A Next-generation Genetically Attenuated Plasmodium falciparum Parasite Created by Triple Gene Deletion

Transcription

1 vccine technology originl rticle A Next-genertion Geneticlly Attenuted Plsmodium flciprum Prsite Creted by Triple Gene Deletion Sebstin A Mikoljczk 1, Viswnthn Lkshmnn 1, Mtthew Fishbugher 1, Nelly Cmrgo 1, Anke Hrup 1, Alexis Kushnsky 1, Alyse N Douglss 1, Michel Bldwin 1, Julie Heler 2, Mtthew O Neill 2, Thun Phuong 2, Aln Cowmn 2 nd Stefn HI Kppe 1,3 1 Settle Biomedicl Reserch Institute (Settle BioMed), Settle, Wshington, USA; 2 The Wlter nd Eliz Hll Institute of Medicl Reserch, Victori, Austrli; 3 Deprtment of Globl Helth, University of Wshington, Settle, Wshington, USA. Immuniztion with live-ttenuted Plsmodium sporozoites completely protects ginst mlri infection. Genetic engineering offers verstile pltform to crete live-ttenuted sporozoite vccine cndidtes. We previously generted geneticlly ttenuted prsite (GAP) by deleting the P52 nd P36 genes in the NF54 wild-type () strin of Plsmodium flciprum (Pf p52 /p36 GAP). Preclinicl ssessment of p52 /p36 GAP in humnized mouse model indicted n erly nd severe liver stge growth defect. However, humn exposure to >2 Pf p52 /p36 GAP-infected mosquito bites in sfety tril resulted in peripherl prsitemi in one of six volunteers, reveling tht this GAP ws incompletely ttenuted. We hve now creted triple gene deleted GAP by dditionlly removing the SAP1 gene (Pf p52 /p36 /sp1 GAP) nd employed flippse (FLP)/flippse recognition trget (FRT) recombintion for drug selectble mrker cssette removl. This next-genertion GAP ws indistinguishble from prsites in blood stge nd mosquito stge development. Using n improved humnized mouse model trnsplnted with humn heptocytes nd humn red blood cells, we show tht despite high-dose sporozoite chllenge, Pf p52 /p36 /sp1 GAP did not trnsition to blood stge infection nd ppered to be completely ttenuted. Thus, clinicl testing of Pf p52 / p36 /sp1 GAP ssessing sfety, immunogenicity, nd efficcy ginst sporozoite chllenge is wrrnted. Received 29 Jnury 214; ccepted 29 April 214; dvnce online publiction 17 June 214. doi:1.138/mt INTRODUCTION Plsmodium prsites tht cuse mlri present formidble thret to humn helth, primrily in resource-poor regions of the world. 1 The tremendous morbidity nd mortlity inflicted by mlri infection could be drmticlly diminished by completely protective mlri vccine. 2 Fortuntely, immuniztions with live-ttenuted sporozoites in niml models nd humns hve consistently demonstrted tht mlri vccine conferring complete, protrcted protection is possible. 3 9 Recent work, involving immuniztions with infectious Plsmodium flciprum sporozoites delivered by reltively smll numbers of mosquito bites under concurrent chloroquine prophylxis (chemoprophylxis with sporozoites), showed robust, longlsting protection ginst infectious sporozoite chllenge. 1 Furthermore, recent clinicl studies with irrdition-ttenuted P. flciprum sporozoites dministered to humns intrvenously showed complete protective efficcy ginst infectious sporozoite chllenge t the highest-dose immuniztion regimen. 11 This work built on previous studies showing tht irrdition-ttenuted P. flciprum sporozoites dministered to humns by the bite of >1, infected mosquitoes conferred robust nd ner-complete protection ginst infectious sporozoite chllenge. 12 However, both chemoprophylxis with sporozoites nd irrdition-ttenuted sporozoites do not llow for ny intrinsic control over the design of the whole prsite immunogen. In contrst, engineered ttenution of prsites (geneticlly ttenuted prsite (GAP)) offers pltform for controlled nd consistent design of whole prsite immunogen. Previous work with the rodent mlri prsites P. berghei nd P. yoelii demonstrted the utility of genetic ttenution by deletion of the pre-erythrocytic stge-expressed genes clled upregulted in infectious sporozoites (UIS)3 nd UIS4, both of which resulted in erly liver stge developmentl rrest Deletion of pre-erythrocytic stge-expressed genes P52 nd P36 similrly resulted in n erly liver stge developmentl rrest. 16,17 Immuniztion of mice with uis3, uis4, or p52 prsites induced complete long-lsting protection ginst infectious sporozoite chllenge, demonstrting tht rodent mlri GAPs re highly effective immunogens ,18 Subsequently, double gene deletion p52 /p36 GAP ws generted in P. yoelii. 19 Infection with Py p52 /p36 sporozoites did not cuse blood stge infection in Blb/cJ mice, nd immuniztion with Py p52 / p36 sporozoites conferred complete protection ginst intrvenous infectious sporozoite chllenge or infectious mosquito bite chllenge. 19 Deletion of P52 nd P36 prevents the prsite from estblishing or mintining prsitophorous vcuole The first two uthors contributed eqully to this work. Correspondence: Stefn HI Kppe, Settle Biomedicl Reserch Institute (Settle BioMed), 37 Westlke Avenue North, Suite 5, Settle WA 9819, USA. E-mil: stefn.kppe@settlebiomed.org Moleculr Therpy vol. 22 no. 9, sep

2 Geneticlly Attenuted Plsmodium flciprum Tble 1 Infections of Plsmodium yoelii p52 /p36 nd P. yoelii sp1 in the Blb/cByJ mouse model Prsite genotype Number of injected sporozoites Number of mice injected/number with blood stge ptency Dy of blood stge ptency P. yoelii wild type /5 3 P. yoelii p52 /p /4 4.7 P. yoelii sp / Mice injected with p52 /p36 nd sp1 sporozoites were followed for 14 dys postinjection. Sporozoites were isolted from mosquito slivry glnds t dy 14 or dy 15 postinfectious blood mel nd were intrvenously injected into the til vein of Blb/cByJ mice. The time to blood stge ptency (defined s >1 infected red bllod cell (RBC)/1, RBCs) ws monitored microscopiclly by Giems-stined thin blood smers strting on dy 3 postinjection. membrne, which cuses the prsite s ttenuted phenotype. To ssess orthologous genetic ttenution in P. flciprum, we previously generted double gene deletion GAP lcking P52 nd P36 (Pf p52 /p36 ) in the NF54 strin. 2 Anlysis of the knockout (2 KO) prsites in vitro nd in humnized mouse model showed tht the Pf p52 /p36 prsites suffered liver stge growth defect erly following heptocyte infection. Subsequent clinicl ssessment showed tht Pf p52 /p36 ws not completely ttenuted in humn infection, nd resulted in peripherl blood stge prsitemi in one of six volunteers exposed to >2 Pf p52 /p36 -infected mosquito bites. 21 Therefore, the next step ws to introduce third gene deletion in these prsites tht, when combined with the Pf p52 /p36 bckground, would potentilly yield complete ttenution. In n erlier study, we identified the sporozoite sprgine-rich protein 1 (SAP1) to be essentil for liver stge development of P. yoelii. 22 The encoded protein loclized to the cytoplsm of slivry glnd sporozoites nd deletion of SAP1 led to reduction in bundnce of numerous UIS trnscripts cused by incresed RNA degrdtion. 22,23 P. yoelii sp1 sporozoites exhibited complete rrest in erly liver stge development nd the bsence of blood stge prsitemi in mice injected with up to sp1 sporozoites. 22 Similr to P. yoelii, sp1 P. berghei lso showed complete ttenution. 24 Thus, we trgeted P. flciprum SAP1 for deletion to generte triple gene KO () GAP. Here, we report the successful genertion of Pf p52 /p36 /sp1 GAP nd its chrcteriztion in preclinicl ssys. RESULTS Complete ttenution of P. yoelii sp1 in the Blb/ cbyj mouse strin Our initil studies of the P. yoelii p52 /p36 GAP indicted complete ttenution becuse injection of sporozoites in Blb/cJ mice did not led to blood stge prsitemi (brekthrough). 19 Similrly, no blood stge brekthrough infection ws observed with the orthologous KO prsite in P. berghei (Pb p52 /p36 ) in Blb/cJ mice. 25 However, occsionl brekthrough infection did occur in C57BL/6 mice injected with Pb p52 /p36 sporozoites, which re more susceptible to P. berghei infection. 25 Thus, we tested if Py p52 /p36 could show brekthrough blood stge infection in more susceptible mouse strin. We recently observed tht Blb/cByJ mice, congenic strin of Blb/cJ mice, crried significntly higher liver stge burden when compred to Blb/cJ mice (unpublished dt). In light of the higher susceptibility of Blb/cByJ mice to P. yoelii, we infected Blb/cByJ mice with high-dose ( ) of P. yoelli p52 /p36 sporozoites nd indeed observed brekthrough blood stge prsitemi in frction of these mice (Tble 1). To evlute the brekthrough cpcity of Py sp1 prsites, we then tested the sme dose of P. yoelii sp1 sporozoites in Blb/cByJ mice nd observed no brekthrough blood stge prsitemi in ny of the chllenged mice (Tble 1). This demonstrted tht Py sp1 prsites re completely ttenuted. Thus, we decided to pursue SAP1 (PF3D7_1147) deletion to chieve complete P. flciprum liver stge ttenution, resoning tht combining deletion cusing full rrest with two deletions cusing severe but incomplete rrest in murine models would generte prsite highly unlikely to cuse brekthrough growth in humns. Genertion of the P. flciprum p52 /p36 /sp1 triple gene KO prsite Our recent phse 1 clinicl study with Pf p52 /p36 GAP showed tht this 2KO prsite ws severely but incompletely ttenuted s one of six volunteers exposed to >2 Pf p52 / p36 -infected mosquito bites developed peripherl blood prsitemi. 21 To further enhnce the ttenuted phenotype, we dditionlly trgeted SAP1 in the Pf p52 /p36 genetic bckground. A clone of Pf p52 /p36 in which the positive selectble mrker HsDHFR hd been removed 26 llowed for the introduction of sequentil gene deletion. SAP1 ws deleted in this clone using the positive-negtive selection strtegy tht ws previously used to delete P52 nd P36 (Figure 1). 2 Following cloning of recombinnt prsites by limiting dilution, Southern blotting of genomic DNA (Southern) nd genotyping by polymerse chin rection (PCR) were performed to confirm the successful deletion of SAP1 in this Pf p52 /p36 /sp1 s well s the bsence of ny residul Pf p52 /p36 prsites (Figure 1,b). Pf p52 /p36 /sp1 prsites show norml blood stge growth nd development in mosquitoes Anlysis of the clonl popultion of the showed no observble defect during sexul blood stge repliction (Figure 1c), in its bility to produce gmetocytes (Figure 1d) or mle gmete exflgelltion (Figure 1e) when compred to wild type (). Evlution of midgut infection in Anopheles stephensi mosquitoes fed with gmetocyte cultures showed no significnt differences between nd prsites in oocyst prevlence (Figure 2) nd oocyst numbers per infected mosquito midgut (Figure 2b). Importntly, the invsion of mosquito slivry glnds ppered unchnged in, s similr numbers of slivry glnd sporozoites were isolted when compred to (Figure 2c). Additionlly, sporozoites showed robust gliding motility on solid substrte (Figure 2d) vol. 22 no. 9 sep. 214

3 Geneticlly Attenuted Plsmodium flciprum p535 E A 3.4 kb 4.3 kb 5.2 kb A p536 E b SAP1 3 probe AflIII SAP1 5 probe EcoRV SAP1 p535 p536 5 probe 3 probe SAP1 HsDHFR ScCDUP kb p535.9 kb 7.5 kb p kb 3.2 kb SAP1 KO locus w/hsdhfr H 5 probe E A E A A C 3 probe HsDHFR HsDHFR probe c d e Prsitemi (%) Time (dys) Gmetocytemi (%) Figure 1 Strtegy for trgeted gene deletion of Pf SAP1, chrcteriztion, nd phenotypic nlysis of the Pf p52 /p36 /sp1. () Schemtic of strtegy for deleting the SAP1 gene in Pf p52 - /p36 - KO clone lcking the HsDHFR mrker. 26 Enzymes nd probes for Southern blotting nd primers for polymerse chin rection (PCR) re shown. Sizes of genomic DNA frgments nd PCR products re indicted in kilobses. (b) Southern blotting (left nd middle pnels) nd PCR (right pnel) to show deletion of SAP1 in Pf p52 /p36 KO clone (Pf p52 /p36 /sp1 ). AflIII-digested genomic DNA ws hybridized with 3 probe yielding 4.3 nd 3.2 kb bnd for wild type () nd DNA, respectively. Hybridiztion of EcoRV-digested genomic DNA with 5 probe detected 5.2 nd.9 kb bnd for nd DNA, respectively. PCR on genomic DNA using primers p535 nd p536 yielded 3.4 nd 4.1 kb bnd for nd DNA, respectively. (c) Comprison of sexul blood stge growth rtes, s mesured by increse in prsitemi over time, between nd. Cultures were initited t.5% prsitemi nd nlyzed dily by Giems-stined thin blood smers until dy 5. Growth ssys were performed in triplicte for ech line nd prsitemi plotted s men ± SEM. Mnn Whitney U-test ws used for sttisticl nlysis. (d) Comprison of % gmetocytemi between nd cultures t the time of feeding in vitro gmetocyte cultures to mosquitoes (14 16 dys postinitition of in vitro gmetocyte cultures). Gmetocytemi ws determined three independent times in duplicte cultures for ech line nd plotted s men ± SEM. Mnn Whitney U-test ws used for sttisticl nlysis. (e) Comprison of number of mle gmete exflgelltion events between nd gmetocyte cultures t the time of feeding to mosquitoes. Exflgelltion (plotted s men ± SEM) ws nlyzed in severl microscopic fields four independent times in duplicte nd triplicte cultures for nd, respectively. Mnn Whitney U-test ws used for sttisticl nlysis. Exflgelltion events/field Pf p52 /p36 /sp1 sporozoites show norml cell trversl nd host cell infection We next investigted if the gene deletions in the clone would ffect the bility of the prsite to trverse or invde heptocytes, the two steps importnt for Plsmodium prsites to initite liver infection. No significnt differences were observed between nd sporozoites in their bility to trverse heptocytes in vitro (Figure 2e). Sporozoites of were lso ble to infect heptocytes in vitro t levels comprble to (Figure 2f). Pf p52 /p36 /sp1 prsites fil to complete liver stge development in FRG-HuHep mice To evlute whether prsites cn complete liver stge development in vivo, FRG-HuHep (FRG) mice were injected intrvenously with one million or sporozoites. Seven dys fter sporozoite infection, mice were injected with humn red blood cells nd subsequently scrificed to collect the livers nd peripherl blood. The blood ws used for in vitro culture to detect the occurrence of blood stge infection. 27 Blood collected from mice injected with Moleculr Therpy vol. 22 no. 9 sep

4 Geneticlly Attenuted Plsmodium flciprum 9 b 21 c 51, Oocyst prevlence (%) 6 3 Oocyst/midgut 14 7 Sporozoites/mosquito 34, 17, d e f Totl wounding (%dextrn(+) cells) Totl infection (%CS(+) cells) Figure 2 Mosquito stge development, nd in vitro sporozoite host cell trversl nd invsion ssys. () Prevlence of oocysts in mosquito midguts infected with nd prsites. Oocyst prevlence (plotted s men ± SEM) ws clculted by dividing the number of dissected midguts contining oocycts by the totl number of midguts on dy 7 postfeeding of mosquitoes with in vitro gmetocyte cultures. Prevlence ws determined in severl mosquitoes four independent times in duplicte nd qudruplicte for nd, respectively. Mnn Whitney U-test ws used for sttisticl nlysis. (b) Comprison of verge number of oocysts per mosquito midgut (plotted s men ± SEM) between nd on dy 7 postfeeding of mosquitoes. Oocyts numbers were determined in severl mosquitoes four independent times in duplicte nd qudruplicte for nd, respectively. Mnn Whitney U-test ws used for sttisticl nlysis. (c) Comprison of verge number of sporozoites per mosquito (plotted s men ± SEM) between nd on dys postfeeding of mosquitoes. Sporozoite numbers were determined three independent times in t lest duplicte for ech line. Mnn Whitney U-test ws used for sttisticl nlysis. (d) Stining of CSP trils using Alex 488-conjugted nti-pfcs 2A1 ntibody in motility ssys of slivry glnd sporozoites from nd. Sporozoites were collected dys postfeeding of mosquitoes. (e) Averge totl of HC-4 cells in trversl ssys with nd slivry glnd sporozoites (plotted s men ± SEM) s mesured by the frction of totl HC-4 cells in the smple tht hd tken up FITC-dextrn. Totl dextrn positive cells were determined four independent times in duplicte for ech line. Mnn Whitney U-test ws used for sttisticl nlysis. (f) Averge totl infection of HC-4 cells by nd slivry glnd sporozoites (plotted s men ± SEM) s mesured by the frction of totl HC-4 cells in the smple tht were positive for intrcellulr prsites s mesured by CS stining. Totl CS-positive cells were determined four independent times in duplicte for ech line. Mnn Whitney U-test ws used for sttisticl nlysis CS, circumsporozoite; FITC, fluorescein isothiocynte. sporozoites consistently produced ptent prsitemi in culture. In contrst, none of the in vitro cultures of blood collected from the sporozoite-injected mice produced positive blood stge prsite culture s evluted by microscopic exmintion of blood smers for period of 3 weeks postcollection (Tble 2). Livers from these mice were lso removed on dy 7 nd nlyzed by immunofluorescence ssy for presence of liver stge prsites. As expected, mture lte stge prsites were detected in livers from sporozoiteinjected mice (Supplementry Figure S2) t n verge density of one prsite per 74 mm 2 re of liver section (Supplementry Tble S1). Conversely, no liver stges of ny size were detected on dy 7 in mice injected with sporozoites (Supplementry Tble S1). A next-genertion GAP vccine cndidte for clinicl studies A GAP tht could enter clinicl development would need to be devoid of ny extrneous DNA nd not exhibit ny drug resistnce. Thus, we removed the HsDHFR selectble mrker from the sp1 locus using the FLP/FRT system 26 in prsites. Two clonl lines tht were positive for Pf p52 /p36 /sp1 triple gene deletion were lso devoid of ny drug selectble mrker cssettes (clones (-H) c1 nd c2) by Southern nd PCR nlyses (Supplementry Figure S1 nd Supplementry Tble S2). These clones were nlyzed for retention of vitl chrcteristics needed to produce sporozoites for Tble 2 Anlysis of blood stge ptency in FRG mice injected with wild type or Pf p52 /p36 /sp1 slivry glnd sporozoites Number of mice infected / Prsite line number with blood stge ptency b Wild type 3/3 3/ 3/3 (-H) c2 3/ Mice were injected with sporozoites. b Ptency ws checked 3 3 dys fter initition of in vitro blood stge culture. experimentl vccintion. As with the prentl prsite, no defects were observed in blood stge growth (Figure 3), gmetocyte production (dt not shown), sporozoite production (Figure 3b), nd gliding on solid substrtes (Figure 3c). Additionlly, (-H) c1 nd c2 were susceptible to the drugs WR9921 nd blsticidin, further demonstrting tht the cssettes contining selectble mrkers conferring resistnce to these drugs were no longer retined (Supplementry Tble S3). Three FRG mice were injected with sporozoites of one of the (-H) clones to determine if the prsites showed the sme degree of liver stge ttenution s observed for the mrker-contining. As expected, no liver stge prsites were detected by immunofluorescence ssys t vol. 22 no. 9 sep. 214

5 Geneticlly Attenuted Plsmodium flciprum Prsitemi (%) ( H) c1 ( H) c2 b Sporozoites/mosquito 9, 6, 3,. 2 4 Time (dys) 6 ( H) c1 ( H) c2 c ( H) c1 ( H) c2 Figure 3 Phenotypic nlysis of Pf p52 /p36 /sp1 (-H) clones. () Comprison of sexul blood stge growth rtes between wild type () nd (-H) clones. Cultures were initited t.5% prsitemi nd nlyzed dily until dy 5. Growth ssys were performed in triplicte for ech line nd prsitemi plotted s men ± SEM. Mnn Whitney U-test ws used for sttisticl nlysis. (b) Comprison of verge number of sporozoites (plotted s men ± SEM) per mosquito between nd (-H) clones dys postfeeding of mosquitoes. Sporozoite numbers were determined three independent times in t lest duplicte for ech line. Mnn Whitney U-test ws used for sttisticl nlysis. (c) Stining of CS trils using Alex Fluor 488-conjugted nti-pfcsp 2A1 ntibody in motility ssys of slivry glnd sporozoites of nd (-H) clones. dy 7 post-(-h) sporozoite injection, nd in vitro culturing of blood collected from the (-H)-infected FRG mice did not show occurrence of blood stge infection during 3 weeks of culture (Tble 2). Control mice injected with sporozoites hd detectble liver stge infection nd in vitro blood cultures were consistently prsitemic shortly fter blood trnsfer from infected FRG mice (Tble 2). Finlly, we used reverse trnscriptse polymerse chin rection (RT-PCR) nlysis to test for the presence of P. flciprum 18S A-type trnscripts in dy 7 liver smples from FRG mice infected with either or (-H) slivry glnd sporozoites. While we detected strong signl for 18S in FRG mice infected with sporozoites t dy 7, no signl for 18S ws detected in liver smples from three independent mice tht hd been injected with (-H) sporozoites (Supplementry Figure S3). DISCUSSION Despite the enormous importnce of mlri, the gol of developing P. flciprum vccine tht demonstrtes high efficcy nd confers long-lsting protection remins unrelized. 28,29 Recent dt however renewed optimism tht highly protective vccine might be ttinble when bsed on live-ttenuted sporozoites. Seder et l. 11 showed tht intrvenous immuniztion with repeted high doses of irrdited, cryopreserved P. flciprum sporozoites protected six out of six volunteers ginst infectious P. flciprum chllenge. This is in greement with historicl dt showing tht intrvenous immuniztions with irrdition-ttenuted sporozoites in niml models nd by mosquito bite delivery in humn volunteers induce sterile protection ginst subsequent sporozoite chllenges. The efficcy of these whole live-ttenuted sporozoite immuniztions hve so fr been unmtched by ny subunit mlri vccines in development, including the most dvnced CSP-bsed RTS,S vccine cndidte currently being tested in phse 3 clinicl studies, which showed 3 5% protective efficcy nd reltively short durtion of protection. 3 Trgeted gene deletion tht llows for producing intrinsiclly nd uniformly ttenuted Plsmodium sporozoites with potentil of incresed vccintion potency, is design-bsed lterntive strtegy to irrdition-bsed prsite ttenution. 31 In our previous study, deletion of two pre-erythrocytic stge-expressed genes (P52 nd P36) in P. flciprum negtively ffected the bility of sporozoites to crete functionl heptocyte infection nd initite liver stge development, resulting in n ttenuted phenotype. 2 Phenotypic nlysis of the P52 nd P36 gene deletions in rodent mlri prsites implicted the lck of prsitophorous vcuole membrne surrounding the prsites erly in heptocyte infection s the cuse for the profound developmentl defect in the p52 /p36 prsites. 19,2 Despite this defect, however, the double gene deletion did not completely ttenute Pf p52 /p36 infections. This cme to light in first-in-humn proofof-concept sfety study in which one of six volunteers developed peripherl blood prsitemi fter exposure to >2 Pf p52 /p36 GAP-infected mosquito bites. 21 Interestingly, it ws recently shown tht P. berghei rodent mlri p52 /p36 prsites could, in rre instnces, develop within the heptocyte without prsitophorous vcuole membrne formtion, 32 providing potentil explntion for the occurrence of brekthrough blood stge infections. We hve here substntited this potentil for brekthrough in the P. yoelii Moleculr Therpy vol. 22 no. 9 sep

6 Geneticlly Attenuted Plsmodium flciprum model by demonstrting the infrequent occurrence of blood stge prsitemi in highly susceptible Blb/cByj mice chllenged with Py p52 /p36 sporozoites. Thus, this model will be n importnt experimentl ddition to the toolbox tht evlutes GAP phenotypes. To optimize ttenution nd prevent blood stge brekthrough of those rre prsites tht could develop without P52 nd P36 expression, we introduced n dditionl gene deletion into Pf p52 /p36 prsites. The deletion ws selected to complete ttenution bsed on dt from rodent mlri prsite models, which showed tht Py nd Pb sp1 prsites suffered complete ttenution of erly liver stge growth nd did not show brekthrough blood stge prsitemi. 22,24 We corroborted this here with the highly sensitive P. yoelii-blb/cbyj model. SAP1 ws lso selected for deletion s the puttive function of the encoded protein significntly differs from the functions of P52 nd P36. While SAP1 is cytoplsmic protein involved in regulting RNA stbility nd s such impcts sporozoite gene expression, 22 P52 nd P36 re secreted proteins both involved in the formtion of the prsitophorous vcuole membrne. 19 Deletions of genes tht re involved in independent biologicl processes should improve the robustness of complete ttenution nd further reduce the possibility of compenstory chnges in the prsite tht might led to loss of ttenution. We demonstrted here tht the Pf p52 /p36 /sp1 triple gene deletion hd no significnt effect on gmetocytogenesis, mosquito infectivity, or sporozoite production. Also, despite undergoing extensive sequentil genetic mnipultion nd drug selection in vitro, Pf p52 /p36 /sp1 sporozoites showed neither reduced vibility nor ltered chrcteristics of initil infection, s mesured by the bility of the sporozoites to glide on solid substrte, trverse nd infect heptocytes. To directly ssess ttenution of the Pf p52 /p36 /sp1 prsite, we employed the robust humnized FRG mouse model hrboring humn heptocytes nd humn red blood cells tht llow for complete development of P. flciprum liver stges nd supports liver stge-to-blood stge trnsition. 27 Unlike prsites, prsites were undetectble in the livers of infected FRG mice t dy 7 fter sporozoite infection. Strikingly, no liver-to-blood trnsition of infection ws observed for the, while infections relibly showed this trnsition. Long-term in vitro culture of the isolted blood lso did not result in detectble prsitemi for. This further suggests tht the prsites re fully ttenuted nd cnnot undergo liver stge development. The removl of the drug resistnce mrker ws chieved using the flippse (FLP)/flippse recognition trget (FRT) system, which llowed for complete excision of ll exogenous DNA from the genome. As result, the (-H) clones re devoid of ny drug resistnce mrkers, which is currently requirement for the use of geneticlly engineered gents in dvnced clinicl testing. It is lso importnt to note tht severl rounds of genetic mnipultion nd prolonged in vitro culturing did not ffect the fitness nd vibility of the (-H) clones. This further emphsizes tht the development of P. flciprum geneticlly ttenuted sporozoites for vccintion is fesible. With the pursuit of triple gene deletion Pf GAP, we hve focused on chieving robust sporozoite production, vibility, nd complete ttenution, which is prerequisite for the use of ttenuted sporozoites in humn vccintion. A recent clinicl study with Pf p52 /p36 GAP showed tht they induce substntil immune responses including functionl ntibody responses tht cn effectively block sporozoite infection in vitro. 21,33 These dt, together with extensive evidence tht sp1 nd p52 /p36 rodent GAPs engender sterile protection ginst sporozoite chllenge in mice, give resons to predict tht the Pf p52 /p36 /sp1 triple gene deletion GAP will induce protective immune responses in humns. 19,22,24 However, erly liver stge rresting GAPs might not yet constitute the optiml live-ttenuted immunogen. It ws previously shown tht lte liver stge rresting P. yoelii GAPs, creted by gene deletions in the ftty cid biosynthesis pthwy (FASII), produce superior immune responses nd protection in mice. 34,35 However, recent nlysis of FASII gene deletions in P. flciprum showed n unexpected detrimentl effect on sporozoite formtion in oocysts, thus currently precluding the production of FASII KO GAPs for humn immuniztion. 36 In conclusion, we hve developed next-genertion triple gene deletion GAP strin of P. flciprum. We used stte-of-thert preclinicl tools to evlute its degree of ttenution nd found tht produced vible, infectious sporozoites tht rrested erly, could not complete liver stge development nd could not trnsition to blood stge infection. A proof-of concept clinicl study with the GAP ssessing sfety, induction of cellulr nd humorl immune responses s well s preliminry efficcy ginst infectious sporozoite chllenge is thus wrrnted. MATERIALS AND METHODS Ethics sttement. All niml studies were pproved by the Institutionl Review Bord t Settle BioMed. In vitro culturing of prsite lines. The P. flciprum NF54 strin nd KO lines were propgted in vitro in O + humn blood (Interstte Blood Bnk, Memphis, TN) in custom-mde Roswell Prk Memoril Institute medium contining hypoxnthine, sodium bicrbonte, nd 4-(2-hydroxyethyl)-1-piperzineethnesulfonic cid (Invitrogen, Life Technologies, Grnd Islnd, NY) supplemented with.4% Albumx (Invitrogen) nd gentmycin (Invitrogen) ccording to stndrd procedures for in vitro P. flciprum culturing. Blood stge growth ssy. In vitro cultures of prsites were set up in triplicte in 12-well pltes t.5% strting prsitemi nd 4% hemtocrit (HCT). Medi ws replced nd prsitemi ws determined dily for 5 dys by Giems (Sigm-Aldrich, St Louis, MO)-stined thin smers to determine rte of growth. Design of gene-trgeting constructs nd genertion of KO prsite lines. Oligonucleotide primers used to generte 5 nd 3 homologous recombintion regions (5 nd 3 flnks, respectively) for simultneous deletion of P52 nd P36 genes were s follows: P52 5 flnk: 5 CCGCGGGGATC TCTATAAATGCATGAGG (sense) nd 5 ; ACTAGTGAAGTTCCTATAC TTTCTAGAGAATAGGAACTTCCTGGGTGAGTTTTTGCCGTA GTACTAAAAGCATCATTC (ntisense); P36 3 flnk: 5 CCTAGGGAA- GTTCCTATTCTCTAGAAAGTATAGGAACTTCGGGAATTTACAT- GCCATTCTATG (sense) nd 5 GGCGCCCCTATACCCTTCCCTT- GTG (ntisense). For P52/P36 trgeting plsmid, the 5 nd 3 flnks were cloned into ScII-SpeI nd AvrII-SfoI, respectively in the pcc1 plsmid. 37 Oligonucleotide primers used to generte 5 nd 3 flnks for deleting SAP1 gene were s follows: 5 flnk: 5 CCGCGGTGAAGAAAAGGG AAACCAAGACATGTG (SAP1 strt codon mutted; itlicized; sense) nd 5 ACTAGTATAACTTCGTATAGCATACATTATACGAAGTT ATGGTGTATTATAACTTTGTGGTGTATTATAAC (ntisense); 3 flnk: 5 GAATTCATAACTTCGTATAATGTATGCTATACGAAGTTATCAGAA vol. 22 no. 9 sep. 214

7 Geneticlly Attenuted Plsmodium flciprum TCAAAATTATATAACCAACC (sense) nd 5 CCTAGGCGTTGTTAAGA TGTGGGTCTATATACG (ntisense). 5 nd 3 flnks for SAP1 trgeting plsmid were cloned into ScII-SpeI nd EcoRI-AvrII, respectively in pcc1. prsites were trnsfected (see below) to generte the p52 / p36 double KO prsites. 2,26 Knockout prsites were cloned by limiting dilution. HsDHFR mrker ws removed by trnsfecting clone of p52 / p36 with p-tet-bsd-flp plsmid. 26 SAP1 ws deleted in clone of p52 / p36 lcking the HsDHFR mrker yielding the p52 /p36 /sp1 triple KO. This line ws further cloned by limiting dilution. One clone () ws selected nd trnsfected with the ptet-bsd-flp plsmid generting popultion lcking the HsDHFR mrker in the SAP1 locus. This popultion ws cloned gin by limiting dilution nd two clones ( (-H) c1 nd c2) were picked. Trnsfection of P. flciprum with KO plsmid constructs. Plsmid DNA ws extrcted by mxi prep (Qigen, Vlenci, CA). P. flciprum NF54 () prsites were sorbitol synchronized 38 nd trnsfected with 3 4 μg plsmid by electroportion t.31 kv nd 95 μf using BioRd Gene Pulser (BioRd, Hercules, CA). 39 To select for trnsfectnts, cultures were plced under either WR9921 (WR; Jcobus Phrmceuticls, Princeton, NJ) or Blsticidin (Invivogen, Sn Diego, CA) 48 hours fter trnsfection depending on whether HsDHFR or AtBSD used s selectble mrker, respectively. 26 WR nd Blsticidin were used t 2.5 nmol/l nd 2.5 μg/ml, respectively. Negtive selection for KO prsites. Knockout prsites were generted by positive-negtive drug selection. 37 Briefly, trnsfectnts positively selected using WR were propgted without WR to enrich for prsites tht hd lost the episoml KO plsmid. Therefter, WR pressure ws repplied to select for prsites with the KO plsmid integrted in the genome. These prsites were plced under weekly cycles with or without 5-fluorocytosine (77 nmol/l) (Sigm-Aldrich) to select for KO prsites tht hd undergone trget gene deletion by double crossover homologous recombintion. 5-fluorocytosine-resistnt prsites were genotyped by Southern nd PCR. Prsite cloning by limiting dilution. Knockout prsites were cloned by limiting dilution in 96-well flt bottom pltes. Prsitemi (using Giemsstined thin smers) nd HCT of cultures were ccurtely determined. Cultures were diluted nd plted t density of.5 prsite per well in 2-μl volume t 2% HCT. Cultures were fed once week with medi contining fresh blood t.5% HCT. Prsitemi in wells ws checked strting 14 dys postinitition of cloning. Southern blotting nd PCR. Southern blotting for the P52/P36 locus ws performed by hybridizing HindIII-ClI-digested genomic DNA from nd KO lines with 3 probe. AflIII or EcoRV-digested DNA ws hybridized with 3 or 5 probe, respectively to chrcterize nd SAP1 KO locus. Digested DNA ws run on.7% Tris-cette-EDTA grose gel t 55 V nd trnsferred to Hybond-N membrne (Amershm, GE Helthcre Life Sciences, Pittsburgh, PA) in 2 SSC overnight t room temperture (RT). DNA ws UV crosslinked to the membrne nd hybridized with digoxygenin-lbeled probes prepred using the DIG kit (Roche Dignostics, Indinpolis, IN). A P52/P36 locus 3 probe ws generted using oligonucleotide primers 5 TATGTACATGTGAAAGTAGCAAAGAC (sense) nd 5 TTCCCTTGTGGGAAATTACAATGAC (ntisense). 3 probe for SAP1 locus ws generted with oligonucleotide primers 5 ATTATGAA CATGACAATACAACTAACG (sense) nd 5 CATATTTATGCTACTGT CAGGGATAG (ntisense). 5 probe for SAP1 locus ws generted using oligonucleotide primers 5 CTAAAATACATAATATACGAAAA AAGTATG (sense) nd 5 TCATATGGCATATAAGATTGTATATCC (ntisense). HsDHFR probe ws generted with primers 5 CCTGGCC ACCGCTCAGGAACG (sense) nd 5 TCCTTGTCACAAATAGTTT AAGATGG (ntisense). Primers 5 CTCAAGAAGAATCCACCCTCATTG (sense) nd 5 CCACACATAACCAGAGGGCAGC (ntisense) were used to mke BSD probe. Primers 5 CAACCTGCAAAATCTAAATTGGT (sm2; sense) nd 5 GTAAATATATAAAACACTACAAATAGTAC (mo41; ntisense) were used for PCR genotyping the P52/P36 locus, nd 5 TCCAAAAATTGACATTCAGAGTTATAG (p353; sense) nd 5 ACACTTATATGTATAGAAATAGTGTTAC (p536; ntisense) for the SAP1 locus. Mosquito infections. Gmetocyte cultures of nd KO lines were propgted in O + humn blood in custom-mde Roswell Prk Memoril Institute medium contining supplemented pooled humn A + serum (Interstte Blood Bnk). Culture medi ws chnged dily nd culture volume ws mintined round 35 ml. Gmetocytogenesis ws checked by Giems-stined thick smers. Exflgelltion ws checked by phse contrst microscopy t 4 mgnifiction beginning 12 dys postinitition of gmetocyte cultures. The cultures were fed to mosquitoes when mjority of the gmetocytes were morphologiclly mture nd vigorously exflgellting. Femle A. stephensi mosquitoes ged 4 7 dys were strved for 1 2 hours nd fed for t lest 3 minutes t 37 C on Budruche membrne feeder pprtus (Joseph Long, Belleville, NJ). Ech cge with 25 3 mosquitoes ws fed with concentrted erythrocytes from the 35 ml gmetocyte culture mixed with n equl volume of fresh red blood cells nd 2 volumes of A + serum. Oocysts prevlence ws determined by microdissecting whole midguts nd exmining them t 1 mgnifiction using phse contrst microscope. Midgut nd slivry glnd sporozoite numbers were determined by microdissecting nd grinding whole midguts nd slivry glnds from mosquitoes on dy 7 nd dys postfeeding with in vitro gmetocyte cultures, respectively, nd counting using hemocytometer. Trversl nd invsion. Trversl nd invsion ssys were performed s previously described. 4 Briefly, HC-4 cells were plted t 3K cells/well in 24-well plte the dy before in Dulbecco's modified Egle's medium (Invitrogen) contining 1% het-inctivted fetl bovine serum (Sigm- Aldrich), penicillin (2 IU/ml)/streptomycin (2 μg/ml) (Corning, Corning, NY), nd 5 ml of mphotericin B (Fungizone) (Corning). Slivry glnd sporozoites of nd KO lines were ctivted in Roswell Prk Memoril Institute medium nd 2% fetl bovine serum t RT for 15 minutes. Sporozoites were trnsferred to HC-4-coted chmber slides t 1K/ well (3:1 rtio of cells:sporozoites). Fluorescein isothiocynte- dextrn (Invitrogen) ws dded to the pproprite wells to ssess totl wounding by sporozoites. The slides were centrifuged t 1,5 rpm for 3 minutes t RT nd incubted t 37 C for hours. Medi ws removed nd cells were fixed nd permebilized in Cytofix/Cytoperm (Becton Dickinson, Frnklin Lkes, NJ). Cells were stined with nti-pfcs mouse monoclonl ntibody (2A1) nd nlyzed by flowcytometry for totl wounding (dextrn + cells) nd totl infection (CS + cells). Motility. Glss coverslips (VWR Interntionl, Rdnor, PA) were precoted with 1 ng/ml 2A1 ntibody in phosphte-buffered sline overnight t RT. Slivry glnd sporozoites of nd KO prsites were ctivted in Roswell Prk Memoril Institute medium contining 2% fetl bovine serum nd llowed to glide on ntibody-coted coverslips t 37 C for 2 hours. Coverslips were fixed in 1% neutrl buffered formlin (Sigm- Aldrich), blocked in 2% bovine serum lbumin phosphte-buffered sline, stined with Alex 488-conjugted 2A1 ntibody, nd mounted in ProLong Gold Antifde Regent (Life Technologies). Motility ws ssessed by detecting CS protein shed in gliding trils on the coverslips. Ptency. FRG-HuHep mice (Yecuris Corportion, Tultin, OR) were injected intrvenously with ech of nd KO slivry glnd sporozoites. On dys 6 nd 7, these mice were injected with 4 μl of wshed O + humn blood t 5% HCT. On dy 7, 3 4 hours following injection of humn blood, the mice were scrificed, nd peripherl blood ws collected by crdic puncture, wshed three times, nd in vitro cultures were set up to determine progression of infection from liver stge to blood stge. Blood stge ptency ws ssessed strting dy 2 postblood collection by Giemsstined thin smers. Moleculr Therpy vol. 22 no. 9 sep

8 Geneticlly Attenuted Plsmodium flciprum Indirect immunofluorescence ssy. After collecting peripherl blood for in vitro culturing, the livers of infected mice were perfused with phosphte-buffered sline, dissected out, wshed with phosphte-buffered sline, nd fixed in 1% neutrl buffered formlin. Fifty micrometer sections were cut using Vibrtome pprtus (Ted Pell, Redding, CA). Immunofluorescence ssys were performed s previously described. 41 Primry ntibodies used were nti-bip (monoclonl) nd nti-acp (polyclonl). RT-PCR ssy. Liver smples were collected in TRIzol (Life Technologies) from FRG mice dy 7 postinfection with intrvenously injected either or (-H) slivry glnd sporozoites. Totl RNA ws extrcted using the Direct-zol MiniPrep Kit (Zymo Reserch, Irvine, CA). cdna synthesis ws performed using the QuntiTect Reverse Trnscription Kit (Qigen). PCR cycling conditions used for mplifiction of cdna were 92 C for 3 seconds for DNA denturtion, 54 C for 3 seconds for primer nneling, nd 62 C for 1 minute for extension (35 cycles). P. flciprum 18S A-type rrna ws mplified using primers 5 CCAGTAGTCATATGCTTGTCTC nd 5 GAAGCGTATTAAAGCGAAAAGC (~7 bp product). Humn ApoA1 ws mplified using primers 5 AGCGTGACCTCCACCTTCAG nd 5 CCTTCACCTCCTCCAGATCCTT (~15 bp product). SUPPLEMENTARY MATERIAL Figure S1. Strtegy for removl of HsDHFR mrker from the SAP1 KO locus in. Figure S2. Immunofluorescence of in vivo liver stge development. Figure S3. RT-PCR of RNA collected on dy 7 of liver stge development. Tble S1. Density of liver stge infection in mice injected with or slivry glnd sporozoites. Tble S2. Sizes of frgments detected by Southern nd PCR products for, nd (-H) clones. Tble S3. Assy for retention of drug resistnce mrkers in nd (-H) clones. ACKNOWLEDGMENTS The reserch ws prtilly funded by Grnt from the Foundtion for the Ntionl Institutes of Helth through the Grnd Chllenges in Globl Helth Inititive (Grnt ID: 1481). This reserch nd development progrm ws lso mde possible by coopertive greement tht ws wrded nd dministered by the US Army Medicl Reserch & Mteriel Commnd nd the Telemedicine & Advnced Technology Reserch Center, t Fort Detrick, MD, under contrct number: W81XWH The uthors declre no conflict of interest. S.H.I.K. is n inventor listed on US Ptent No. 7,22,179, US Ptent No. 7,261,884, nd interntionl ptent ppliction PCT/ US24/4323, ech titled Live Geneticlly Attenuted Mlri Vccine. REFERENCES 1. World Helth Orgniztion. Mlri Control Deprtment. (213). Mlri report Greenwood, BM, Fidock, DA, Kyle, DE, Kppe, SH, Alonso, PL, Collins, FH et l. (28). Mlri: progress, perils, nd prospects for erdiction. J Clin Invest 118: Nussenzweig, RS, Vnderberg, J, Most, H nd Orton, C (1967). Protective immunity produced by the injection of x-irrdited sporozoites of plsmodium berghei. Nture 216: Clyde, DF, Most, H, McCrthy, VC nd Vnderberg, JP (1973). Immuniztion of mn ginst sporozite-induced flciprum mlri. Am J Med Sci 266: Clyde, DF, McCrthy, VC, Miller, RM nd Hornick, RB (1973). Specificity of protection of mn immunized ginst sporozoite-induced flciprum mlri. Am J Med Sci 266: Clyde, DF (1975). Immuniztion of mn ginst flciprum nd vivx mlri by use of ttenuted sporozoites. Am J Trop Med Hyg 24: Rieckmnn, KH, Crson, PE, Beudoin, RL, Cssells, JS nd Sell, KW (1974). Letter: Sporozoite induced immunity in mn ginst n Ethiopin strin of Plsmodium flciprum. Trns R Soc Trop Med Hyg 68: Rieckmnn, KH, Beudoin, RL, Cssells, JS nd Sell, KW (1979). Use of ttenuted sporozoites in the immuniztion of humn volunteers ginst flciprum mlri. Bull World Helth Orgn 57 (suppl. 1): Rieckmnn, KH (199). Humn immuniztion with ttenuted sporozoites. Bull World Helth Orgn 68: Roestenberg, M, McCll, M, Hopmn, J, Wiersm, J, Luty, AJ, vn Gemert, GJ et l. (29). Protection ginst mlri chllenge by sporozoite inocultion. N Engl J Med 361: Seder, RA, Chng, LJ, Enm, ME, Zephir, KL, Srwr, UN, Gordon, IJ et l.; VRC 312 Study Tem. (213). Protection ginst mlri by intrvenous immuniztion with nonreplicting sporozoite vccine. Science 341: Hoffmn, SL, Goh, LM, Luke, TC, Schneider, I, Le, TP, Dooln, DL et l. (22). Protection of humns ginst mlri by immuniztion with rdition-ttenuted Plsmodium flciprum sporozoites. J Infect Dis 185: Mueller, AK, Cmrgo, N, Kiser, K, Andorfer, C, Frevert, U, Mtuschewski, K et l. (25). Plsmodium liver stge developmentl rrest by depletion of protein t the prsite-host interfce. Proc Ntl Acd Sci USA 12: Mueller, AK, Lbied, M, Kppe, SH nd Mtuschewski, K (25). Geneticlly modified Plsmodium prsites s protective experimentl mlri vccine. Nture 433: Trun, AS, Dumpit, RF, Cmrgo, N, Lbied, M, Liu, P, Tkgi, A, et l. (27). Protrcted sterile protection with Plsmodium yoelii pre-erythrocytic geneticlly ttenuted prsite mlri vccines is independent of significnt liver-stge persistence nd is medited by CD8+ T cells. J Infect Dis 196: vn Schijk, BC, Jnse, CJ, vn Gemert, GJ, vn Dijk, MR, Gego, A, Frnetich, JF et l. (28). Gene disruption of Plsmodium flciprum p52 results in ttenution of mlri liver stge development in cultured primry humn heptocytes. PLoS One 3: e Ishino, T, Chinzei, Y nd Yud, M (25). Two proteins with 6-cys motifs re required for mlril prsites to commit to infection of the heptocyte. Mol Microbiol 58: Dourdinh, B, vn Dijk, M, vn Gemert, GJ, Khn, SM, Jnse, CJ, Wters, AP et l. (211). Immuniztion with geneticlly ttenuted P52-deficient Plsmodium berghei sporozoites induces long-lsting effector memory CD8+ T cell response in the liver. J Immune Bsed Ther Vccines 9: Lbied, M, Hrup, A, Dumpit, RF, Coppens, I, Mikoljczk, SA nd Kppe, SH (27). Plsmodium yoelii sporozoites with simultneous deletion of P52 nd P36 re completely ttenuted nd confer sterile immunity ginst infection. Infect Immun 75: VnBuskirk, KM, O Neill, MT, De L Veg, P, Mier, AG, Krzych, U, Willims, J et l. (29). Preerythrocytic, live-ttenuted Plsmodium flciprum vccine cndidtes by design. Proc Ntl Acd Sci USA 16: Spring, M, Murphy, J, Nielsen, R, Dowler, M, Bennett, JW, Zrling, S et l. (213). First-in-humn evlution of geneticlly ttenuted Plsmodium flciprum sporozoites dministered by bite of Anopheles mosquitoes to dult volunteers. Vccine 31: Aly, AS, Mikoljczk, SA, River, HS, Cmrgo, N, Jcobs-Loren, V, Lbied, M et l. (28). Trgeted deletion of SAP1 bolishes the expression of infectivity fctors necessry for successful mlri prsite liver infection. Mol Microbiol 69: Aly, AS, Lindner, SE, McKellr, DC, Peng, X nd Kppe, SH (211). SAP1 is criticl post-trnscriptionl regultor of infectivity in mlri prsite sporozoite stges. Mol Microbiol 79: Silvie, O, Goetz, K nd Mtuschewski, K (28). A sporozoite sprgine-rich protein controls initition of Plsmodium liver stge development. PLoS Pthog 4: e Annour, T, Ploemen, IH, vn Schijk, BC, Sjid, M, Vos, MW, vn Gemert, GJ et l. (212). Assessing the dequcy of ttenution of geneticlly modified mlri prsite vccine cndidtes. Vccine 3: O Neill, MT, Phuong, T, Heler, J, Richrd, D nd Cowmn, AF (211). Gene deletion from Plsmodium flciprum using FLP nd Cre recombinses: implictions for pplied site-specific recombintion. Int J Prsitol 41: Vughn, AM, Mikoljczk, SA, Wilson, EM, Grompe, M, Kushnsky, A, Cmrgo, N et l. (212). Complete Plsmodium flciprum liver-stge development in liverchimeric mice. J Clin Invest 122: Duffy, PE, Shu, T, Akue, A, Milmn, N nd Anderson, C (212). Pre-erythrocytic mlri vccines: identifying the trgets. Expert Rev Vccines 11: Sheehy, SH, Dougls, AD nd Drper, SJ (213). Chllenges of ssessing the clinicl efficcy of sexul blood-stge Plsmodium flciprum mlri vccines. Hum Vccin Immunother 9: Agnndji, ST, Lell, B, Fernndes, JF, Abossolo, BP, Methogo, BG, Kbwende, AL et l.; RTS,S Clinicl Trils Prtnership. (212). A phse 3 tril of RTS,S/AS1 mlri vccine in Africn infnts. N Engl J Med 367: Khn, SM, Jnse, CJ, Kppe, SH nd Mikoljczk, SA (212). Genetic engineering of ttenuted mlri prsites for vccintion. Curr Opin Biotechnol 23: Ploemen, IH, Croes, HJ, vn Gemert, GJ, Wijers-Rouw, M, Hermsen, CC nd Suerwein, RW (212). Plsmodium berghei Δp52&p36 prsites develop independent of prsitophorous vcuole membrne in Huh-7 liver cells. PLoS One 7: e Finney, OC, Keitny, GJ, Smithers, H, Kushnsky, A, Kppe, S nd Wng, R (214). Immuniztion with geneticlly ttenuted P. flciprum prsites induces long-lived ntibodies tht efficiently block heptocyte invsion by sporozoites. Vccine 32: Vughn, AM, O Neill, MT, Trun, AS, Cmrgo, N, Phuong, TM, Aly, AS et l. (29). Type II ftty cid synthesis is essentil only for mlri prsite lte liver stge development. Cell Microbiol 11: Butler, NS, Schmidt, NW, Vughn, AM, Aly, AS, Kppe, SH nd Hrty, JT (211). Superior ntimlril immunity fter vccintion with lte liver stge-rresting geneticlly ttenuted prsites. Cell Host Microbe 9: vol. 22 no. 9 sep. 214

9 Geneticlly Attenuted Plsmodium flciprum 36. vn Schijk, BC, Kumr, TR, Vos, MW, Richmn, A, vn Gemert, GJ, Li, T et l. (214). Type II Ftty Acid Biosynthesis Is Essentil for Plsmodium flciprum Sporozoite Development in the Midgut of Anopheles Mosquitoes. Eukryot Cell 13: Mier, AG, Brks, JA, Wters, AP nd Cowmn, AF (26). Negtive selection using yest cytosine deminse/urcil phosphoribosyl trnsferse in Plsmodium flciprum for trgeted gene deletion by double crossover recombintion. Mol Biochem Prsitol 15: Lmbros, C nd Vnderberg, JP (1979). Synchroniztion of Plsmodium flciprum erythrocytic stges in culture. J Prsitol 65: Wu, Y, Sifri, CD, Lei, HH, Su, XZ nd Wellems, TE (1995). Trnsfection of Plsmodium flciprum within humn red blood cells. Proc Ntl Acd Sci USA 92: Kushnsky, A, Rezkhni, N, Mnn, H nd Kppe, SH (212). Development of quntittive flow cytometry-bsed ssy to ssess infection by Plsmodium flciprum sporozoites. Mol Biochem Prsitol 183: Vughn, AM, Mikoljczk, SA, Cmrgo, N, Lkshmnn, V, Kennedy, M, Lindner, SE et l. (212). A trnsgenic Plsmodium flciprum NF54 strin tht expresses GFP-luciferse throughout the prsite life cycle. Mol Biochem Prsitol 186: Moleculr Therpy vol. 22 no. 9 sep