems over the past 5 years.

Vaccine, Vol. 11, Issue 5, 1993 591

Adjuvants and delivery systems for malaria vaccine: D.M. Gordon

ADJUVANTS AND DELIVERY SYSTEMS One of the first malaria vaccine candidates to undergo safety, immunogenicity and efficacy testing consisted of the alum-adjuvanted, Escherichia coli-expressed, recombinant protein R32tet325. This study evaluated five doses ranging from 10 to 800/~g of R32tet32. Although this preparation was safe, it was less immunogenic in human volunteers than expected. Only one volunteer in the 800 ~Lg dose group demonstrated boosting with subsequent doses of vaccine. The geometric mean (GM) anti-R32 specific lgG concentration of the three individuals in the 800ttg dose group was only 3.7 Itg ml 1, as measured by ELISA. For reference the anti-R32 specific IgG antibody level in the one protected individual, at time of challenge, was 9.8 ltg ml 1. One possible explanation for the generally poor antibody response to R32tet32 may have been an absence of adequate T-helper cell epitopes in the tet32 expression partner of this fusion protein. Subsequently, the expression system was re-engineered so that the R32 fragment would be expressed fused to the amino terminal of an 81 amino acid fragment derived from the non-structural protein 1 (NSIs~), of influenza A, a fragment previously shown to contain T-helper cell epitopes and enhance fusion protein expression ~. R32NS1sl, at 11.5, 115 or ll50ttg per dose, was also well tolerated when administered as an alum-adjuvanted vaccine. Improved immunogenicity was demonstrated by the observation that three doses of R 3 2 N S I alum, administered at 11.5/xg per dose, produced a GM anti-R32 IgG antibody level comparable to that seen with the highest dose of R32tet32 (2.3 versus 3.7/~gml 1). No significant increase in GM antibody levels, however, was observed in the 115 or 1150/~g dose groups. At this point, two different approaches were taken to improve the overall immunogenicity of the R32 fragment of the P../Mciparum CS protein. One approach involved the systematic evaluation of repeat-based synthetic peptides and recombinant proteins chemically conjugated to various bacterial proteins which served as carrier molecules. Initial studies, conducted in volunteers, designed to optimize the carrier protein, peptide-tocarrier conjugation ratios, and peptide orientation, revealed that the recombinant peptide R32-Leu-Arg, chemically conjugated to detoxified pseudomonas toxin A (R32TA), resulted in the highest seroconversion rate (21 of 22 volunteers), with the best overall response being observed in the 4001tg dose group (GM anti-R32 antibody levels of ll.0/tg m1-1)7. Subsequently, an expanded phase I/phase IIa safety, immunogenicity and efficacy trial of alum-adjuvanted R32TA was conducted. This recently completed study confirmed the relative safety of this formulation. The GM anti-R32 antibody level in the 20 volunteers 2 weeks after the second dose of vaccine was 12.2 lLg ml 1 (Ref. 8). Concurrent with these studies, another series of experiments was conducted evaluating new adjuvant preparations ~, or combinations of potent immunomodulators and delivery systems 1°. For these studies the recombinant fusion protein R32NS81 was used. Five volunteers were vaccinated with 1230/~g of R32NS181 adiuvanted with the Ribi DETOX" adjuvant consisting of 5 l~g monophosphoryl lipid A (M PL ), 50/~g cell wall cytoskeleton of Mvcohacterium phlei (CWS), and 0.5 itl squalane. Similar doses of vaccine were given 2 and

592

Vaccine, Vol. 11, Issue 5, 1993

4 months after the first. R32NS181-DETOX" was more reactogenic than R32NSIsl alum by historic comparison. All five volunteers receiving R32NS181-DETOX" complained of moderate pain at the site of injection 1 2 days after each dose. Four individuals demonstrated erythema and warmth at the injection site and one volunteer developed enlarged axillary lymph nodes after the third dose of vaccine. None of these signs or symptoms were seen with the alum-adjuvanted preparation. R32NS 181 DETOX" was more immunogenic than R32NS181-alum based on comparison to archived sera run concurrently in a standard ELISA assay. The GM anti-R32 antibody level 2 weeks after three doses of R32NSI81 DETOX '~' was 25.4/tg ml 1 compared with 2.3/xg ml- 1 measured in sera obtained at a comparable time from volunteers immunized with R32NS181-alum (p < 0.05, Mann-Whitney U test). Although R32NS181DETOX" was more reactogenic than the alumadjuvanted preparation, none of the volunteers declined further participation in the study. In light of these observations and the significant improvement in antibody levels, current plans include efficacy studies, using a laboratory challenge model, of the R32NS181DETOX" vaccine formulation. We have also evaluated the use of liposomes as an antigen/adjuvant delivery vehicle for malarial antigens. The rationale for their use has recently been reviewed elsewhere 11 A phase I safety and immunogenicity study, in humans, again using R32NS18~ as the antigen, has recently been completed. A single formulation containing 1260/~gm1-1 R32NSI81, and 4400#gml 1 M P L encapsulated in liposomes consisting of 25400/~g (37.5/~mol) dimyristoyl phosphatidylcholine, 2900/xg (4.4/~mol) dimyristoyl phosphatidylglycerol, and 13 000 #g (33.6 FLmol) cholesterol was prepared. Actual vaccine doses were prepared at the time of vaccination by mixing a 1"100, 1"10, 1:4, 1:2, or an undiluted sample of the original R32NSI sl-liposome formulation with an equal volume of aluminium hydroxide. This approach resulted in vaccine doses with constant ratios of liposome components and an increasing liposome'alum ratio. Thirty adult, malaria-naive volunteers were randomly divided into five groups with each subsequent group receiving an increasing dose of vaccine. The vaccine was generally well tolerated. Objective local reactions were limited to erythema or induration and were observed after 11 of the 88 administered doses. These reactions were seen only in individuals receiving vaccine doses containing 1100 #g or more of MPL. All vaccine doses were highly immunogenic. The five individuals receiving the most dilute preparation (6.3/xg R32NS181 and 22 #g MPL ) had a GM anti-R32 antibody level of 22.7/xg ml(range 6.4 to 287.2~lgml 1). The GM decrease to 12/~g ml ~ if the extremely high responder is excluded. Increasing the amount of R32NS18~ and MPL to 630/~g and 2200/xg respectively resulted in an increase in the GM anti-R32 antibody level to 33 tlg ml 1 Fiqure 1 summarizes the GM anti-R32 antibody levels from clinical phase 1 safety and immunogenicity studies evaluating vaccine formulations designed to improve the humoral immune response to the repeat region of the P. ./alciparum circumsporozoite (CS) protein. Although three doses of R32NSlsl alum, given at 11,5, 115 or 1150/~g per dose, were able to induce antibody levels comparable with three doses of R32tet32 administered at 800 lLg per dose, markedly better antibody levels were

Adjuvants and delivery systems for malaria vaccine: D.M. Gordon

cytoplasmic mechanisms and delivered to class I restricted cytolytic T cells 14. This latter route of presentation may facilitate the induction of cytotoxic T cells which have previously been shown to be important in a P. berghei rodent model of malaria 15.

35

3o

i

25 20 > _o

~

15

s 800 pg R32tet32 Alum

Figure 1

1150 IJg R32NS1 Alum

m

400 ~Jg R32TA Alum

REFERENCES 1 2 1230 pg R32NSI

Detox(~

630 pg R32NS1 Liposome

Geometric mean anti-R32 antibody levels

observed after only two 400#g doses of R32TA. Significant improvements in antibody levels were observed with 1230 #g of R32NS 1s ~ DETOX~"~compared with either the R32tet32 or R32NS18~ alum formulations; however, the increased reactogenicity of the DETOX ~ adjuvant may limit its usefulness. The liposome formulation tested offered several advantages. Limited amounts of antigen (6.3 #g per dose), when combined with the potent immunomodulator MPL, are able to induce a GM anti-R32 level of 2 2 . 7 # g m l ~ (range 6.4 to 287.2#gml ~, n = 5 ) . However, liposomes are capable of incorporating relatively large amounts of 'antigen', thus a variety of different antigens providing target epitopes from molecules expressed during the various stages of the parasite life cycle may be incorporated in future preparations. The liposome formulation was significantly less reactogenic than R 3 2 - D E T O X ~', particularly if one considers the difference in the amount of MPL used in all but the most dilute liposome dosage. Incorporation of the MPL into liposomes greatly reduces its inherent toxicity 12. The highest dose of MPL administered, 2200/tg per dose, represents a 12-fold increase in the previously reported maximal safe dose of free MPL 13. The improved humoral immune response observed is, in part, dependent on the use of MPL ; however, processing of the liposome-delivered material is also crucial. Liposomes are actively cleared from the site of injection by macrophages. Subsequently the antigen is processed through endosomal mechanisms with presentation to class II restricted T-helper cells as well as through

3

4

5

6

7

8

9

10

11 12

13 14

15

Gordon, D.M. Malaria vaccines. Infect. Dis. Clin. North Am. 1990, 4(2), 299 313 UNDP/World Bank/WHO Special Program for Research and Traiping in Tropical Diseases. Asexual blood stage and transmissionblocking antigens of plasmodia. Report of the Sixth Meeting of the Scientific Working Group on Immunology of Malaria. Geneva, 26-28 March, 1984, TDR/IMMAL/SWG(6)/84.3 UNDP/World Bank/WHO Special Program for Research and Training in Tropical Diseases. Exoerythrocytic and asexual blood-stage antigens of human malaria parasites. Report of the Tenth Meeting of the Scientific Working Group on Immunology of Malaria. Geneva, 13-15 April, 1988, TDR/IMMAL/SWG(10)/88.3 Lyon, J.A., Thomas, A.W. et al. Specificities of antibodies that inhibit merozoite dispersal from malaria-infected erythrocytes. Mol. Biochem. Parasitol. 1989, 36, 77-86 Ballou, W.R., Hoffman, S.L. et al. Safety, and efficacy of a recombinant DNA Plasmodium falciparum sporozoite vaccine. Lancet 1987, i, 1277-1281 Young, J.F. Efficient expression of influenza virus NS1 nonstructural proteins in Escherichia coil Proc. Natl Acad. Sci. USA 1983, 80, 6105-6109 Ballou, WR. Progress in malaria vaccines. In: Vaccines and Immunotherapy (Ed. Cryz, S.J.) Pergamon Press, New York, 1991, pp. 373-380 Fries, L., Gordon, DM. et al. Safety, immunogenicity, and efficacy of a Plasmodium falciparum vaccine comprised of a cirumsporozoite protein repeat based peptide conjugated to Pseudomonas aerugenosa toxin A. Infect. Immun. 1992, 60, 1834-1839 Rickman, L.S., Gordon D.M et al. A novel adjuvant containing the cell wall skeleton of mycobacteria, monophosphoryl lipid A, and squalane substantially improves the immunogenicity in humans of a malaria circumsporozoite protein vaccine. Lancet 1991,337, 988-1001 Fries, L., Gordon, D.M. et al. Liposome malaria vaccine in humans: a novel, safe, and potent adjuvant strategy. Proc. Natl. Acad. Sci. USA 1992, 87, 35~362 Alving, C.R. Liposomes as carriers of antigens and adjuvants. J. Immunol. Meth. 1991, 140, 1-13 Richards, R.L., Swartz, G.M. et al. Immunogenicity of liposomal sporozoite antigen in monkeys: adjuvant effects of aluminium hydroxide and non-pyrogenic liposomal lipid A. Vaccine 1989, 7, 506-512 Vosika, G.J., Barr, C. et al. Phase-1 study of intravenous modified lipid-A. Cancer Immunol. Immunother. 1984, 18, 107 112 Verma, J.N., Wassef, N.M. et al. Phagocytosis of liposomes by macrophages: intracellular fate of liposomal malaria antigen. Biochem. Biophys. Acta 1991, 1066, 229-231 Romero, P., Maryanski, J.L. et al. Cloned cytotoxic T cells recognize an epitope in the circumsporozoite protein and protect against malaria. Nature 1989, 341,232 326

Vaccine, Vol. 11, Issue 5, 1993

593