Parasite Immunology 1991, 13, 509-5 16

Genetic control of the immune response to a synthetic vaccine against Plasmodium falcipavum M A N U E L E. PATARROYO, JAVIER VINASCO, ROBERTO AMADOR, FABIOLA ESPEJO, YOLANDA SILVA, A L B E R T 0 M O R E N O , M A U R I C I O ROJAS, A N A LUCIA MORA, M A R G A R I T A SALCEDO, VICTORIA VALERO, A N A KARLA G O L D B E R G * & J O R G E KALIL* Instituto de lmmunologia, Hospital San Juan de Dibs, Universidad Nacional de Colombia, Bogota D.E., Colombia and *Laboratorio de Imunologia de Transplantes, lnstituto do Corapio, Faculdade de Medicina, Universidade de Scio Paulo, Scio Paulo, B r a d Accepted for publication 27 March 1991 Summary Two independent vaccination trials using a hybrid synthetic polypeptide containing epitopes from four proteins of Plasmodium falciparum were performed. In the first trial 63 and in the second 122 volunteers were vaccinated, using different immunization schedules. The analysis of the humoral response to the vaccine, measured by IgG antibody titres to the polypeptide showed a bimodal distribution in both cases suggesting genetic control of the immune response to this protein. There was a small group of low or non-responders and a large group of good responders. HLA phenotyping of the two groups disclosed an association of the low responders to HLA-DR4 antigens with chi-square P value of 0.00039 when compared with the good responders group. These findings provide evidence for the genetic control of the immune response to the synthetic vaccine by the association of this response with particular alleles of the HLA class 11 antigens; such findings may lead to an explanation of the mechanism involved in disease susceptibility and need to be used in the design of a totally effective vaccine. Keywords: synthetic malaria vaccine, immune response, HLA class I1 antigens, P . falciparum

Introduction Malaria continues to be one of the greatest public health problems in developing countries (Sturchler 1989). In the search for immunoprophylactic methods to control this disease (Ballou et al. 1987, Collins et al. 1988, Jendoubi et al. 1987, Nussenzweig & Nussenzweig Correspondence: Dr Manuel E. Patarroyo

509

510

M.E.Patarroyo et al.

1984),particularly in its most aggressive form caused by Plasmodiumfalciparum, vaccines that provide a certain degree of protection to experimentally challenged volunteers are being developed (Nussenzweig & Nussenzweig 1984, Patarroyo et al. 1988). However, genetic restriction of the immune response to these vaccines in experimental models have been found (Good et al. 1986). A synthetic vaccine, named SPf 66, has been developed in our laboratory and is directed mainly against the P. fakiparum asexual blood stages (Patarroyo et al. 1988). Preliminary studies in a small group of human volunteers have shown that two or three inoculationswith SPf66 elicited protection against the experimentalchallenge with a wild strain of P. falciparum (Patarroyo et al. 1988). In order to analyse the possible influence of genetic factors in the individual immune response, an extended study of immunogenicity in two larger groups was conducted with careful monitoring of the vaccinees and controls, in whom complete HLA typing was performed. Materials and methods HUMAN VOLUNTEERS

399 healthy young male volunteers (18-21 years old, most of them of black race), from the Colombian military forces from whom, after a careful explanation of the project, written consent was obtained, were enrolled in the two trials. All volunteers, after the induction period, were assigned to an area endemic for malaria named Tumaco for approximately 12 months of their military service. The two studies Tumaco A and Tumaco B, begun one year apart, were conducted on different individuals undergoing military service. The first trial, Tumaco A, involved 193 individuals; 63 were choosen at random to be immunized with the SPf 66 vaccine, the remaining 130 in the platoon received placebo injections according to the same immunization protocol: on days 0, 20 and 220. The Tumaco B trial, in which the immunization schedule was modified by shortening the time to the third dose, consisted of 206 volunteers, of whom 122 were immunized. The remaining 84 in the platoon received placebo injections. Three doses of 2 mg of the hybrid synthetic protein, dissolved in 0.9% saline solution and adsorbed to A1(OH)3 were administered subcutaneously to each vaccinee. The placebo group received the same treatment except for the absence of the immunizing peptide in the innoculum. The doses were given on days 0,20 and 90 in the second trial. The study was approved by the Colombian Ministry of Public Health and was performed with the aid of the local Malaria Eradication Service (SEM) and the Colombian military forces. All volunteers underwent rigorous clinical examination and extensive laboratory testing which included: complete haematological analysis; serological tests for HIV, VDRL, hepatitis B. Rheumatoid factor, Antinuclear antibodies and a Coombs test; blood chemistry included gamma glutamyl transferase, bilirubins, transaminases, alkaline phosphatase and creatinine, measured before and after each immunization in order to evaluate changes induced by vaccination. SYNTHETIC VACCINE

The peptide used in the production of the vaccine was chemically synthetized, based on

HLA control to a malaria synthetic vaccine

5 11

Merrifield’s procedure (Merrifield 1963), and according to Houghten’s multiple solid phase synthesis method (Houghten 1985). The aminoacid sequence of this peptide is: Cys-

Gly-Asp-Glu-Leu-Glu-Ala-Glu-Thr-Gln-Asn-Val-Tyr-Ala-Ala-Pro-Asn-Ala-Asn-ProTyr-Ser-Leu-Phe-Gln-Lys-Glu-Lys-Met-Val-Leu-Pro-Asn-Ala-Asn-Pro-Pro-Ala-AsnLys-Lys-Asn-Ala-Gly-Cys. The peptide was allowed to polymerize under air oxidation for 6 h in order to obtain a final product composed by 3-5 individual peptides joined by disulphide bridges. Since the molecular weight is between 18 kD and 23 kD, the use of a carrier was not required. Purity of the product was checked by HPLC, SDS PAGE, aminoacid analysis and aminoacid sequence. Furthermore, after extensive dialysis against double distilled water, pyrogenicity in rabbits, sterility and cytotoxicity in vitro, safety in Aotus monkeys and stability tests were performed before use in human volunteers. (Patarroyo ef al. 1988) MONITORING OF ANTIBODY TITRES

Venous blood was drawn on days 0,19,37,52,98,200 and 240 in the Tumaco A trial and on days 0, 14, 65, 110, 150 and 280 in the Tumaco B trial. Sera were aliquoted and immediately stored at - 20°C for future analysis. On certain occasions programmed blood samples for antibody titres determination or HLA typing could not be obtained due to guard or patrol assignments of some of the volunteers to places far from the collection site. Sera were tested in duplicate by Falcon Assay Screening Test ELISA (FAST ELISA) (Campbell et al. 1987, Salcedo et al. 1991). Plate lids (Falcon ELISA plates, Falcon, Oxnard, CA) were coated with 10 ,ug/ml of the peptide in PBS for 2+ h. Antigen coated lids were then washed with PBS/Tween 0.5% and dried at room temperature. The lids were then incubated with previously determined serum dilutions in the individual wells for ten min, washed and then incubated with goat anti-human IgG peroxidase conjugate (Sigma Chemical Co., St Louis, MO) for 1 h. After washing and reaction with the substrate Tetramethylbenzidine (TMB), (Kirkegaard & Perry Lab. Inc., Gaithersburg, MD) for five min, lids were removed and optical densities determined at 620 nm, on an ELISA plate reader. The antibody titres were expressed as the end point of O D read at 620 nm. These were considered to be significant if they were three standard deviations above the average obtained from the preimmune sera. HLA TYPING

HLA A, B, DR and DQ typing was performed on 48 vaccinees and 47 platoon controls in the first trial; and on 67 vaccinees in the second trial by the microlymphocytotoxicity technique (Terasaki et al. 1978) on Terasaki Third HLA Trays (One Lambda Inc., Los Angeles, CA). 58 antisera to HLA-A and B, 41 to HLA-DR and DQw were used. Chisquare tests were performed and P values in the HLA-DR analysis corrected multiplying P by ten (the number of alloantigens) (Wayne 1980). Yates correction was applied whenever the number of individuals in any of the four positions of the 2 x 2 table was below five. Results Clinical laboratory tests and medical examinations were normal before and after

M.E.Patarroyo et al.

512

synthetic vaccine or saline solution injections. None of the volunteers developed systemic reactions, and only minimal local erythema was found at the injection site in three volunteers in the Tumaco A trial following the third immunization, a phenomenon also seen in some of the placebo volunteers. Minimal variation in the blood chemistry tests were observed in both vaccinated and control groups showing that vaccine was safe for human use. In the Tumaco A trial, analysis of the antibody reponse by ELISA assay of the IgG titres elicited after the second immunization showed two types of responses. The two groups varied in the time that elapsed between immunizations and the increase in antibody levels to the highest titre reached. Four determinations of titres, were performed on days 17,32,78 and 180 after the second immunization, with the results ranging from 0 to 1 :25 600. When the highest antibody titres obtained for each individual are plotted on a histogram (Figure I), a bimodal distribution can be clearly observed. A low responder group of 16 out of 63 of the vaccinees had antibody titres below to 1:200. The remaining vaccinees exhibited antibody titres above 1 :200, reaching in some cases 1 :25 600. HLA typing of the two groups of vaccinees and of the non-immunized control platoon members are shown in Table 1. No significant differences in class I antigen frequencies were observed in these groups (data not shown). When class I1 antigen were compared, HLA-DR and DQ antigen distribution in the good responders group was similar to that of the control group. However, a definite increase of HLA-DR4 ( 1 1/16,68.7%) in the low responders group was evident when compared to good responders (8/32, 25.0%) or to controls (17/47, 36.2%) (Figure 2). The Chi-square for low versus good responders which, corrected for the number of antigens tested gave a showed a P value of 6.06 x P value of 1.53 x Moreover a significant number of the low responder vaccinees typed DR4, blank (DR4,-) (P=6.039 x suggests an increase of homozygotes in this group. DQw antigens were very similar in all three groups, without a significant difference.

25

,

20

t? a

E 15 -

g % 10 d

z 5 0

1/100

V200

1/4W

1/800

1/1600 1/3200 1/6400 1/12800 1/25600

IgG antibody

iitres

.;

Figure 1. Bimodal distribution of the IgG antibody titres to the synthetic vaccine SPf 66 in vaccinees of Tumaco A and B trials. The individuals were classified as low responders (antibody titres below 1 :200) and good responders (antibody titres greater than 1 :200). Tumaco A Tumaco B D.

HLA control to a malaria synthetic vaccine

5 13

Table 1. Frequencies of the different HLA class I1 antigens in the individuals included in the Turnaco A and B trials.

Type

Vaccinated Low responders Good responders A B A B (16) (13) (32) (54)

DR 1 2 3 4 5 W6 7 W8 w9 w10 -Blank

6.2" ( 6.2 (1) 0 68.7 (1 1) 0 18.7 (3) 25.0 (4) 6.2 (1) 0 0 68.7 (1 1)

DQ 1 2 3 4 - Blank

62.5 (10) 30.8 (4) 18.7(3) 7.7(1) 62.5 (10) 76.9(12) 0 7.7 (1) 56.3 (9) 76.9 ( I 2)

0 30.7 (4) 7.7 (1) 69.2 (9) 7.7 (1) 0 7.7 ( I ) 0 15.4 (2) 0 61.5 (8)

Control (47)

15.6 (5) 25.0 (8) 28.1 (9) 25.0 (8) 15.6 (5) 18.7 (6) 9.4 (3) 15.6 (5) 6.2 (2) 3.1 ( I ) 37.5 (12)

7.3 (4) 22.6 (12) 32.0 (17) 28.3 (15) 18.9 (10) 18.9 (10) 30.2 (16) 9.4 (5) 13.3 (7) 1.9(1) 18*9(10)

10.6 (5) 27.6 (1 3) 25.5 (12) 36.2 (1 7) 14.9 (7) 14.9 (7) 21.3 (10) 8.5 (4) 8.5 (4) 0 31.9(15)

59.4 (19) 37.5 (12) 53.1 (17) 9.4 (3) 40.6 (13)

39.6 (21) 52.8 (28) 50.9 (27) 11.3 (6) 45.3 (24)

57.4 (27) 31.6 (15) 43.1 (20) 12.8 (6) 38.3 (18)

Vaccinated individuals are classified as low or good responders on the basis of a peak antibody titre after vaccination of < or > 1 in 200. a =percentage of the group with each HLA type. =actual number of individuals.

These results have been confirmed in a second independent trial (Tumaco B) where a different immunization schedule was employed. In this trial, 122 volunteers received the three doses of the vaccine in a shorter time interval, and samples for antibody titration were obtained 20 days after the third immunization. Again, a low responders group was observed comprising 19.4% of the vaccinees (Figure 2). HLA class I1 typing showed a similar distribution of HLA-DR and DQ antigens in the good responders and control groups, while the low responders group presented an excess of HLA-DR4 (9/13, 69.2%) (Figure 2). Chi-square for low versus good responders in this second trial, showed a P value of 6.06 x

Discussion

The role of the major histocompatibility complex (MHC) in immune responses to parasite infections has been well studied in experimental parasitology (Wassom 1990). Knowledge

514

M.E.Patarroyo et al. Turnaco B

Turnaco A DR4

DR4

.3

5

"@o"-DR4

31.3

Low responders

71 *7

"4@0"-DR4

Non-DR4 75

30.8

Good responders

Low responders

Non-DR4

Good responders

Controls

63.8

Figure 2. Distribution of DR4 antigen between individuals in the Tumaco A and B vaccine trials. Observe the similar proportions in the good responders in both trials and the controls, and the significant increase of DR4 antigen in both low responders groups.

in this area has progressed rapidly due to unique characteristics of the murine models, namely, genetic homogeneity of mouse strains, availability of H-2 congenic mice, and also the control of injected mass, level of virulence, timing and pathways of parasite inoculation. On the other hand, the influence of MHC in the immune response to parasite infections in humans is still badly defined. Several diseases have been associated with particular HLA haplotypes especially class I1 antigens. For instance, in viral diseases, a much more simple model, the influence of HLA has been clearly demonstrated (Batchelor & McMichael 1987). In our model, we were able to evaluate the association of HLA with the humoral immune response against parasite P.falciparum epitopes presented as a hybrid molecule used as a vaccine in humans. For this purpose, we have analysed the antibody titres elicited against the vaccine after a second or third immunization, in two independent trials and we have searched for an association with HLA haplotypes. In the first trial, we have analysed the kinetics of IgG response in four consecutive samples obtained 17,32,78 and 180 days after the second immunization (data not shown). A group comprising 33.3% of the vaccinees presented either a very weak response or none at all. These were named low responders. The remaining individuals formed a group of variable positive responses and were named good responders. Based on levels and time interval needed to raise antibody titres, as well as response to a third immunization, the latter group was split further into other groups called high and intermediate. These data are detailed elsewhere (Salcedo et al. 1991). The low responders group was associated with the increased presence of the HLADR4 antigen when compared to good responders (P=0.009), controls (P=0.048), or even to the IX Histocompatibility Workshop antigen frequencies in negro and Caucasian populations (data not shown) (Albert et al. 1984). Furthermore, in the Tumaco A trial, the non-DR4 low responders were mostly positive for DR7, an antigen highly

HLA control to a malaria synthetic vaccine

5 15

homologous to DR4. Another interesting finding was the number of blanks as the second allele of HLA-DR, suggesting homozygosity at the HLA-DR locus. Essentially the same results were found in the second trial in spite of the single determination of antibody titres performed. The same excess of HLA-DR4 in the low responders group could be clearly observed. Several explanations for these findings are possible: (a) the peptide may not be recognized in the context of HLA-DR4 by T cells; (b) there may be a preferential association of variable regions of the T cell receptor with certain HLA-DR haplotypes; and (c) the HLA-DR4 plus peptide complex may elicit a suppressive response. Further investigation may clarify these points. In another system, the vaccination of humans with a recombinant hepatitis B protein discloses an association of the extended haplotype A l , B8, DR3, SCOl with unresponsiveness to the vaccine (Alper et a]. 1989). The use of synthetic and recombinant vaccines will probably reveal other examples of genetic control since, in these cases, a smaller number of epitopes instead of the enormous complexity of whole pathogens is presented to the immune system. The precise identification of the mechanisms involving the low responders pattern may give new insights for the comprehension of the development of infectious diseases and ultimately lead to some modification of the vaccine in order to achieve an even broader efficacy.

References ALBERTE.D., BAURM.P., MAYRW.R. (eds) (1984) Histocompatibility Testing 1984, SpringerVerlag, Berlin ALPERC.A., KRUSKALL M.S., MARCUS-BAGLEY D. et al. (1989) Genetic prediction of nonresponse to hepatitis B vaccine. New England Journal of Medicine 321, 708 BALLOUW.R., SHERWOOD J.A., NEVA F.A. et al. (1987) Safety and efficacy of a recombinant D N A Plasmodium falciparum sporozoite vaccine. The Lancet i, 1277 BATCHELOR J.R. & MCMICHAELA.J. (1987) Progress in understanding HLA and disease associations. British Medical Bulletin 43, I56 CAMPBELL G.. ALEYS., BALLOUW.R. et al. (1987) Use of synthetic and recombinant peptides in the study of host-parasite interactions in the malarias. American Journal of Tropical Medicine and Hygiene 37,428 COLLINS W.E., PAPPAIOANOU M., ANDERS R.F. et al. (1988) Immunization trials with the ringinfected erythrocyte surface antigen of Plasmodium falciparum in owl monkeys (Aotus vociferans). American Journal of Tropical Medicine and Hygiene 38, 268 GOODM.F., BERZOFSKY J.A., LEEMALOYW. et al. (1986) Genetic control of the immune response in mice to a Plasmodiumfalciparum sporozoite vaccine. Journal of Experimental Medicine 164, 655 HOUGHTEN R.A. (1985) General method for the rapid solid phase synthesis of the large numbers of peptides: specificity of antigen-antibody interaction of the level of individual aminoacids. Proceedings of the National Academy of Science of the United States of America 82, 5131 JENDOUBI M. & PEREIRA DA SILVA L. (1987) Polypeptide antigens M, 90000 and 72000 related to protective immunity against the blood form of P. falciparum in the squirrel monkey show stable characteristics in strains from different geographic origins. American Journal of Tropical Medicine and Hygiene 37, 9 MERRIFIELD R.B. (1963) Solid phase peptide synthesis. I. The synthesis of a tetrapeptide. American Chemical Society 85, 2149

516

M.E.Patarroyo et al.

NUSSENZWEIG R. & NUSSENZWEIG V. (1984) Development of sporozoite vaccines. Philosophical Transactions of the Royal Society, London, 307, 117 PATARROYO M.E., AMADOR R., CLAVIJO P. et al. (1988) A synthetic vaccine protects humans against challenge with asexual blood stages of Plasmodiumfakiparum malaria. Nature (London) 332, 358 SALCEDO M., B m L., ROJASM.et al. (1991) Studies on the humoral immune response to a synthetic vaccine against Plasmodiumfalciparum malaria. Clinical and Experimental Immunology 84,122 STURCHLERD. (1989) How much malaria is there worldwide. Parasitology, Today 5,39 TERASAKIP.I., BERNOWD., PARKM.S.et al. (1978) Microdroplet testing of HLA-A, -B, -C and -D antigens. American Journal of Clinical Pathology 69, 103 WASSOND.L. & KELLY E.A.B. (1990) The role of the major histocompatibility complex in resistence to parasite infections. Cricial Reviews in Immunology 10, 31 WAYNEW.D. (1980) Bioslatistics, pp. 325-357, Editorial Limusa, Mexico.