Genetically attenuated P36p-deficient Plasmodium berghei sporozoites confer long-lasting and partial cross-species protection

www.elsevier.com/locate/ijpara

International Journal for Parasitology 37 (2007) 1511–1519

Genetically attenuated P36p-deficient Plasmodium berghei sporozoitesconfer long-lasting and partial cross-species protection

Bruno Douradinha a,b,1, Melissa R. van Dijk c,1, Ricardo Ataide a,Geert-Jan van Gemert d, Joanne Thompson e, Jean-Francois Franetich f,g,

Dominique Mazier f,g,h, Adrian J.F. Luty d, Robert Sauerwein d, Chris J. Janse c,Andrew P. Waters c, Maria M. Mota a,b,*

a Unidade Malaria, Instituto de Medicina Molecular, Faculdade de Medicina da Universidade de Lisboa, 1649-028, Lisboa, Portugalb Instituto Gulbenkian de Ciencia, 2781-156 Oeiras, Portugal

c Department of Parasitology, Leiden University Medical Center, P.O. Box 9600, 2300 RC, Leiden, The Netherlandsd Department of Medical Microbiology, University Medical Centre, P.O. Box 9101, 6500 HB Nijmegen, The Netherlands

e Institute of Immunology and Infection Research, School of Biological Science, University of Edinburgh, The Kings Buildings, Edinburgh EH9 3JT,

United Kingdomf Inserm, U511, and Universite Pierre et Marie Curie-Paris 6, Faculte de Medecine Pitie-Salpetriere, France

g Universite Pierre et Marie Curie-Paris 6, Faculte de Medecine Pitie-Salpetriere, Franceh Assistance Publique-Hopitaux de Paris, Centre Hospitalo-Universitaire Pitie-Salpetriere, F-75013, Paris, France

Received 20 February 2007; received in revised form 18 April 2007; accepted 8 May 2007

Abstract

Immunisation with live, radiation-attenuated sporozoites (RAS) or genetically attenuated sporozoites (GAS) of rodent plasmodialparasites protects against subsequent challenge infections. We recently showed that immunisation with Plasmodium berghei GAS thatlack the microneme protein P36p protects mice for a period of up to 4 months. Here, we show that the period of full protection inducedby p36p�-sporozoites lasts 12 and 18 months in C57Bl6 and BALB/c mice, respectively. Full protection is also achieved with three dosesof only 1000 p36p� (but not RAS) sporozoites. Subcutaneous, intradermal or intramuscular routes of administration also lead to partialprotection. In addition, immunisation with either P. berghei RAS- or, to a lesser extent, p36p�-sporozoites inhibits parasite intrahepaticdevelopment in mice challenged with Plasmodium yoelii sporozoites. Since naturally acquired malaria infections or subunit-based vac-cines only induce short-term immune responses, the protection conferred by immunisation with p36p�-sporozoites described here furtheremphasises the potential of GAS as a vaccination strategy for malaria.� 2007 Australian Society for Parasitology Inc. Published by Elsevier Ltd. All rights reserved.

Keywords: Malaria; Plasmodium; Sporozoites; RAS; GAS; Vaccine

1. Introduction

Naturally acquired infections with plasmodial parasites,as well as human immunisation strategies using parasite

0020-7519/$30.00 � 2007 Australian Society for Parasitology Inc. Published b

doi:10.1016/j.ijpara.2007.05.005

* Corresponding author. Address: Unidade Malaria, Instituto de Med-icina Molecular, Faculdade de Medicina da Universidade de Lisboa, 1649-028 Lisboa, Portugal. Tel.: +351 21 799 9509; fax: +351 21 799 9504.

E-mail address: [email protected] (M.M. Mota).1 These authors contributed equally to this work.

peptides or proteins, induce partial and mainly short-livedbut not sterile protection in humans (Moorthy et al., 2004;Schofield and Grau, 2005; Hill, 2006; Marsh and Kin-yanjui, 2006; Schofield and Mueller, 2006). To date, theonly single immunisation approach that has been shownto confer sterile, lasting protection against Plasmodium inmice, monkeys and humans, is immunisation with livesporozoites that have been attenuated by c-radiation (Nus-senzweig et al., 1967; Collins and Contacos, 1972; Clydeet al., 1973). These radiation-attenuated sporozoites

y Elsevier Ltd. All rights reserved.

1512 B. Douradinha et al. / International Journal for Parasitology 37 (2007) 1511–1519

(RAS) are able to invade but not fully mature inside hepa-tocytes (Silvie et al., 2002). Plasmodium falciparum RASimmunisations have conferred long-lasting protection tohuman volunteers against challenge with infectious spor-ozoites, lasting up to 101

2months (Hoffman et al., 2002).

Recently, it has been shown in the Plasmodium bergheirodent model of malaria that protective immunity can alsobe induced by immunisation with live sporozoites that areattenuated by genetic modification. (Mueller et al.,2005a,b; van Dijk et al., 2005). These genetically attenuatedsporozoites (GAS) that lack specific proteins expressed inthe sporozoite/young liver stage, were able to induce as yetuncharacterised but protective and stage-specific immuneresponses that remained effective for periods between 1and 4 months against subsequent challenge with infectioussporozoites. Interestingly, the investigation of invasion andgrowth characteristics of three different GAS (Muelleret al., 2005a,b; van Dijk et al., 2005) has shown these to beshort-lived (6–12 h) compared with the longer-lived RAS(Scheller and Azad, 1995). In this study, we analysed severalaspects of the protection induced by GAS, which lack themicronemal protein P36p. (Ishino et al., 2005; van Dijket al., 2005), including the influence of the dose and admin-istration routes of the GAS, as well as its longevity andcross-species protective nature. This study demonstratesthat as few as 1,000 p36p� GAS generate fully protectiveimmunity in C57Bl6 mice when administered intravenously.Furthermore, partial protection is obtained through clini-cally acceptable routes of sporozoite administration andagainst a heterologous species, Plasmodium yoelii. The find-ings highlight the potential of attenuated sporozoite vacci-nation for induction of long-lasting and cross-speciesprotection and have important implications for the develop-ment of new strategies for vaccination against malaria.

2. Materials and methods

2.1. Mice

BALB/c and C57BL6 were bred in the animal facility ofInstituto Gulbenkian de Ciencia (Oeiras, Portugal) or pur-chased from Harlan Iberica (Barcelona, Spain). Six- toeight-week old mice were used (except in the long-lastingprotection experiments). All animal care and procedureswere in accordance with European regulations.

2.2. Parasites

Immunisations were performed with P. berghei sporozo-ites lacking P36p and expressing GFP (p36p� GAS) (vanDijk et al., 2005). Challenge of the immunised mice with‘wild-type’ sporozoites were performed either with P. berg-

hei sporozoites expressing green fluorescent protein (GFP)(PbGFPcon; ANKA strain) (Franke-Fayard et al., 2004)or with P. yoelii sporozoites (265 BY strain). All parasitelines were maintained by alternating passage of the parasitesin Anopheles stephensi mosquitoes and mice. Sporozoites for

immunisation/challenge studies were obtained by dissectionof infected mosquitoes 18–21 days (P. berghei) or 14–17 days(P. yoelii) after the infectious blood meal (Ozaki et al., 1984).

2.3. Immunisation with P. berghei p36p� GAS or RAS and

challenge of immunised mice with PbGFPcon sporozoites

BALB/c and C57Bl6mice were immunised by i.v. injec-tion of p36p� GAS, PbGFPcon RAS (16 krad) (Orjihet al., 1982) or PBS (control). Immunised mice were mon-itored to confirm complete absence of blood stage parasite-mia by fluorescence activated cell sorting (FACS) analysisof tail blood (Franke-Fayard et al., 2004) from day 3 upto the challenge. Different immunisation doses, immunisa-tion protocols and administration routes were used asdescribed. Immunised mice were challenged by i.v. injec-tion of purified PbGFPcon sporozoites or by the bites ofinfectious mosquitoes. Blood stage parasitemias were mon-itored by FACS analysis (Franke-Fayard et al., 2004) fromday 3 up to 2–3 weeks after challenge. The prepatent periodis defined as the period of time between challenge and thefirst appearance of blood-stage parasites.

2.4. Immunisation with p36p� GAS and challenge of

immunised mice with P. yoelii sporozoites

In two independent experiments, groups of BALB/c andC57Bl6 mice were immunised i.v. with p36p� GAS orPbGFPcon RAS or with PBS. Mice were challenged i.v.with 100, 1,000 or 10,000 sporozoites of P. yoelii 3 weekslater. Animals were monitored either for blood stage para-sitemia by Giemsa stained blood smears starting from day3 until 2 weeks after challenge or were examined for P. yoe-

lii liver infection using quantitative RT-PCR (qRT-PCR).Control immunised groups were challenged with 10,000P. berghei PbGFPcon sporozoites. Peak parasitemia isthe maximum parasitemia reached during the period afterchallenge and recovery of the mice from the P. yoelii

blood-stage infection.

2.5. Quantification of P. yoelii liver-infection by qRT-PCR

Livers were removed from mice immunised with P. berg-

hei p36p� GAS, PbGFPcon RAS or PBS 40 h after chal-lenge with P. yoelii sporozoites. Liver infection wasquantified by qRT-PCR of parasite-specific 18S rRNA asdescribed elsewhere (Bruna-Romero et al., 2001). Protec-tion is defined as the percentage of liver infection inhibi-tion, and was calculated as follows: (1 � rRNAimmunised/rRNAnaive) · 100. rRNAimmunised and rRNAnaive representthe number of copies of parasite 18S rRNA in the liversof immunised and control mice, respectively.

2.6. Statistical analysis

Student’s t test (paired, two tails) was performed withthe Prism version 4.03 software (GraphPad Software, San

B. Douradinha et al. / International Journal for Parasitology 37 (2007) 1511–1519 1513

Diego, CA). P values of less than 0.05 were considered sta-tistically significant.

3. Results

3.1. Long-lasting protective immune responses in both

BALB/c and C57Bl6 mice after immunisation with p36p�

GAS

p36p� GAS-immunised BALB/c mice that were pro-tected up to 4 months against infection with wild-type P.

berghei sporozoites (van Dijk et al., 2005) were re-chal-lenged i.v. with 10,000 PbGFPcon parasites 6 months(180 days) after the final immunisation and remained fullyprotected. None developed a blood-stage parasitemia up to3 weeks after challenge whereas blood-stage parasites couldbe observed in all control mice on day 3 (Table 1, group1).

The same group of protected BALB/c mice was subse-quently re-challenged i.v. with 20,000 PbGFPcon sporozo-ites nearly 12 months (351 days) and 18 months afterimmunisation. Since the age of the mice might influence par-asitemia (Greenberg et al., 1953; Singer and Hadfield, 1955;Pierrot et al., 2003), a higher dose of sporozoites (20,000instead of 10,000) was used. All p36p�immunised mice werefully protected up to 18 months (Table 1; group 1).

Three immunisation doses of RAS or p36p� GAS wererequired to protect C57Bl6 mice up to at least 1 monthafter the final immunisation (van Dijk et al., 2005). We alsoinvestigated the persistence of these protective immuneresponses. After the final immunisation, the mice were sub-sequently challenged either 2 and 3 months later (10,000sporozoites) or 6 and 12 months later (20,000 sporozoites).Full protection was observed in all p36p� GAS- or RAS-immunised mice challenged at 2, 3 or 6 months (Table 1,group 2). Of the mice that were challenged after 12 months,50% were fully protected whilst the others developed para-sitemia. RAS-immunised mice were 100% protected.Patency in all p36p� GAS-immunised mice that developed

Table 1Protection of mice immunised with p36p� GAS or RAS against challenge wit

Group Mousestrain

Immunisationa,RAS/p36p� · 103

Challengeb · 103 Daf

1d BALB/c 50 10 101 BALB/c 50 10/20/20 182d C57Bl6 50/20/20 10/10 102 C57Bl6 50/20/20 10/10/20/20 603 BALB/c 50/20/20 10/20/20 604 BALB/c 50/20/20 10/20/20 905 BALB/c 50/20/20 20/20 18

ND, not determined.a Groups of mice were immunised i.v. with one dose of RAS or p36p� GAS is

or p36p� GAS were performed with weekly intervals.b Mice were challenged with PbGFPcon sporozoites, isolated from different m

after challenge.c Control mice used were from the same age as immunised mice at the timed These groups of mice have been previously challenged at indicated time poe During the course of the experiment some mice died from natural, malar

groups.

parasitemia was significantly delayed compared with thecontrol mice (4.5 versus 3.0 days, P < 0.01), suggesting per-sistence of immune responses conferring partial protection.

3.2. BALB/c mice immunised with p36p� GAS do not require

repeated challenge with sporozoites to remain fully protected

up to 6 months

Above, immunised mice were repeatedly challenged withsporozoites. These repetitive challenge infections maythemselves boost the immune response, with a consequentinfluence on the long-lasting protection observed. There-fore, RAS- and p36p� GAS- immunised BALB/c mice(Table 1, groups 3–5; three subsequent immunisationdoses) were challenged only once, at different time pointsafter immunisation, with either 10,000 (at 2 or 3 months)or with 20,000 sporozoites (at 6 months). None of theimmunised mice developed parasitemia whereas all controlmice developed parasitemias at day 3 or 4, demonstratingthat protection does not require repeated sporozoite chal-lenge to be maintained for up to 6 months in BALB/c micethat have received one primary and two boost immunisa-tions. However, not all p36p� GAS-immunised mice werefully protected when challenged after 12 months. Regard-less of the challenge history, two out of six mice developedparasitemia (groups 3–5, Table 1). Thus, protection wanesafter 6 months but is not abolished. However, the prepat-ent period in mice challenged after 12 months (group 4,Table 1) is significantly longer than in controls (5.5 versus3.0 days, P < 0.01).

3.3. Three doses of 1,000 p36p� GAS are able to fully protect

mice against challenge with sporozoites

In three independent experiments, C57Bl6 mice wereimmunised with different doses of RAS or p36p� GAS(3 · 10,000; 3 · 5,000 or 3 · 1,000, at weekly intervals),and then challenged i.v. with 10,000 PbGFPcon sporozo-

h PbGFPcon sporozoites

ay of challengeter the last immunisation

No. protected (no. challenged)

Controlc RAS p36p�

–120 0(5) ND 5(5)0/351/531 0(5) ND 4(4)/3(3)e

/30 0(5) 5(5) 4(4)/90/180/365 0(5) 5(5) 2(4)/180/365 0(5) 3(3) 4(4)/1(2)e

/180/365 0(5) 3(3)/NDe 3(3)/2(2)e

0/365 0(5) ND 4(4)/1(2)e

olated from different mosquito batches. Multiple immunisations with RAS

osquito batches. All non-immunised mice become positive on day 3 or 4

of challenge.ints after immunisation (van Dijk et al., 2005).

ia-unrelated causes, which reduced the number of mice in the respective

1514 B. Douradinha et al. / International Journal for Parasitology 37 (2007) 1511–1519

ites 10 days after the last immunisation. Mice immunisedwith three doses of 5,000 or 10,000 RAS or p36p� GASwere fully protected against sporozoite challenge (Table2), whereas all control mice developed parasitemia. Lowdoses of only 1,000 p36p� GAS can also confer full protec-tion. Interestingly, three doses of 1,000 RAS sporozoitesdid not result in full protection (Table 2). All RAS-immun-ised mice developed parasitemia, although with a signifi-cantly prolonged prepatent period (6.2 days versus 3.2days in the control mice).

3.4. Different routes of administration of p36p� GAS

partially protect mice against challenge with sporozoites

We investigated whether p36p� GAS or RAS sporozo-ites confer protection when administered by non-i.v.routes. In two independent experiments C57Bl6 mice wereimmunised by i.v., i.m., s.c. or intradermal (i.d.) injectionof RAS or p36p� GAS (50,000/20,000/20,000, at weeklyintervals). Ten days after the final immunisation, mice werechallenged either by i.v. injection of 10,000 PbGFPconsporozoites or by the bites of infectious mosquitoes thatwere highly infected with PbGFPcon (Table 3). Miceimmunised i.v. with either RAS or p36p� GAS were fullyprotected against sporozoite challenge as expected (Table

Table 2Protection of mice immunised with p36p� GAS or RAS against challenge with

Immunisationa, RAS/p36p� · 103 Prepatent periodb

RAS

50/20/20 –10/10/10 –5/5/5 –1/1/1 6.2None (Control) 3.2

a Groups of mice were immunised i.v. with RAS or p36p� GAS isolated fromshown.

b Mice were challenged with 10,000 PbGFPcon sporozoites, isolated from ditime between challenge and the first appearance of blood-stage parasites as de

Table 3Protection of mice against challenge with PbGFPcon sporozoites after immunisubcutaneously or intradermally

Type of immunisationa Prepatent periodb No. protected/no. challe

RAS p36p� RAS p

Intravenous – – 5/5 (100) 5Intramuscular 4.5 5.3 1/5 (20) 1Subcutaneous 4.8 5.0 0/5 (0) 0Intradermal 4.5 4.8 1/5 (20) 0None (Control) 3.2 0/8 (0)

a Groups of mice were immunised (as described in Section 3) with RAS oindependent experiments are shown.b,c Mice were challenged with 10,000 PbGFPcon isolated from different mosquchallenge and the first appearance of blood-stage parasites as determined bychallenged by i.v. injection of sporozoites.

d Mice were challenged by mosquito bite. Highly infected mosquitoes (50,000to feed on mice during 10–15 min (5 mosquitoes per mouse).

3). None of the mice immunised s.c. with either p36p�

GAS or RAS and challenged by i.v. injection of sporozo-ites were fully protected, although the prepatent periodwas prolonged by almost 2 days. Interestingly, small pro-portions of mice were fully protected when immunisedi.m. (20% for RAS; 13% for p36p� GAS) or i.d. (20% forRAS) (Table 3). The remaining mice (immunised witheither RAS or p36p� GAS) developed parasitemia but witha significantly prolonged prepatent period of between 4.5and 5.3 days (control mice: 3.2 days, P < 0.01; Table 3).These results show that non-i.v. routes of immunisationare less efficient. However, the significantly prolonged pre-patent periods suggest that immune responses that conferpartial protection are elicited. Interestingly, when challengewas performed by delivery of sporozoites via the bites ofinfectious mosquitoes, much higher numbers of mice wereprotected: 83%, 80% and 75% of the mice were fully pro-tected after immunisation by either i.m., s.c., or i.d. deliv-ery of sporozoites, respectively (Table 3).

3.5. Immunisation with p36p� GAS partially protects mice

against challenge with P. yoelii sporozoites

We investigated whether p36p� GAS or RAS couldinduce cross-species protection by challenging P. berghei

PbGFPcon sporozoites

No. protected/no. challenged

p36p� RAS p36p�

– 5/5 8/8– 5/5 5/5– 5/5 8/8– 0/5 10/10

0/8

different mosquito batches. Data from three independent experiments are

fferent mosquito batches. The prepatent period is defined as the period oftermined by FACS analysis.

sation with p36p� GAS or RAS, which were administered intramuscularly,

nged i.v. (%)c No. protected/no. challenged by mosquito bited (%)

36p� p36p�

/5 (100) 5/5 (100)/8 (13) 5/6 (83)/10 (0) 8/10 (80)/9 (0) 6/8 (75)

0/8 (0)

r p36p� GAS isolated from different mosquito batches. Data from two

ito batches. The prepatent period is defined as the period of time betweenFACS analysis and is only determined in experiments were mice were

–80,000 salivary gland PbGFPcon sporozoites per mosquito) were allowed

B. Douradinha et al. / International Journal for Parasitology 37 (2007) 1511–1519 1515

p36p� GAS- or RAS-immunised mice with P. yoelii spor-ozoites. In two independent experiments, BALB/c micewere immunised i.v. with p36p� GAS or RAS using threedifferent immunisation protocols: (i) a single immunisationwith 50,000 sporozoites, (ii) two immunisations with50,000/20,000 sporozoites (at weekly intervals), and (iii)three immunisations with 50,000/20,000/20,000 sporozoites(at weekly intervals). Three weeks after the last immunisa-tion, mice were challenged with 100, 1,000 or 10,000 P. yoe-

lii sporozoites. Relatively low numbers of sporozoites wereused in the challenge since BALB/c mice are known to be2,000-fold more susceptible to P. yoelii than to P. berghei

sporozoites (Khan and Vanderberg, 1991). In these exper-iments, protection was determined both by measuring theprepatent period and by quantifying liver infection usingqRT-PCR. The results show that two immunisations withRAS or p36p� GAS significantly reduced P. yoelii liverdevelopment. Challenge with 100 P. yoelii sporozoites ledto 99.9% and 94.6% inhibition, while challenge with 1,000sporozoites led to 94.8% and 88.6% inhibition in RAS-and p36p� GAS-immunised mice, respectively (Fig. 1). Asingle immunisation with p36p� GAS followed by chal-lenge with 100 and 1,000 P. yoelii sporozoites resulted innon-significant reductions in liver infection of 58.5% and51.1%, respectively (Fig. 1) and did not lead to a significantdelay in appearance of parasitemia (Fig. 2a; Table 4,groups 1 and 4). Partial protection, however, was observedin mice that received two or three p36p� GAS immunisa-tions, as shown by the 1–2 day increase in the prepatentperiod and, additionally, the lower peak parasitemia

Fig. 1. Level of inhibition of liver stage development of Plasmodium yoelii

sporozoites in mice immunised with Plasmodium berghei c-irradiatedsporozoites (RAS) or p36p� GAS. Groups of five mice were eitherimmunised once (I) or twice (I + B) with p36p� GAS or RAS or PBS andchallenged 3 weeks later with 100 (a) or 1,000 (b) P. yoelii sporozoites.Inhibition of development was determined by qRT-PCR quantification of18S ribosomal RNA copies in liver stage parasites 40 h after challengewith P. yoelii parasites. *Significant difference (P < 0.05).

Fig. 2. Maximum (peak) blood-stage parasitemias in Plasmodium berghei

p36p� GAS and RAS immunised BALB/c (a and b) or C57Bl6 mice (c)(five mice per group) which were subsequently challenged with Plasmo-

dium yoelii sporozoites. BALB/c and C57Bl6 mice were immunised witheither p36p� GAS or RAS and challenged 3 weeks after the lastimmunisation with 100, 1,000 or 10,000 P. yoelii sporozoites. I, singleimmunisation; I + B, two immunisations; I + 2B, three immunisations.Parasitemias are expressed as the mean ± SD of individual animals. Datafrom two independent experiments are shown. *Significant difference(P < 0.05) between maximum parasitemias.

compared with control mice (Table 4, groups 1–6, 8;Fig. 2a). Interestingly, immunisation with RAS fully pro-tected 85% (six out of seven) of the BALB/c mice againstchallenge with 100 P. yoelii sporozoites (Fig. 2b and Table4, groups 2 and 3) and 57% (four out of seven) when chal-lenged with 1,000 P. yoelii sporozoites (Table 4, groups 5and 6). The mice that developed parasitemia had a pro-longed prepatent period and a significantly lower peak par-asitemia (P < 0.05) compared with control mice (Fig. 2b;Table 4, group 5). BALB/c mice immunised three timeswith RAS and challenged with 10,000 P. yoelii sporozoiteswere not protected, although they displayed 1 day delay inpatency as well as significantly lower peak parasitemias

Table 4Protection of mice immunised with p36p� GAS or RAS against challenge with different doses of Plasmodium yoelii sporozoites

Mouse strain ImmunisationRAS/p36p� · 103a

Challengedoseb (P. yoelii)

Prepatent period (days)c No. protected (no. challenged)

p36p� RAS Control p36p� RAS Control

1 BALB/c 50 100 5.0 ND 5.2 0 (3) ND 0 (3)2 BALB/c 50/20 100 7.0 – 5.2 1 (3) 3 (3) 0 (3)3 BALB/c 50/20/20 100 5.0 6.0 5.2 0 (5) 3 (4) 0 (5)4 BALB/c 50 1,000 4.0 ND 4.0 0 (3) ND 0 (3)5 BALB/c 50/20 1,000 4.7 5.5 4.0 0 (3) 1 (3) 0 (3)6 BALB/c 50/20/20 1,000 5.2 7.0 4.0 0 (5) 3 (4) 0 (5)7 C57Bl6 50/20/20 1,000 4.8 5.5 4.0 0 (5) 0 (5) 0 (5)8 BALB/c 50/20/20 10,000 4.6 5.0 4.0 0 (5) 0 (4) 0 (5)

ND, not determined.a Groups of mice were immunised (as described in Section 3) with different doses of Plasmodium berghei RAS or p36p� GAS. Data from two

independent experiments are shown.b In all experiments shown control mice were used that were immunised similarly with either RAS or p36p� GAS and challenged with 10,000 PbGFPcon

sporozoites. All control mice did not develop a blood stage parasitemia, confirming that RAS or p36p� GAS used were fully protective in the homologouschallenge studies. The control naıve mice that were challenged either with P. yoelii or P. berghei sporozoites all developed a normal blood stageparasitemia.

c The prepatent period is defined as the period of time between challenge and the first appearance of blood-stage parasites as determined by FACSanalysis.

1516 B. Douradinha et al. / International Journal for Parasitology 37 (2007) 1511–1519

(P < 0.01; Fig. 2b). Such partial protection was alsoobserved in C57Bl6 mice immunised three times with eitherP. berghei p36p� GAS or RAS sporozoites and challengedwith 1,000 P. yoelii sporozoites, with 0.8 and 1.5 days delayin patency, respectively (Table 4, group 7) and significantlylower peak parasitemias (P < 0.05 for both RAS or p36p�

GAS), compared with control mice (Fig. 2c). These resultsshow that immunisation with P. berghei p36p� GAS con-fers partial protection against challenge with P. yoelii

whereas RAS immunisation may fully protect someanimals.

4. Discussion

Despite its efficacy, vaccination with live RAS presents anumber of difficulties including the logistics of productionbut more importantly the reproducibility and safety of theirradiation process. The radiation dose must be strictlycontrolled, since overdose kills the sporozoite, with conse-quent failure to generate protective immunity, while under-irradiated parasites may remain infective (Suhrbier et al.,1990). Additionally, since the production of sufficient num-bers of sporozoites is a limitation (Luke and Hoffman,2003), the minimum dose of attenuated parasites confer-ring protection becomes an important issue. Moreover,current regulations stipulate that all vaccines must beadministered by s.c., i.m. or i.d. routes (Luke and Hoffman,2003). Therefore, a vaccine based on live-attenuated para-sites should also fulfil this requirement. Studies in humanvolunteers need to be performed to validate the feasibilityof these routes of administration of a P. falciparum RAS-based vaccine (Luke and Hoffman, 2003). Immunisationof mice with RAS (Nussenzweig et al., 1967) and GAS(Mueller et al., 2005a,b; van Dijk et al., 2005) of the rodentmalaria parasite P. berghei has been shown to protect

against challenge with wild-type sporozoites. Immunisationwith RAS is well-established and the strength of theimmune response (and hence the level of protection ofmice) is influenced by multiple factors, such as immunisa-tion regime (dose of sporozoites, number of immunisa-tions) (Jaffe et al., 1990; Chatterjee et al., 2001), route ofadministration of the sporozoites (Kramer and Vander-berg, 1975) and the host–rodent used (Chatterjee et al.,2001). Here, we have started to analyse immunisation withGAS in more detail, specifically of GAS which lack themicroneme protein P36p (Ishino et al., 2005; van Dijket al., 2005) and have directly compared the results ofour GAS immunisation procedures with those of RAS.

The results of this study clearly show that a p36p� GASimmunisation strategy confers long-term protection. Fullprotection could be achieved for periods of up to 12–18months (last time point tested for two strains of mice). Inaddition, we found that the protection lasting up to 6months does not require the regular boosting of theimmune response that has previously been thought to becritical for maintenance of long-lived protective immunity(Chatterjee et al., 2001). p36p� GAS conferred full protec-tion against wild-type sporozoite challenge for up to 6months in BALB/c mice. Moreover, 50% of the mice re-challenged 12 months after their last boost immunisationand 6 months after the last challenge were still fully pro-tected. The rest were partially protected, while 100% ofRAS immunised mice were fully protected.

Previously it has been shown that the presence of para-site antigen is needed to maintain the persistence of intra-hepatic memory T-cells and, therefore, immunity againstthe parasite (Berenzon et al., 2003). Whereas persistinghepatocytes containing arrested RAS might serve as an‘antigen-depot’ that maintains activation of the immunesystem, p36p�GAS have been shown to induce higher rates

B. Douradinha et al. / International Journal for Parasitology 37 (2007) 1511–1519 1517

of apoptosis of infected hepatocytes compared with RAS(van Dijk et al., 2005), resulting in the rapid disappearanceof p36p�-infected cells from the liver.

Infected hepatocytes reportedly persist up to 6 monthsafter immunisation with P. berghei RAS (Scheller andAzad, 1995), however similarly low numbers of RAS-and p36p�-infected hepatocytes have been observed 1 weekafter infection of mice (van Dijk et al., 2005).

Perhaps as a consequence of the persistence of infectedhepatocytes, immunisation with RAS induces protectiveimmune responses that seem to last longer than GAS-induced immunity. Whether the specific process of apopto-sis leads to qualitative and/or quantitative differences in theimmune response, as reflected by the more rapid loss ofprotection after immunisation with GAS compared withRAS, requires further investigation. Differences in theimmune response induced by GAS versus RAS may alsoresult from mechanisms of apoptosis-induced inflamma-tion, with consequently more effective antigen presentation(Leiriao et al., 2005). GAS immunisation appears to gener-ate a stronger initial protective immune response comparedwith RAS, as suggested by the relatively low numbers ofp36p� GAS that are able to confer full protection againstsubsequent challenge: three doses of only 1,000 p36p� spor-ozoites provide full protection against challenge with10,000sporozoites in C57Bl6 mice, whereas the same doses ofRAS only confer partial protection. These data confirmthose of previous studies showing that low doses of RASwere unable to provide full protection in C57Bl6 mice (Jaffeet al., 1990). Currently, it is unclear which factors might beresponsible for the difference in protective immunity withlow doses of GAS versus RAS, but the lower number ofGAS required to elicit full protection may be an importantadvantage when production and delivery of live attenuatedorganism-based vaccines are considered. Therefore, futurestudies in mice need to address in more detail aspects ofthe level and longevity of protective immune responsesresulting from immunisation with low doses of GAS andhow these might be enhanced.

In most reports to date, immunisations with RAS andGAS have relied on i.v. injection of sporozoites, but novaccine currently used as a public health measure is admin-istered via this route (Luke and Hoffman, 2003). Studieswith GAS indicate that although the three alternativeimmunisation routes tested (i.m., s.c. or i.d.) induce a lowerlevel of protective immunity, some were able to provide sig-nificant protection against sporozoite challenge. For exam-ple, full protection was achieved in a proportion of micewhen RAS or p36p� GAS were administered i.m., confirm-ing previous studies with RAS (Kramer and Vanderberg,1975) and partial protection, as reflected by a significantlyprolonged prepatent period, was achieved by s.c. delivery.Previous studies with RAS, where different mouse strainsand immunisation regimens were used, were less successful(Kramer and Vanderberg, 1975). In contrast, s.c. adminis-tration of GAS of another mutant P. berghei parasite thatlacks the UIS3 protein confers full protection in all mice

against challenge with sporozoites (Mueller et al., 2005b).A more detailed comparison of the different GAS is thuswarranted in order to clarify the factors related to differentdelivery routes that may enhance or decrease the develop-ment of protective immune responses. As noted above,none of the alternative routes of delivery confer the samelevel of sterile protection as i.v. injection of sporozoites,but the partial protection obtained with both i.m. ands.c. administration of GAS is nevertheless encouraging.Interestingly, much higher levels of protection wereobserved when mice immunised with p36p� GAS usingthese different routes of administration were challengedby infection via mosquito bite. This difference in sensitivityto infection might relate to the fact that mosquito bites arethought to introduce much lower numbers of sporozoitesthan those used in the challenge regimes using purifiedsporozoites (Ponnudurai et al., 1989; Rosenberg et al.,1990; Vaughan et al., 1999; Medica and Sinnis, 2005;Amino et al., 2006). Injection of sporozoites via mosquitobite may also induce additional immune responses as aresult of their migration/survival kinetics (Amino et al.,2006).

The known polymorphisms in proteins of human plas-modial populations suggest that immunisation approachesneed to consider possible differences in induction of effec-tive immune responses against different parasite subpopu-lations (isolates, strains). For P. berghei only a few fieldisolates have been collected and the number of proteinpolymorphisms reflected by sequence comparison of sev-eral genes appears to be low (Carter, 1978; Eichingeret al., 1986). However, P. berghei is closely related toanother rodent parasite, P. yoelii, with which it shares anoverall protein identity of 88.2% (Hall et al., 2005), makingit a suitable candidate for assessments of immunity againstheterologous sporozoite challenge (Nussenzweig et al.,1969, 1972). In immunisation studies with either Plasmo-

dium vivax or P. falciparum RAS performed with humanvolunteers, no cross-species protection was observed (Hoff-man et al., 2002). However, this conclusion relies on resultsfrom a single volunteer who was immunised via the bites ofP. falciparum RAS-infected mosquitoes, shown to be fullyprotected against challenges with P. falciparum sporozoitesbut not against a challenge infection performed later withintact P. vivax sporozoites. Given these conditions it istherefore difficult to portray any concrete conclusion onthe existence of cross-species protection induced by P. fal-ciparum and P. vivax RAS. Our results show that immuni-sation with RAS or, to a lesser extent, p36p� GASinhibited the development of P. yoeliiliver stages as deter-mined both by the level of liver infection, the length ofthe prepatent period and the subsequent peak parasitemias.This work and previous reports (Nussenzweig et al., 1969,1972), clearly show that immunisation with live-attenuatedsporozoites elicit a species-transcending immune responsein malarial rodent models. The mechanisms that elicit suchprotection are presently unclear. One may not exclude thatinnate immune response could exert an effect in heterologous

1518 B. Douradinha et al. / International Journal for Parasitology 37 (2007) 1511–1519

protection, as observed in both infection and homologouslive-attenuated immunisation with malarial P. yoelii spor-ozoites (Gramzinski et al., 2001; Roland et al., 2006). Fur-ther work will unravel the mechanisms of immune responseresponsible for RAS and p36p�mediated cross-species pro-tection and to determine the importance of either innateand adaptive (or both) immune response in suchprotection.

Taken together, our results show that immunisationwith p36p� GAS induces strong protective immuneresponses using different immunisation regimes. Therefore,this specific GAS is a useful tool to unravel the detail ofprotective immune responses against sporozoites/liverstages. P36p is conserved throughout the genus Plasmo-dium and orthologues are present in the human malariaparasites, P. falciparum and P. vivax (Thompson et al.,2001). Further work with rodent-infectious GAS can thusbe expected to provide data highly important to exploringthe potential of GAS as an immunisation approach forhuman malaria.

Acknowledgements

We thank Dr. Joao Pedro Simas for his assistance in thedifferent routes of immunisation used in this work, and Ali-na Costa and Sally Moore, for technical support. Thiswork has been funded by Fundacao para a Ciencia e Tecn-ologia (Portugal), European Science Foundation, TheNetherlands Organisation for Health Research and Devel-opment (ZonMW) and The Wellcome Trust FunctionalGenomics Initiative. B.D. is recipient of a fellowship fromFundacao para a Ciencia e Tecnologia (SFRH/16813/BD/2004), and received support from European Science Foun-dation (COST STSM 857 00743) and from EuropeanMolecular Biology Organization (EMBO Short Term Fel-lowship ASTF 242-2005) for this work. M.M.M. is aEMBO YIP, is a recipient of a EURYI Award from Euro-pean Science Foundation and is a Howard Hughes MedicalInstitute International Scholar.

References

Amino, R., Thiberge, S., Martin, B., Celli, S., Shorte, S., Frischknecht, F.,Menard, R., 2006. Quantitative imaging of Plasmodium transmissionfrom mosquito to mammal. Nat. Med. 12, 220–224.

Berenzon, D., Schwenk, R.J., Letellier, L., Guebre-Xabier, M., Williams,J., Krzych, U., 2003. Protracted protection to Plasmodium berghei

malaria is linked to functionally and phenotypically heterogeneousliver memory CD8+ T cells. J. Immunol. 171, 2024–2034.

Bruna-Romero, O., Gonzalez-Aseguinolaza, G., Hafalla, J.C., Tsuji, M.,Nussenzweig, R.S., 2001. Complete, long-lasting protection againstmalaria of mice primed and boosted with two distinct viral vectorsexpressing the same plasmodial antigen. Proc. Natl. Acad. Sci. USA98, 11491–11496.

Carter, R., 1978. Studies on enzyme variation in the murine malariaparasites Plasmodium berghei, P. yoelii, P. vinckei and P. chabaudi bystarch gel electrophoresis. Parasitology 76, 241–267.

Chatterjee, S., Ngonseau, E., van Overmeir, C., Correwyn, A., Druilhe, P.,Wery, M., 2001. Rodent malaria in the natural host-irradiatedsporozoites of Plasmodium berghei induce liver-stage specific immune

responses in the natural host Grammomys surdaster and protectimmunized Grammomys against P. berghei sporozoite challenge. Afr. J.Med. Med. Sci. 30, 25–33.

Clyde, D.F., Most, H., McCarthy, V.C., Vanderberg, J., 1973. Immuni-zation of man against sporozoite-induced falciparum malaria. Am. J.Med. Sci. 266, 398–403.

Collins, W.E., Contacos, P.G., 1972. Immunization of monkeys againstPlasmodium cynomolgi by X-irradiated sporozoites. Nat. New Biol.236, 176–177.

Eichinger, D.J., Arnot, D.E., Tam, J.P., Nussenzweig, V., Enea, V., 1986.Circumsporozoite protein of Plasmodium berghei: gene cloning andidentification of the immunodominant epitopes. Mol. Cell Biol. 6,3965–3972.

Franke-Fayard, B., Trueman, H., Ramesar, J., Mendoza, J., van derKeur, M., van der Linden, R., Sinden, R.E., Waters, A.P., Janse, C.J.,2004. A Plasmodium berghei reference line that constitutively expressesGFP at a high level throughout the complete life cycle. Mol. Biochem.Parasitol. 137, 23–33.

Gramzinski, R.A., Doolan, D.L., Sedegah, M., Davis, H.L., Krieg, A.M.,Hoffman, S.L., 2001. Interleukin-12- and gamma interferon-dependentprotection against malaria conferred by CpG oligodeoxynucleotide inmice. Infect. Immun. 69, 1643–1649.

Greenberg, J., Nadel, E.M., Coatney, G.R., 1953. The influence of strain,sex and age of mice on infection with Plasmodium berghei. J. Infect.Dis. 93, 96–100.

Hall, N., Karras, M., Raine, J.D., Carlton, J.M., Kooij, T.W.A., Berriman,M., Florens, L., Janssen, C.S., Pain, A., Christophides, G.K., James, K.,Rutherford, K., Harris, B., Harris, D., Churcher, C., Quail, M.A.,Ormond, D., Doggett, J., Trueman, H.E., Mendoza, J., Bidwell, S.L.,Rajandream, M.-A., Carucci, D.J., Yates III, J.R., Kafatos, F.C., Janse,C.J., Barrell, B., Turner, C.M.R., Waters, A.P., Sinden, R.E., 2005. Acomprehensive survey of the Plasmodium life cycle by genomic,transcriptomic, and proteomic analyses. Science 307, 82–86.

Hill, A.V., 2006. Pre-erythrocytic malaria vaccines: towards greaterefficacy. Nat. Rev. Immunol. 6, 21–32.

Hoffman, S.L., Goh, L.M., Luke, T.C., Schneider, I., Le, T.P., Doolan,D.L., Sacci, J., de la Vega, P., Dowler, M., Paul, C., Gordon, D.M.,Stoute, J.A., Church, L.W., Sedegah, M., Heppner, D.G., Ballou,W.R., Richie, T.L., 2002. Protection of humans against malaria byimmunisation with radiation-attenuated Plasmodium falciparum spor-ozoites. J. Infect. Dis. 185, 1155–1164.

Ishino, T., Chinzei, Y., Yuda, M., 2005. Two proteins with 6-cys motifsare required for malarial parasites to commit to infection of thehepatocyte. Mol. Microbiol. 58, 1264–1275.

Jaffe, R.I., Lowell, G.H., Gordon, D.M., 1990. Differences in susceptibil-ity among mouse strains to infection with Plasmodium berghei (ANKAclone) sporozoites and its relationship to protection by gamma-irradiated sporozoites. Am. J. Trop. Med. Hyg. 42, 309–313.

Khan, Z.M., Vanderberg, J., 1991. Role of host cellular response indifferential susceptibility of nonimmunized BALB/c mice to Plasmo-

dium bergheiand Plasmodium yoelii sporozoites. Infect. Immun. 59,2529–2534.

Kramer, L.D., Vanderberg, J., 1975. Intramuscular immunisation of micewith irradiated Plasmodium berghei sporozoites. Enhancement ofprotection with albumin. Am. J. Trop. Med. Hyg. 24, 913–916.

Leiriao, P., Mota, M.M., Rodriguez, A., 2005. Apoptotic Plasmodium-infected hepatocytes provide antigens to liver dendritic cells. J. Infect.Dis. 191, 1576–1581.

Luke, T.C., Hoffman, S.L., 2003. Rationale and plans for developing anon-replicating, metabolically active, radiation-attenuated Plasmo-

dium falciparum sporozoite vaccine. J. Exp. Biol. 206, 3803–3808.Marsh, K., Kinyanjui, S., 2006. Immune effector mechanisms in malaria.

Parasite Immunol. 28, 51–60.Medica, D.L., Sinnis, P., 2005. Quantitative dynamics of Plasmodium

yoelii sporozoite transmission by infected anopheline mosquitoes.Infect. Immun. 73, 4363–4369.

Moorthy, V.S., Good, M.F., Hill, A.V., 2004. Malaria vaccine develop-ments. Lancet 363, 150–156.

B. Douradinha et al. / International Journal for Parasitology 37 (2007) 1511–1519 1519

Mueller, A.K., Camargo, N., Kaiser, K., Andorfer, C., Frevert, U.,Matuschewski, K., Kappe, S.H., 2005a. Plasmodium liver stagedevelopmental arrest by depletion of a protein at the parasite–hostinterface. Proc. Natl. Acad. Sci. USA 102, 3022–3027.

Mueller, A.K., Labaied, M., Kappe, S.H., Matuschewski, K., 2005b.Genetically modified Plasmodium parasites as a protective experimen-tal malaria vaccine. Nature 433, 164–167.

Nussenzweig, R.S., Vanderberg, J., Most, H., Orton, C., 1967. Protectiveimmunity produced by the injection of X-irradiated sporozoites ofPlasmodium berghei. Nature 216, 160–162.

Nussenzweig, R.S., Vanderberg, J.P., Most, H., Orton, C., 1969. Speci-ficity of protective immunity produced by X-irradiated Plasmodium

berghei sporozoites. Nature 222, 488–489.Nussenzweig, R.S., Vanderberg, J., Spitalny, G.L., Rivera, C.I.O., Orton,

C., Most, H., 1972. Sporozoite-induced immunity in mammalianmalaria. A review. Am. J. Trop. Med. Hyg. 21, 722–728.

Orjih, A.U., Cochrane, A.H., Nussenzweig, R.S., 1982. Comparativestudies on the immunogenicity of infective and attenuated sporozoitesof Plasmodium berghei. Trans. R Soc. Trop. Med. Hyg. 76, 57–61.

Ozaki, L.S., Gwadz, R.W., Godson, G.N., 1984. Simple centrifugationmethod for rapid separation of sporozoites from mosquitoes. J.Parasitol. 70, 831–833.

Pierrot, C., Adam, E., Lafitte, S., Godin, C., Dive, D., Capron, M.,Khalife, J., 2003. The age-related resistance of rats to Plasmodium

berghei infection is associated with differential cellular and humoralimmune responses. Int. J. Parasitol. 33, 1067–1078.

Ponnudurai, T., Lensen, A.H., van Gemert, G.J., Bensink, M.P., Bolmer,M., Meuwissen, J.H., 1989. Sporozoite load of mosquitoes infectedwith Plasmodium falciparum. Trans. R Soc. Trop. Med. Hyg. 84,67–70.

Roland, J., Soulard, V., Sellier, C., Drapier, A.M., Di Santo, J.P.,Cazenave, P.A., Pied, S., 2006. NK cell responses to Plasmodium

infection and control of intrahepatic parasite development. J. Immu-nol. 177, 1229–1239.

Rosenberg, R., Wirtz, R.A., Schneider, I., Burge, R., 1990. Infectivity ofPlasmodium berghei sporozoites delivered by intravenous inoculationversus mosquito bite: implications for sporozoite vaccine trials. Infect.Immun. 67, 4285–4289.

Scheller, L.F., Azad, A.F., 1995. Maintenance of protective immunityagainst malaria by persistent hepatic parasites derived from irradiatedsporozoites. Proc. Natl. Acad. Sci. USA 92, 4066–4068.

Schofield, L., Grau, G.E., 2005. Immunological processes in malariapathogenesis. Nat. Rev. Immunol. 5, 722–735.

Schofield, L., Mueller, I., 2006. Clinical immunity to malaria. Curr. Mol.Med. 6, 205–221.

Silvie, O., Semblat, J.P., Franetich, J.F., Hannoun, L., Eling, W., Mazier,D., 2002. Effects of irradiation on Plasmodium falciparum sporozoitehepatic development: implications for the design of pre-erythrocyticmalaria vaccines. Parasite Immunol. 24, 221–223.

Singer, I., Hadfield, R., Lakonen, M., 1955. The influence of age on theintensity of infection with Plasmodium berghei in the rat. J. Infect. Dis.97, 15–21.

Suhrbier, A., Winger, L.A., Castellano, E., Sinden, R.E., 1990. Survivaland antigenic profile of irradiated malarial sporozoites in infected livercells. Infect. Immun. 58, 2834–2839.

Thompson, J., Janse, C.J., Waters, A.P., 2001. Comparative genomics inPlasmodium: a tool for the identification of genes and functionalanalysis. Mol. Biochem. Parasitol. 118, 147–154.

van Dijk, M.R., Douradinha, B., Franke-Fayard, B., Heussler, V., vanDooren, M.W., van Schaijk, B., van Gemert, G.J., Sauerwein, R.W.,Mota, M.M., Waters, A.P., Janse, C.J., 2005. Genetically attenuated,P36p-deficient malarial sporozoites induce protective immunity andapoptosis of infected liver cells. Proc. Natl. Acad. Sci. USA 102,12194–12199.

Vaughan, J.A., Scheller, L.F., Wirtz, R.A., Azad, A.F., 1999. Infectivityof Plasmodium berghei sporozoites delivered by intravenous inocula-tion versus mosquito bite: implications for sporozoite vaccine trials.Infect. Immun. 67, 4285–4289.