Antigenicity studies in humans and immunogenicity studies in mice: an ...

Microbes and Infection 16 (2014) 419e428 www.elsevier.com/locate/micinf

Original article

Antigenicity studies in humans and immunogenicity studies in mice: an MSP1P subdomain as a candidate for malaria vaccine development Yang Cheng a, Eun-Hee Shin b,c, Feng Lu a,1, Bo Wang a, Jongseon Choe d, Takafumi Tsuboi e, Eun-Taek Han a,* a

Department of Medical Environmental Biology and Tropical Medicine, School of Medicine, Kangwon National University, Chuncheon, Gangwon-do 200-701, Republic of Korea b Department of Parasitology and Tropical Medicine, Seoul National University College of Medicine, and Institute of Endemic Disease, Seoul National University Medical Research Center, Seoul 110-799, Republic of Korea c Seoul National University Bundang Hospital, Seongnam 463-707, Republic of Korea d Department of Microbiology and Immunology, School of Medicine, Kangwon National University, Chuncheon, Gangwon-do 200-701, Republic of Korea e Division of Malaria Research, Proteo-Science Center, Ehime University, Matsuyama, Ehime 790-8577, Japan Received 5 June 2013; accepted 11 February 2014 Available online 20 February 2014

Abstract The newly identified GPI-anchored Plasmodium vivax merozoite surface protein 1 paralog (MSP1P) has a highly antigenic C-terminus that binds erythrocytes. To characterize the antigenicity and immunogenicity of two regions (PvMSP1P-19 and -33) of the highly conserved Cterminus of MSP1P relative to PvMSP1-19, 30 P. vivax malaria-infected patients and two groups of mice (immunized with PvMSP1P-19 or -33) were tested for IgG subclass antibodies against PvMSP1P-19 and -33 antigens. In the patients infected with P. vivax, IgG1 and IgG3 levels were significantly higher than those levels in healthy individuals, and were the predominant response to the two C-terminal fragments of PvMSP1P ( p < 0.05). In mice immunized with PvMSP1P-19, IgG1 levels were the highest while IgG2b levels were similar to IgG1 levels. The levels of Th1 cytokines in mice immunized with PvMSP1P-19 or -33 were significantly higher than those in mice immunized with PvMSP1-19 ( p < 0.05). Our results indicate that: (i) IgG1 and IgG3 (IgG2b in mice) are predominant IgG subclasses in both patients infected with P. vivax and mice immunized with PvMSP1P-19 or -33; (ii) the C-terminus of MSP1P induces a Th1-cytokine response. This immune profiling study provides evidence that MSP1P may be a potential candidate for vivax vaccine. Ó 2014 Institut Pasteur. Published by Elsevier Masson SAS. All rights reserved.

Keywords: Plasmodium vivax; MSP1P; IgG subclasses; Cytokines; Immunogenicity

1. Introduction Vivax malaria threatens almost 40% of the world’s population, but only two Plasmodium vivax vaccines are in preliminary (phase _) clinical trials [1]. Meanwhile, resistance of P. vivax to first-line antimalarial drugs such as chloroquine is increasing, and the resistance of mosquitoes to insecticide has exacerbated the threat of malaria. To date, there have been no effective * Corresponding author. Tel.: þ82 33 250 7941; fax: þ82 33 255 8809. E-mail addresses: [email protected], [email protected] (E.-T. Han). 1 Present address: Jiangsu Institute of Parasitic Diseases, Wuxi, Jiangsu, People’s Republic of China.

malaria vaccines [2]. These issues emphasize the urgent need for an effective malaria vaccine. The P. vivax merozoite surface protein 1 paralog (MSP1P) is a glycosylphosphatidylinositol (GPI)-anchored blood-stage protein, but is not closely related to MSP1 (11% identity and 22% similarity) [3,4]. PvMSP1P is located on the surface of merozoites, and the C-terminus of PvMSP1P is highly antigenic and binds erythrocytes (RBCs) [5]. This study indicates that MSP1P has potential as a candidate for developing a vivax malaria vaccine. MSP1 shares some characteristics with MSP1P including: (i) a conserved antigenic C-terminal; (ii) location; (iii) an EGF-like domain involved in GPI-anchoring; (iv) ability to

http://dx.doi.org/10.1016/j.micinf.2014.02.002 1286-4579/Ó 2014 Institut Pasteur. Published by Elsevier Masson SAS. All rights reserved.

420

Y. Cheng et al. / Microbes and Infection 16 (2014) 419e428

bind RBCs; and (v) similar molecular weight (about 200 kDa) [3,5,6]. EGF-like domains contain conserved cysteines that can potentially form disulfide bonds, and are predicted to be structurally homologous to surface proteins of Plasmodium falciparum [7]. These regions are obvious candidates for mediating proteineprotein interactions, including receptorbinding function. Some GPI-anchored merozoite surface proteins of P. falciparum are generally refractory to gene deletion, suggesting that they must play important roles in blood-stage development [8]. The C-terminus of MSP1, which includes these common regions, is a leading vaccine candidate for erythrocytic-stage parasites [9]. Moreover, recombinant MSP1-19 can induce immunity in monkey and mouse models [10e12]. The PvMSP1P C-terminus possesses the highest antigenicity and binding capability within MSP1P [5]. However, the question of whether MSP1P is a rational candidate for vivax malaria vaccine development should be addressed in more detail. To characterize the immunogenicity of the two C-terminus regions of MSP1P, splenocytes of mice immunized with these peptides, and the levels of Th1/Th2 cytokines in culture supernatants of splenocytes from the immunized mice were determined. The C-terminus of MSP1P was selected based on: (i) the identification of critical highly antigenic fragments, (ii) the ability of these fragments to bind to RBCs, and (iii) the highly conserved nature of these regions. 2. Materials and methods 2.1. Human sera samples Samples of sera were collected from 30 patients positive for vivax malaria (mean age 27.5 years, range 18e52 years) with symptoms and positivity (mean parasitemia 0.118%, range 0.027%e0.477%) identified by microscopy at local health centers and clinics in the Gyeonggi and Gangwon Provinces in endemic areas of the Republic of Korea (ROK). Eight serum samples of healthy individuals (mean age 25 years, range 11e44 years) that were negative for vivax malaria by microscopy were collected in nonendemic areas of the ROK. Sixteen samples were selected randomly from those from the 30 patients positive for vivax malaria, and four normal samples were selected from the eight healthy individuals were used for determination of IgG subclass level. Written informed consent to participate in this study was obtained from all subjects. Approval for the use of the blood samples for this study was obtained from the Kangwon National University Hospital Institutional Review Board.

expressed using a wheat germ cell-free (WGCF) system (CellFree Sciences, Matsuyama, Japan) and purified using Niaffinity chromatography under nondenaturing conditions. Experimental protocols involving animals were approved by the Institutional Animal Care and Use Committee of Kangwon National University, and the experiments were conducted according to the Kangwon National University Ethical Guidelines for Animal Experiments. 2.3. Immunization of mice with MSP1-19, MSP1P-19, and -33 Six- to eight-week-old female BALB/c mice (DBL, Seoul, ROK) were injected intraperitoneally with about 20 mg of recombinant MSP1-19, MSP1P-19, or -33 proteins in phosphate-buffered saline (PBS) with complete Freund’s adjuvant (SigmaeAldrich, St. Louis, MO) in a final volume of 100 ml. Three mice were used per group. Booster injections were given after 3 and 6 weeks using the same amount of antigen in PBS with incomplete Freund’s adjuvant (SigmaeAldrich). Mouse blood samples were taken 2 weeks after the final booster. 2.4. SDS-PAGE and Western blot analysis Ten micrograms of each recombinant PvMSP1P protein was prepared in reducing sample buffer, separated by 12% SDS-PAGE, and then stained with Coomassie Brilliant Blue. For Western blot analysis, the proteins were transferred electrophoretically to PVDF membranes (Millipore, Bedford, MA), and incubated with blocking buffer (5% nonfat dry milk in PBS containing 0.2% Tween 20, PBS/T) for 1 h at 37 C. The blots containing recombinant proteins were then incubated for 1 h at 37 C with either anti-penta-His antibody (Qiagen, Hilden, Germany) or antibodies against each recombinant PvMSP1P fragment diluted 1:1000 in PBS/T. The membranes were washed with PBS/T and incubated with a 1:2000 dilution of alkaline phosphatase-conjugated goat antimouse IgG (MP Biomedicals, Solon, OH) for 1 h at 37 C. The blots were washed with PBS/T and developed with 5bromo-4-chloro-3-indolyl phosphate/nitro blue tetrazolium (BCIP/NBT, SigmaeAldrich). After the primary antibody reactions and washing, the membrane was incubated with secondary IRDye goat anti-mouse (1:10,000 dilution) (LI-COR Bioscience, Lincoln, NE) antibodies for detecting His-tagged recombinant protein to determine immune serum specificity. Data were obtained using the Odyssey infrared imaging system (LI-COR Bioscience) and analyzed with Odyssey software (LI-COR Bioscience).

2.2. Expression of PvMSP1 and PvMSP1P protein fragments

2.5. Vivax patient serum screening using protein arrays

Previously, we divided PvMSP1P (PlasmoDB PVX_099975) into eight small fragments based on MSP1 structure [4], and recombinantly expressed a tag of six histidines as described previously [5]. In brief, the two MSP1P Cterminus fragments, MSP1P-19 and -33 (Fig. 1A) were

Sera from 30 patients infected with vivax malaria were tested against the recombinant PvMSP1P-19 and -33 proteins using protein arrays as previously described [5]. Briefly, 1 ml of 50 ng/ml recombinant PvMSP1P-19 or -33 proteins were spotted on PVDF membranes. Then, nonspecific binding sites


A

TR

SP

421

EGF EGF GPI

PR

NH

COOH MSP1P-33 (274 aa)

B

rProtein 33

C

19

α-M

α-His

50

kDa 250 150 100 75 50

37

37

25 20

25

kDa 250 150 100 75

33

19

MSP1P-19 (86 aa)

kDa

33 19

250 150 100 75 50 37

20 15

15

10

10

25 20 15

10

Fig. 1. Schematic diagram showing expression of PvMSP1P. (A) Schematic representation of PvMSP1P, the two constructs of MSP1P C-terminus used for serological and cellular analysis. (B) Expression and purification of recombinant PvMSP1P-33 and -19 fragments. Both PvMSP1P-33 (about 40 kDa) and -19 (about 14 kDa) tagged with His were synthesized using a wheat germ cell-free protein expression system, then purified using a Ni-Sepharose column. These two purified proteins existed in soluble elution fractions and were resolved by 12.5% SDS-PAGE. (C) Recombinant PvMSP1P-33 and -19 proteins were probed with anti-His antibody (a-His) and immune mouse sera (a-M) under reducing conditions. Arrows indicate the target bands for native and recombinant proteins. rProtein, recombinant protein; aa, amino acid; kDa, kilodalton; TR, tandem repeat region of heptapeptide; PR, polymorphic region; SP, signal peptide; EGF, epidermal growth factor-like domains; GPI, glycophosphatidylinositol.

on the arrays and their matrix were blocked with 5% BSA in PBS/T for 1 h at 37 C. After probing with human serum (1:10), antibodies were visualized with 10 ng/ml Alexa Fluor 546 goat anti-human IgG (Invitrogen, Carlsbad, CA) in PBS/T and scanned in a fluorescence scanner (ScanArray Express, PerkinElmer, Boston, MA). Fluorescence intensities of array spots were quantified using the fixed circle method with ScanArray Express software (version 4.0, PerkinElmer). 2.6. Indirect immunofluorescence assay (IFA) Blood rich in schizont-stage parasites of P. vivax was collected from a malaria patient in Thailand. Slides smeared with parasite-infected blood were fixed with ice-cold acetone for 3 min, dried, and stored at 80 C. Before use, the slides were thawed on blue silica gel (Samchun Chemical, Pyeongtaek, Gyeonggi, ROK) and nonspecific binding sites blocked with PBS containing 5% nonfat dry milk at 37 C for 30 min. Then, the samples were incubated with primary antibodies (rabbit anti-MSP1-19 [1:100 dilution] and mouse anti-MSP1P19 [1:50 dilution]) at 37 C for 1 h. After the primary antibody reactions, the samples were treated with Alexa 546-conjugated goat anti-mouse IgG secondary antibody or Alexa 488-conjugated goat anti-rabbit IgG secondary antibody (Invitrogen), and 40 ,6-diamidino-2-phenylindole (DAPI, Invitrogen) to stain nuclei, at 37 C for 30 min. The slides were mounted with ProLong Gold antifade reagent (Invitrogen) and visualized under oil immersion using a confocal laser scanning microscope (FV200; Olympus, Tokyo, Japan) equipped with 20 dry and 60 oil objectives. Images were captured using

FV10-ASW 3.0 viewer software and prepared for publication with Adobe Photoshop CS5 (Adobe Systems, San Jose, CA). 2.7. Enzyme-linked immunosorbent assay (ELISA) To investigate the prevalence of various IgG subclasses against MSP1P-19 and -33, sera from 30 vivax-positive patients and eight negative serum samples were selected. Briefly, 2.5 mg/ml of MSP1-19, 5 mg/ml of MSP1P-19, and 5 mg/ml of MSP1P-33 in coating buffer (50 mM bicarbonate buffer, pH 9.5) were incubated in 96-well ELISA plates (Costar, Corning, NY) for 2 h at room temperature. Nonspecific binding sites of the plates were blocked with 5% nonfat dry milk in PBS/T (0.1% Tween-20) for 1 h at 37 C, and then the plates incubated with 100 ml of individual sera diluted 1:400 in PBS/T. Horseradish peroxidase (HRP)-conjugated antihuman IgG1, IgG2, IgG3, and IgG4 antibodies (Invitrogen) diluted 1:1000 in PBS/T were used for detection. Chromogenic reactions were developed as described previously [13]. The cutoff value was the mean plus two standard deviations (SD) of the optical density at 450 nm (OD450) of eight negative samples. Sixteen serum samples from patients infected with P. vivax were randomly selected from the 30 patient samples; and four samples from unexposed subjects were selected as controls. The IgG subclass distribution was determined using a human IgG subclass profile kit (Invitrogen), according to the manufacturer’s instructions. Briefly, a series of concentrations of IgG1, IgG2, IgG3, and IgG4 standards was established. The concentrations used were IgG1 (13.72, 6.86, 3.43, 1.72, 0.86, and 0.43 mg/ml), IgG2 (5.32, 2.66, 1.33, 0.67, 0.33, and


2.8. T-cell assays Spleens were removed from mice 2 weeks after their third immunization. Splenocytes were resuspended at 5 105 cells/ ml in RPMI (Gibco, Invitrogen) supplemented with 1 antibioticeantimycotic (Gibco), and 10% fetal bovine serum (Gibco). One-hundred-microliter aliquots of cell suspensions were distributed into round-bottom 96-well microculture plates (Costar), and 100 ml (5 mg/ml) of each antigen was added. Meantime, splenocytes were stimulated with 5 mg/ml of concanavalin A (SigmaeAldrich) or 10 mg/ml of lipopolysaccharides (LPS, SigmaeAldrich), which functioned as positive controls, whereas medium alone was a negative control. Cultures were assayed in triplicate. After 72 h with the cells at 2.5 105/well (at 37 C and 5% CO2), around 150 ml of supernatant/well was collected and stored at 70 C for cytokine determination. Fifty microliters of enhanced cell

viability buffer (Daeil Lab Service, Seoul, ROK) was added to 50 ml of splenocyte culture medium for the final 4 h of a 72 h incubation period. The OD at 450 nm of each well was measured with reference to the OD at 650 nm using an ELISA plate reader. The stimulation index (SI ) was calculated as follows: mean OD450 of triplicate test wells/mean OD450 þ two standard deviations of six control wells. Proliferation was considered positive when the SI was >1. Cytokine concentrations in culture supernatants and sera from immunized mice were assayed using the BD CBA Flex Set kit (BD Biosciences, San Jose, CA). The cytokines assayed included mouse gamma interferon (IFN-g), tumor necrosis factor (TNF), interleukin-12p70 (IL-12p70), IL-2, IL-4, and IL-10. The results were acquired on a FACSAriaII Cell Sorter (BD Biosciences) according to the manufacturer’s

A

15000

R2 < 0.1

Human IgG M SP1P-33 (M FI)

0.17 mg/ml), IgG3 (1.34, 0.67, 0.34, 0.17, 0.084, and 0.042 mg/ml), and IgG4 (0.76, 0.38, 0.19, 0.095, 0.048, and 0.024 mg/ml) in 96-well ELISA assay plates. Human subclassspecific antibodies (50 ml/well) were added to wells coated with each antibody and then incubated with 50 ml of a 1:2000dilution of serum samples, standards, and the ready-to-use human serum controls. Following aspiration of the samples, the wells were washed with 200 ml of washing buffer, and 100 ml of a 1:50 dilution of HRP-conjugated anti-human IgG antibody solution was added. Color was developed with 100 ml of 3,30 ,5,50 -tetramethylbenzidine solution (Invitrogen) and absorbance at 450 nm was measured within 1 h of adding the stop solution. The results were calculated using a logelog curve fit. To detect mouse immune serum titers, MSP1-19 (2.5 mg/ ml), MSP1P-19 and -33 (5 mg/ml/sample) were coated on ELISA plates as previously described [14]. After blocking with 5% nonfat dry milk in PBS/T, a twofold serial dilution of anti-MSP1-19, -MSP1P-19, and -MSP1P-33 mouse sera was add to each well. HRP-conjugated anti-mouse IgG antibody (H þ L) (dilution 1:10,000) (Pierce Biotechnology, Rockford, IL) was added to each well as a secondary antibody for 1 h at 37 C. The reaction was developed by adding 100 ml of diluted TMB solution (Invitrogen) for 15 min at 37 C and then stopped with 100 ml 1 N HCl, before optical density (OD) was measured at 450 nm. All samples were tested in duplicate, and the mean absorbance was calculated. The ELISA titer was the dilution at which the absorbance unit was nearest to 1.0. An ELISA was also developed for quantitative analysis of the mouse IgG subclass. Briefly, to construct a standard curve, 100 ml of purified mouse IgG1, IgG2a, IgG2b, or IgG3 (Invitrogen) was coated onto 96-well plates at 256, 128, 64, 32, 16, 8, and 4 ng/ml, and incubated with immune mouse sera diluted 1:1000 in PBS/T. HRP-conjugated anti-mouse IgG1, IgG2a, IgG2b, and IgG3 antibodies (Invitrogen) at 1:1000, 1:1000, 1:2000, and 1:1000 dilutions, respectively, were used to detect reactions. The color intensity was measured and calculated using a logelog curve fit.

10000

5000

0 0.0

0.2

0.4

0.6

Parasitemia %

B

25000

R2 < 0.1

Human IgG to M SP1P-19 (M FI)

422

20000

15000

10000

5000

0 0.0

0.2

0.4

0.6

Parasitemia %

Fig. 2. Correlation between patient peripheral blood parasitemia and mean immunoreactive fluorescence intensity (MFI). Correlation between immunoreactivity of total IgG against PvMSP1P-33 (A) and -19 (B) and parasitemia of each vivax patient sample was determined for 30 vivax patient samples. R2 was calculated using a polynomial.


423

instructions, and analyzed using FCAP array software (Soft Flow, Kedves, Hungary).

significant. Simple scatter-regression was used to make a standard curve.

2.9. Statistical analysis

3. Results

Data were analyzed using GraphPad Prism software (San Diego, CA), SigmaPlot (Systat Software, San Jose, CA) and Microsoft Excel 2007 (Microsoft, Redmond, WA). ManneWhitney U tests were used to compare the differences between the means of each group in terms of their statistical significance. Differences of p < 0.05 were considered

3.1. Expression of C-terminus of PvMSP1P, and production of immune sera

A

Recombinant proteins were successfully purified under nondenaturing conditions as shown in Fig. 1B. The corresponding immunoblots were probed with an anti-His tag

70%

1.2

73% 27%

1.0

43%

0.8

20%

O.D. 450 nm

0.6

23%

56%

0.4

57%

0.3

10%

20% 0%

0.2 23% 0.1

0.0

IgG1

IgG2

IgG3

IgG4

IgG1

MSP1-19

IgG2

IgG3

IgG4

IgG1

MSP1P-19

IgG2

IgG3

IgG4

MSP1P-33

Human IgG subclasses to PvMSP1P sub-domain antigens

B

25

p < 0.01

IgGs conc. (mg/ml)

20

p = 0.1

15 10 p < 0.05

5 2.0

p = 0.12

1.5 1.0 0.5 0.0

Healthy

Patients

IgG1

Healthy

Patients

IgG2

Healthy

Patients

IgG3

Healthy

Patients

IgG4

Fig. 3. The levels of IgG subclasses in sera from patients from the Republic of Korea who were positive for P. vivax infection and in sera from healthy individuals in response to PvMSP1P. (A) Immunoreactivity and levels of IgG subclasses against each antigen in the sera of malaria patients from ROK was determined. The level of antibodies to PvMSP1-19, PvMSP1P-19, and -33 in sera from vivax malaria patients was determined by ELISA. Vivax patients (n ¼ 30; mean age 27.5 years, range 18e52 years) and unexposed healthy controls (n ¼ 8; mean age 28.5 years, range 23e45 years) were randomly selected for IgG subclass prevalence studies for antibodies against MSP1-19, MSP1P-19, and -33. The cutoff value was the mean plus two standard deviations (SD) of the optical density at 450 nm (OD450) of eight negative samples. (B) IgG subclass levels were determined in 16 of the 30 vivax patient samples and four of the samples from eight healthy individuals. The mean levels of IgG subclasses in vivax patient samples were IgG1 16.7 (range 11.9e19.7) mg/ml, IgG2 6.7 (4.1e11.2) mg/ml, IgG3 1.7 (1.1e2.2) mg/ml, and IgG4 0.9 (0.1e1.9) mg/ml. IgG levels in the sera of healthy individuals were: IgG1 9.2 (8.2e12.0) mg/ml, IgG2 4.8 (3.2e7.5) mg/ml, IgG3 0.3 (0.3e0.6) mg/ml, and IgG4 0.2 (0.1e0.3) mg/ml. Bars indicate the IgG subclass concentration (mg/ml) and its percentage in each sample. Differences between IgG subclass levels in the negative and positive groups were analyzed using ManneWhitney U tests. P values for IgG1 and IgG3 were 0.05. P < 0.05 was considered a significant difference.

424


monoclonal antibody, and anti-rPvMSP1P immune mouse serum revealed a similar and specific pattern of migration for PvMSP1P (Fig. 1C). Preimmune mouse serum samples were used as a negative control (data not shown). 3.2. Correlation between anti-PvMSP1P antibody levels and parasitemia To clarify whether anti-PvMSP1P antibody levels correlate with protection, antibody responses against two antigens were compared with the individual parasitemia of the 30 patients (Fig. 2). Although there was no significant inverse correlation between parasitemia and antibody levels against either PvMSP1P-33 (R2 < 0.1) or 19 (R2 < 0.1) antigens, there was no case with high antibody levels with high parasitemia (Fig. 2A and B, top right quadrant). These findings suggest

that the level of anti-PvMSP1P antibody may not strongly contribute to the inhibition of parasite growth. 3.3. IgG subclass against PvMSP1P C-terminus in vivax patients We analyzed the prevalence of each IgG subclass for antibodies against MSP1-19, MSP1P-19, and -33. We found that IgG3 was the predominant IgG subclass for antibodies against MSP1P-19 (73%) and MSP1-19 (70%) and IgG1 was the subdominant IgG subclass for antibodies against MSP1P-33 (prevalence 57%) and MSP1P-19 (43%) (Fig. 3A). Compared with MSP1-19, all the IgG subclasses were more prevalent for antibodies against MSP1P-19, but only IgG1 and IgG4 antibodies were more prevalent against MSP1P-33. Nevertheless, the IgG subclass that predominated antibodies

A 600 500 400 300

IgGs conc. (ug/ml)

200

100

50

0

IgG1 IgG2a IgG2b IgG3

MSP1-19



MSP1P-19

MSP1P-33

B

C

1000000

IgG3 IgG2b IgG1 IgG2a

Titers of total IgG

Mouse total IgG

100000

10000

1000

100

10

1

MSP1-19 MSP1P-19 MSP1P-33

MSP1-19 MSP1P-19 MSP1P-33

Fig. 4. IgG subclass levels in immune mouse samples, their percentages, and total IgG titers. Three groups of three mice were immunized with PvMSP1-19, PvMSP1P-33, or -19, respectively (A). Isotypic distribution of the IgG responses to the MSP1-19, MSP1P-19 and -33 antigens. Geometric means were calculated for each of the four subclasses in each group of three mice and are presented as percentages of the total IgG responses in BALB/c mice (B). Total IgG responses against PvMSP1-19 and the C-terminal of the PvMSP1P in three BALB/c mice per group immunized with PvMSP1-19 and the PvMSP1P C-terminalspecific antigens, respectively (C). IgG responses were analyzed by ELISA after the final immunizations against MSP1-19, MSP1P-19 and -33. The results are expressed as mean titers SD.


against MSP1P-19 and -33 differed, whereas MSP1P-19 and MSP1-19 shared a similar predominant subclass of antibodies. To determine the IgG subclass concentrations in the sera of vivax patients, we evaluated 16 vivax patients and four serum samples from healthy individuals. The mean concentrations of IgG1, IgG2, IgG3, and IgG4 were 16.7, 6.7, 1.0, and 0.5 mg/ ml, respectively, compared with 9.2 mg/ml, 4.8 mg/ml, 0.4 mg/ml, and 0.25 mg/ml, respectively, in the four samples from healthy individuals (Fig. 3B). Importantly, despite the high prevalence of IgG3 antibodies against MSP1-19 and MSP1P-19 in vivax patients, the concentration of IgG3 in vivax patient sera was not high. Nevertheless, our data clearly show that the levels of IgG1 ( p < 0.01) and IgG3 ( p < 0.05) in patients with vivax infections were significantly greater than those in healthy controls. 3.4. IgG1 and IgG2b are the predominant class of antibodies against the C-terminus of MSP1P in immunized mice We determined the levels of IgG subclasses for antibodies in sera from mice immunized against three antigens. IgG1 concentrations were highest in the three groups of immunized mice, although IgG1 concentrations were similar to those of IgG2b in mice immunized with MSP1P-19. The level of IgG2b antibodies against MSP1P-19 (mean concentration 447.0 mg/ml) was notably higher than those of IgG2b antibodies against MSP1-19 (39.3 mg/ml) and MSP1P-33 (22.0 mg/ml) (Fig. 4A). Fig. 4B shows that noncytophilic antibodies, IgG1 plus IgG2b, were major components of the antibody response in mice immunized with MSP1-19 and MSP1P-33. There was no significant difference between the

425

levels of cytophilic and noncytophilic isotype from sera of mice immunized with MSP1P-19. 3.5. Titers of IgG antibodies against the C-terminus of MSP1P compared with those against MSP1-19 The titers of IgG antibodies against MSP1-19, MSP1P-19, and -33 were 130,000 46,000, 80,000 10,000 and 90,000 50,000 (mean SD), respectively (Fig. 4C). The two C-terminal fragments of MSP1P (MSP1P-19 and -33), clearly induced an immune response as large as that seen in mice immunized with MSP1-19. 3.6. PvMSP1P is a merozoite surface protein IFA was conducted using anti-PvMSP1 and anti-PvMSP1P sera. In parasites, PvMSP1P was localized to the merozoite surface (Fig. 5A). To confirm the merozoite surface localization of PvMSP1P, we compared with merozoite surface protein PvMSP1-19 in the schizont stage (Fig. 5B) using preimmune mouse serum as a negative control (Fig. 5C). 3.7. Effect of PvMSP1-19, PvMSP1P-19, and -33 on the Th1/Th2 ratio We assessed the lymphoproliferation of splenocytes taken from immunized BALB/c mice in response to in vitro stimulation with 5 mg/ml of MSP1-19, MSP1P-19, or -33 (Fig. 6). Splenocytes from mice immunized with MSP1-19 showed low proliferative responses (SI range 0.9e2.0), whereas higher lymphoproliferation was observed in response to stimulation with MSP1P-19 (SI range 1.6e3.5), and MSP1P-33 (SI range

Fig. 5. Localization of PvMSP1P in the mature schizont stage. Schizont-stage parasites were labeled with antisera against PvMSP1P-19 (A), dual-labeled with antisera against PvMSP1P-19 and PvMSP1-19 (merozoite surface marker) (B), or dual-labeled with antisera against PvMSP1-19 and preimmune mouse serum (C). Nuclei are visualized with DAPI in merged images. PI, preimmune mouse serum; DIC, differential interference contrast. Bar represents 5 mm.

426


A

5

Stimulation index (SI)

4

3

2

1

MSP1P-19

B

-3 3

M

S

P 1P

M

S

S M

MSP1P-33

P 1P

-1 9

P 119

LP S

on A C

P 1P S

M

M

MSP1-19

-3 3

LP S

on A

S

C

-1 9

LP S

on A

P 1P

S M

C

P 119

LP S

C

on A

0

NI

Immunized mice p < 0.05

p < 0.001

Cytokine conc. (pg/ml)

800

p < 0.001

p < 0.01

p < 0.05 p < 0.001

600 400 200

10 5

M S P 119 M S P 1P -1 9 M S P 1P -3 3 M S P 119 M S P 1P -1 9 M S P 1P -3 3 M S P 119 M S P 1P -1 9 M S P 1P -3 3

0

IL-2

IFN-

C

TNF

p < 0.05 p < 0.01

Cytokine conc. (pg/ml)

400 300 200 100 10 5

19 M S P 1P -1 9 M S P 1P -3 3

M S P 1-

M S P 119 M S P 1P -1 9 M S P 1P -3 3

0

IL-4

IL-10

Fig. 6. Cytokine levels in 72-h culture supernatants of splenocytes from BALB/c mice immunized with either MSP1-19, MSP1P-19, or -33, stimulated in vitro with each protein. (A) The stimulation index (SI ) of splenocytes from mice immunized with MSP1-19, MSP1P-19, or -33 following stimulation with culture medium and 5 mg/ml of recombinant MSP1-19, MSP1P-19, or -33, respectively. Proliferation of splenocytes stimulated with concanavalin A or LPS as positive controls is indicated. The SI was calculated as: (mean OD450 of triple test wells)/(mean OD450 þ two standard deviations of six control wells). Proliferation was considered positive when the SI was >1. The results are expressed as mean SI SD. Th1 cytokine profile (B), and Th2 cytokine profile (C). Shown are individual results for each mouse tested in duplicate with samples taken after the third immunization; the black horizontal bars represent the geometric means of groups of mice. The IL4 levels against MSP1-19 were too low to detect. The p values, calculated using Student’s t test, indicate the number of sera analyzed; p values of