Chimeric Plasmodium falciparum parasites expressing Plasmodium vivax circumsporozoite protein fail to produce salivary gland sporozoites

Plasmodium falciparum and Plasmodium berghei parasites and in vitro cultivation of P. falciparum blood stages

Plasmodium falciparum parasites from the NF54 strain [29] were obtained from the Radboud University Medical Center (Nijmegen, The Netherlands). Parasites were cultured following the standard conditions in RPMI-1640 culture medium supplemented with l-glutamine and 25 mM HEPES (Gibco Life Technologies), 50 mg/l hypoxanthine (Sigma). Culture medium was supplemented with 10% human serum and 0.225% NaHCO₃. Parasites were cultured at a 5% haematocrit under 4% O₂, 3% CO₂ and 93% N₂ gas-conditions at 75 rpm at 37 °C in a semi-automated culture system in 10 ml flasks (Infers HT Multitron and Watson Marlow 520U). Fresh human serum and human red blood cells (RBC) were obtained from the Dutch National Blood Bank (Sanquin Amsterdam, the Netherlands; permission granted from donors for the use of blood products for malaria research and microbiology test for safety). RBC of different donors were pooled every 2 weeks, washed twice in serum free RPMI-1640 and suspended in complete culture medium to 50% haematocrit. Human serum of different donors were pooled every 4–6 months and stored at − 20 °C until required.

In addition, P. falciparum gametocytes cultures were generated using standard culture conditions (see above) with some modifications [30]. Briefly, parasites from asexual stage cultures were diluted to a final parasitaemia of 0.5% and cultures were followed during 14 days without refreshing RBC. After 9 days these cultures were treated with 50 mM of N-acetyl-D-glucosamine (Sigma) to kill asexual stages and to enrich for gametocytes. At day 14 the cultures were analysed for mature, stage V, gametocytes.

Four different mutant lines of the rodent parasite P. berghei were used that have been previously reported. (i) A transgenic reference line of P. berghei ANKA, expressing the fusion protein GFP-Luciferase (line 676m1cl1; PbΔp230p; RMgm-29; https://www.pberghei.eu) [31]; (ii) A mutant that expresses P. vivax CSP (VK210 allele). In this mutant the pbcsp gene has been replaced with the pvcsp-vk210 gene (line 2196cl1; RMgm-4136; https://www.pberghei.eu); [27]; (iii) A mutant that expresses PvCSP (VK247 allele). In this mutant the pbcsp gene has been replaced with the pvcsp-vk247 gene (line 2199cl1; RMgm-4137; https://www.pberghei.eu; [27];

Generation and selection of the chimeric lines pf-pvcsp(vk210) and pf-pvcsp(vk247)

In order to create pf-pvcsp(vk210) and pf-pvcsp(vk247), the previously described pLf0019 construct, containing the cas9 gene was used [32] and 2 different sgRNA-donor DNA containing plasmids, pLf0042 (targeting PfCSP and containing the pvcsp-vk210 gene) and pLf0043 (containing the pvcsp-vk247). The pLf0019 construct contains a blasticidin (BSD) drug-selectable marker cassette and both sgRNA-donor DNA constructs (pLf0042 and pLf0043) contain a hdhfr drug-selectable marker cassette for selection with WR99210. To generate pLf0042 and pLf0043, plasmid pLf0033 (see Additional file 1) was modified by introducing two homology regions targeting Pfcsp (PF3D7_0304600). Homology region 1 (HR1) was amplified using primers P1/P2 and homology region 2 (HR2) with P3/P4 from P. falciparum NF54 genomic DNA (see Additional file 2 for primer details). HR1 was cloned in pLf0033 using restriction sites kpnI/EcoRI and HR2 using EcoRI/AatII. The pvcsp alleles pvcsp-vk210 (GenBank accession number P08677; Belem strain) and pvcsp-vk247 (GenBank accession number M69059.1; Papua New Guinea strain) were amplified from existing plasmids PbG01-PvCSP-vk247 and PbG01-PvCSP-vk210 [27] using the primers P5/P6 and cloned into pL0033 containing the HR using restriction sites EcoRV/EcoRI, resulting in intermediate plasmids pLf0040 and pLf0041 (pvcsp-vk210 and pvcsp-vk247; see Additional file 1). An additional plasmid, AS301 (see Additional file 1) was used to clone the guide sgRNA (AS301-sgRNA2) specific for pfcsp. The sgRNA sequence was identified using the Protospacer software (alphaversion; https://sourceforge.net/projects/protospacerwb/files/Release/) and was amplified using the primers P7/P8. This sgRNA was selected based on the best-off targets hits score throughout the genome given by Protospacer and the total number of mismatches of the sgRNA with respect to the PAM site. A 20 bp guide sgRNA sequence, flanked on both sides by a 15 bp DNA sequence necessary for In Fusion cloning (HD Cloning Kit; Clontech), was annealed and used to replace the BtgZI adaptor as previously described [33]. The construct was then digested with BlnI and NruI to evaluate the successful cloning of the sgRNA and later confirmed by Sanger sequencing using primers P9/P10. Finally, the AS301-sgRNA2 was digested with EcoRV/ApaI and cloned into the vectors pLf0040 and Plf0041 using the restriction sites StuI/ApaI resulting in the final constructs pLf0042 and pLf0043 (see Additional file 1).

Plasmids for transfection were isolated from 250 ml cultures of Escherichia coli XL10-Gold Ultracompetent Cells (Stratagene) by maxi-pep (using HiSpeed^® Plasmid Maxi Kit (Qiagen^®)) to generate 25–50 µg of DNA used per transfection. Transfections of P. falciparum NF54 parasites were performed using ring stage parasites obtained from cultures with a parasitaemia of 6–15% that were synchronized by 5% d-sorbitol treatment 2 days before transfection [34]. Infected RBC were pelleted by centrifugation (1150g, 5 min) and 300 µl of the pelleted cells were transferred to a 0.2 cm cuvette and mixed with ~ 50 µg of each circular plasmid (Cas9 and sgRNA/Donor DNA constructs) in 100 µl cytomix [35]. Electroporation was performed with a single pulse (310 V and 950 µF) in the Biorad Gene Pulser Xcell electroporator (including CE- and PC module) and cells were immediately transferred in a 10 ml culture flask and cultures were maintained under standard conditions in the semi-automated culture system (see above). Selection of transformed parasites was performed by applying ‘double’ positive selection 24 h after transfection using the drugs WR99210 (2.6 nM) and BSD (5 µg/ml). For WR99210 100 µl of a stock solution (2.6 µM) was added to 100 ml complete culture medium resulting in a final concentration of 2.6 nM. To prepare the WR99210 stock-solution WR99210 was dissolved in DMSO (100 mM). For BSD 50 µl of a stock solution (10 mg/ml) was added to 100 ml complete culture medium resulting in a 5 µg/ml final concentration. Drug pressure in the cultures was maintained until thin blood-smears were parasite-positive (usually after 14–26 days). Positive selection will select for the parasites that were transfected successfully with both plasmids (Cas9 and sgRNA/Donor constructs). Subsequently, both drugs were removed from the cultures for 2–4 days, followed by applying negative selection by addition of 5-Fluorocytosine (5-FC; 130 µl of a stock solution (0.77 mM) in 100 ml complete medium with a final concentration of 1 µM; [36]) in order to eliminate parasites that retained the crRNA/Donor construct as episomal plasmid and enrich for the transfected parasites containing the donor DNA integrated into the genome. Negative drug pressure in the cultures was maintained until thin blood-smears were parasite-positive (usually after 7 days). After negative selection infected RBC (iRBC) were harvested from cultures with a parasitaemia of 4–10% for genotyping by diagnostic PCR and Southern blot analysis. Subsequently, selected parasites were cloned by limiting dilution.

Generation and selection of the PfΔcsp line

In order to create the PfΔcsp line, a new plasmid (pLf0070: that contain both the crRNA and the Cas9-expression cassette, was modified in order to introduce sgRNAs against Pfcsp. This plasmid, kindly obtained from Dr. Marcus Lee (Wellcome Trust Sanger Institute, Wellcome Genome Campus, Hinxton, Cambridgeshire, UK) is based on plasmid pDC2-cam-Cas9-U6.2-hdhfr [37] with a smaller U6 cassete (693 bp) and with the Cas9 gene harmonized to P. falciparum. Two plasmids with two different sgRNAs (026 and 012) were generated. Briefly, pLf0070 was digested with BbsI and sgRNA026 was cloned using the primers P11/P12 and sgRNA012 using the primers P13/P14. The primers (100 µM each primer) were phosphorylated with T4 polynucleotide kinase (10 Units per reaction) during 30 min at 37 °C, followed by an annealing program of 5 min incubation at 94 °C and a ramp down to 25 °C at 5 °C per min, and subsequently ligated into the vector using T4 ligase (5 units) resulting in plasmid pLf0071 and plasmid pLf0072 (see Additional file 1). A second DNA donor plasmid was generated by replacing the Pvcsp-vk210 gene of the pLf0040 construct (see above) by an mCherry expression cassette obtained from an intermediate plasmid pLf0055 (see Additional file 1). In this cassette mCherry is under control of the promoter region of gapdh (GeneID PF3D7_1462800) and the 3′UTR of histidine-rich protein II(GeneID PF3D7_0831800). The complete mCherry expression cassette was removed from pLf0055 by digestion with the EcoRI/NruI and cloned into pLf0040 digested with EcoRI/EcoRV resulting in the DNA donor vector pLf0083 (see Additional file 1). This donor DNA construct has a drug selectable marker cassette containing a fusion of the positive selectable marker hdhfr and the negative selectable marker yfcu (yeast cytosine deaminase/uridyl phosphoribosyl transferase).

Transfections of P. falciparum NF54 parasites was performed by spontaneous plasmid uptake from plasmid-loaded red blood cells cultured under static conditions [38]. Briefly, 300 µl of pelleted, uninfected RBC were transferred to a 0.2 cm cuvette and mixed with 50 µg of both sgRNA constructs (25 µg of pLf0071 and 25 µg of pLf0072) and 50 µg of the donor construct (pLf0083) suspended in 200 µl of cytomix. Electroporation was performed as described in the previous section. After electroporation of the uninfected RBC, iRBC containing P. falciparum NF54 parasites were added to a concentration of 0.1%. Selection of transformed parasites was performed when cultures reached a parasitaemia of 3% (after approx. 3 days) with 100 µl of WR992010 (2.6 nM) during a period of 6 days. Subsequently the drug was removed and parasites were harvested at 0.8% of parasitaemia for mCherry fluorescence microscopy analysis to determine the ratio of wild type and mutant parasites present in the population. Parasites were collected from cultures that contained > 80 mCherry-positive parasites (at a 4–10% parasitaemia) for genotyping by diagnostic PCR and Southern blot analysis and for cloning (see next section).

Cloning of transfected P. falciparum parasites

Based on the PCR confirmation of the integration, the transfected parasites were cloned by the method of limiting dilution as previously described [39] with minor modifications. Briefly, infected RBC from cultures with a 4 to 10% parasitaemia were diluted with uninfected RBC to 10⁵ iRBC/100 µl in 2 ml culture medium (1% haematocrit and 20% serum). Serial dilutions were then performed with uninfected RBC in complete medium (1% haematocrit and 20% serum) and cultured in a total volume of 100 µl incubated in 96 well plates, resulting in 8 rows with the following numbers of iRBC per well: 100, 10, 5, 2.5, 1.25, 0.6, 0.3, 0.15. Plates were incubated in a Candle Jar at 37 °C and culture medium was changed every other day. Every 5 days RBC were added resulting in an increase of the haematocrit from 1 to 5%. Between days 10–14 samples were collected for thick smear analysis from the rows with the highest numbers of iRBC/well; 50 µl medium was removed and from the remaining culture 5 µl was used directly for preparing thick smears. At day 21 thick smears were made from all rows. Clones were selected from dilutions/row with less than 30% of the wells parasite positive. These clones were transferred in 10 ml culture flasks at 5% haematocrit under standard culture conditions (see previous sections) in the semi-automated culture system for collection of parasites for further genotype and phenotype analyses.

Genotyping of the pf-pvcsp(vk210), pf-pvcsp(vk247) and pfΔcsp lines

For genotyping of the chimeric pf-pvcsp lines and the pfΔcsp line diagnostic PCR and Southern blot analysis of digested DNA were performed from material isolated from iRBC obtained from 10 ml cultures (parasitaemia 3–10%), pelleted by centrifugation (1150g; 5 min.). RBC were then lysed with 5–10 ml of cold (4 °C) erythrocyte lysis buffer (10 × stock solution 1.5 M NH₄Cl, 0.1 M KHCO₃, 0.01 M Na₂EDTA; pH 7.4; [34]) and parasites were treated with RNAse and proteinase-K before DNA isolation by standard phenol–chloroform methods. Correct integration of the donor constructs was analysed by standard and long-range PCR (LR-PCR). In brief, for the chimeric pf-pvcsp lines integration of the pvcsp cassettes was confirmed by LR-PCR using the primers P15/P16 (and analysed by EcoRV digestion). The PCR-amplified product was cloned in a TopoTA vector for sequencing (see Additional file 3 for details of the primers and Additional file 2 for sequence data). The LR-PCR fragments were amplified using KOD Hot start polymerase following standard conditions with an annealing temperature of 53.5 °C for 15 s and an elongation step of 68 °C for 9 min. For the PfΔcsp line, 5′-integration PCR was performed using the primers P15/P19 and to confirm the presence of the mCherry gene PCR was performed with the primers P20/P21. The PCR fragments were amplified using Go-taq^® DNA polymerase (Promega) following standard conditions with an annealing temperature of 56 °C for 20 s and a elongation step of 72 °C for 4 min. All other PCR settings were according to manufacturer’s instructions.

Southern blot analysis for the chimeric pf-pvcsp lines was performed with genomic DNA digested with AvaII (4 h at 37 °C) in order to confirm integration of the replacement of pfcsp by the pvcsp genes Fig. 1c). Digested DNA was hybridized with probes targeting the Pfcsp homology region 2 (HR2), amplified from NF54 genomic DNA by PCR using the primers P3/P4 and a second probe targeting ampicillin (Amp) gene, obtained by digestion of the intermediate plasmid pLf0040 with AatII/PvuI (550 bp). For Southern blot analysis of PfΔcsp, genomic DNA was digested with AvaII and XhoI (4 h at 37 °C) and digested DNA was hybridized with the same probes used with the Pf-Pvcsp lines (HR2 and Amp probes see Additional file 7C).

Phenotype analysis of P. falciparum parasites: blood stages, gametocytes, oocysts and sporozoites

The growth rate of asexual blood stages of the pf-pvcsp and pfΔcsp lines was monitored in 10 ml cultures maintained in the semi-automated culture system under standard culture conditions (see above). Briefly, a 0.5% parasitaemia culture was established in complete culture medium at a haematocrit of 5%. Medium was changed twice daily and the culture maintained for a period of 5 days without refreshing RBC. For determination of the course of parasitaemia, triplicate samples of 100 µl were collected daily from all cultures and cells pelleted by centrifugation (9485g; 30 s) and stained with Giemsa. mCherry expression of in PfΔcsp blood stages was analysed by standard fluorescence microscopy. In brief, 200 µl samples were collected from 10 ml cultures with a parasitaemia between 4 and 10% and stained with the DNA-specific dye Hoechst-33342 by adding 4 µl of a 500 µM stock-solution (final concentration 10 µM) for 20 min at 37 °C. Subsequently, a 5 µl drop was placed on a microscopic slide (mounted under a cover slip) and fluorescence in live iRBC analysed using a Leica fluorescence MDR microscope (100× magnification). Pictures were recorded with a DC500 digital camera microscope using Leica LAS X software and with the following exposure times: mCherry 0.6 s; Hoechst-33342 0.136 s; bright field 0.62 s (1× gain).

Gametocyte production by the pf-pvcsp and pfΔcsp lines was analysed in gametocyte cultures, established as described in the previous sections. To activate gametocytes for exflagellation 20 µl samples of the P. falciparum stage V gametocyte cultures at day 14 were diluted 1:1 with FCS at room temperature. Gametes and exflagellation centres were examined and quantified 10–20 min after activation using a Bürker cell counter. For analysis of mosquito stages (oocysts and sporozoites) of the chimeric pf-pvcsp lines, An. stephensi were infected using the standard membrane feeding assay (SMFA) [40, 41]. Oocysts were analysed between day 8 and 14 for sporozoite production and the percentage of degenerated oocyst determined. Oocysts were qualified as degenerated based on the following criteria: no sporozoite formation visible and oocyst cytoplasm vacuolated. Salivary gland sporozoites were counted at day 14 and 21 post feeding. For counting sporozoites, salivary glands from 30 to 60 mosquitoes were dissected and homogenized using a grinder in 100 µl of RPMI pH 7.2 and sporozoites were analysed in a Bürker cell counter using phase-contrast microscopy.

Analysis of PvCSP expression in oocyst-derived sporozoites of the pf-pvcsp(vk210) and pf-pvcsp(vk247) lines

To analyse CSP expression in oocyst-derived sporozoites of the pf-pvcsp lines by immunofluorescence microscopy, midguts from 30 to 60 An. stephensi mosquitoes were collected in an eppendorf tube at day 10 after feeding in RPMI-1640 medium containing 3% BSA. Midguts were mechanically crushed using a grinder and centrifuged with low speed (62g) for 3 min at 4 °C. Subsequently, the supernatant containing oocyst-derived sporozoites was collected for fluorescence microscopy and samples (20 µl) were placed on a 8-well black cell-line diagnostic microscope slide (Thermo Scientific), dried for 10 min, and fixed with 4% paraformaldehyde for 30 min at room temperature. After fixation the slides were washed three times with 1× PBS and permeabilized with 20 µl of 0.5% triton in 1× PBS and then blocked with 10% of FCS in 1× PBS for 1 h. Fixed cells were washed with 1× PBS and incubated with monoclonal antibodies against PvCSP(vk210) (mouse, anti-PvCSP-VK210 MAb (MR4); 1:200 dilution of 109 µg/ml stock solution [27]) PvCSP-(vk247) (mouse, anti-PvCSP-VK247 MAb (MR4); 1:200 dilution of 125 µg/ml [27]), PfCSP (mouse, anti-PfCSP (210 A) MAb(MR4); 1:200 dilution of 8 µg/ml stock solution REF) and PfHSP70 (rabbit, anti-PfHSP70; 1:200 dilution of 100 µg/ml stock solution StressMarqBiosciences) for 1 h at room temperature. Subsequently, cells were rinsed three times with 1× PBS and incubated with the secondary antibodies Alexa FLuor^®488/594-conjugated chicken anti-mouse and anti-rabbit (Invitrogen Detection technologies at 1:200). Finally, the cells were washed again three times with 1× PBS and stained with the DNA-specific dye Hoechst-33342 at a final concentration of 10 µM. Fixed cells were covered with 1–2 drops of an anti-fading agent (Vectashield), and a coverslip placed on the cells and sealed with nail polish. Stained cells were analysed for fluorescence using a Leica fluorescence MDR microscope (100× magnification). Pictures were recorded with a DC500 digital camera microscope using Leica LAS X software with the following exposure times: Alexa 488: 0.7 s; Alexa 594: 0.6 s Hoechst 0.136 s; bright field 0.62 s (1× gain).

Phenotype analysis of Plasmodium berghei parasites: oocysts, sporozoites and sporozoite infectivity

Feeding of An. stephensi mosquitoes with P. berghei parasites, determination of oocyst production and sporozoite collection were performed as described [27]. Determination of parasite liver load by in vivo imaging and determination of the prepatent period in mice after intravenously injection of 1000 salivary gland sporozoites was performed as described [27].

Statistics

Data were analysed using GraphPad Prism software package 5.04 (GraphPad Software, Inc). Significance difference analyses between WT, pf-pvcsp and the pfΔcsp lines was performed using the unpaired Student’s t-test.

Generation of two chimeric P. falciparum parasites lines expressing PvCSP-210 or PvCSP-247

Using CRISPR/Cas9 gene editing, two chimeric P. falciparum parasites lines were created, pf-pvcsp(vk210) and pf-pvcsp(vk247), where the P. falciparum csp gene has been replaced by either one of the two major P. vivax csp alleles, VK210 and VK247. A previously described Cas9 construct (pLf0019), containing the Cas9 expression cassette with a blasticidin (BSD) drug-selectable marker cassette [32], was used in combination with two sgRNA donor-DNA containing plasmids, pLf0042 and pLf0043. These constructs are used to target the pfcsp gene locus and as ‘DNA-donor’ sequences they replace pfcsp with either pvcsp-vk210 or the pvcsp-vk247 full-length gene coding sequences (Fig. 1). The coding sequences of pvcsp-vk210 (GenBank accession number P08677; Belem strain) and pvcsp-vk247 (GenBank accession number M69059.1; Papua New Guinea strain) were amplified from existing plasmids P. bergheiG01-PvCSP-vk247 (pL1943) and P. bergheiG01-PvCSP-vk210 (pL1942) [27]. The two homology regions targeting pfcsp (PF3D7_0304600) were amplified from genomic P. falciparum DNA (NF54 strain). Both constructs contain a hdhfr-yfcu drug-selectable marker cassette.

Transfections of P. falciparum NF54 parasites were performed using synchronized ring stage parasites that were transfected with ~ 50 µg of each circular plasmid (Cas9 and sgRNA/donor-DNA constructs; see Additional file 1) and selection of transformed parasites containing both plasmids (Cas9 and sgRNA/donor-DNA constructs) was performed by applying ‘double’ positive selection with the drugs WR99210 and BSD, until parasites were detectable by thin blood-smear analysis (between day 14 and 26 post transfection). Subsequently, parasites were cultured for 2–4 days without drugs, followed by applying negative selection to eliminate parasites that retained the transfection constructs (i.e. donor-DNA) as episomal plasmids and to enrich for transfected parasites in which the donor-DNA construct has integrated into the parasite genome. Genotyping of selected parasite populations by long-range PCR revealed that both pvcsp(vk210) and pvcsp(vk247) cassettes had integrated into the P. falciparum genomes (Fig. 1b) and Southern blot analysis of cloned lines confirmed correct integration of the constructs (Fig. 1c). Phenotype analyses, as described below, were performed using pf-pvcsp(vk210) clone 7 and pf-pvcsp(vk247) clone 5. Sequence analysis of the long-range PCR products confirmed the correct sequence of pvcsp genes and replacement of the pfcsp gene in both pf-pvcsp lines (see Additional file 3).

Pf-PvCSP210 and Pf-PvCSP210 parasites form oocysts but salivary gland sporozoites are absent in An. stephensi mosquitoes

Anopheles stephensi mosquitoes were fed with either WT gametocytes or gametocytes of the pf-pvcsp lines using the standard membrane feeding assay and the number of oocysts in mosquito midguts was determined at day 10 post infection and the presence of sporozoites in the haemocoel and salivary glands was analysed at day 14 and day 21. Dissection of WT and pf-pvcsp infected mosquitoes revealed that all lines produced comparable numbers of oocysts; however, no sporozoites were detected in salivary glands in mosquitoes infected with either of the pf-pvcsp lines (Table 1). Moreover, no sporozoites from the pf-pvcsp lines were found in haemocoel fluid after mosquito dissection; in contrast, when WT infected mosquitoes were dissected many sporozoites were observed in haemocoel fluid.

These observations indicate that the chimeric parasites expressing PvCSP are unable to produce sporozoites that are competent in invading salivary glands.

pf-pvcsp lines produce sporozoites in oocysts but most oocysts degenerate before sporozoite formation and release of sporozoites

For two Plasmodium species it has been shown that csp deletion mutants can form oocysts; however by light-microscopy, these oocysts are highly vacuolated and do not show signs of sporozoite formation [42, 43]. Therefore, oocyst development of pf-pvcsp was analysed in greater detail by light microscopy and PvCSP expression was examined by immunofluorescence microscopy. After feeding with both pf-pvcsp lines, oocysts were readily detected at day 10, in which sporozoite formation occurred (Fig. 2a). However, analysis of oocysts between day 10 and 14 showed that most oocysts started to degenerate before full maturation as shown by the absence of clear sporozoite formation and the presence of large vacuoles in the cytoplasm of maturing oocysts (see Additional file 5). In WT-infected mosquitoes, 30% of the oocysts were characterized as degenerate at day 10, whereas in Pf-PvCSP(vk210) and Pf-PvCSP(vk247) infected mosquitoes 52 and 87% degenerate oocysts were countedy (Fig. 2b). No spontaneous release of sporozoites from mature pf-pvcsp oocysts was detected and no sporozoites could be found free in the haemocoel or in salivary glands, even up to day 21 after feeding. Only through the application of force to oocysts in dissected midguts could mechanically liberated oocyst-derived pf-pvcsp sporozoites be recovered (Fig. 2a).

Chimeric sporozoites of both pf-pvcsp(vk210) and pf-pvcsp(vk247) express PvCSP

To analyse expression of PvCSP in pf-pvcsp sporozoites, oocyst-derived sporozoites were collected by differential centrifugation of extracted, mechanically crushed, infected midguts. These sporozoites were stained with antibodies specific for PvCSP-VK210, PvCSP-VK247 and PfCSP. Sporozoites of pf-pvcsp(vk247) and pf-pvcsp(vk210) only reacted with their cognate anti-PvCSP-VK247 or anti-PvCSP-VK210 antibodies and WT sporozoites only with anti-PfCSP antibodies (Fig. 3; Additional file 6). These results indicate that the corresponding PvCSP is expressed in developing oocysts and oocyst-derived sporozoites of the pf-pvcsp lines and that the failure of formation of fully competent sporozoites is not due to the absence of PvCSP expression.

A P. falciparum NF54 mutant that lacks expression of PfCSP forms oocysts but sporozoite formation is absent

To investigate whether the formation of sporozoites in chimeric pf-pvcsp oocysts was due to partial complementation of PfCSP by PvCSP, a P. falciparum mutant lacking CSP expression was generated. This mutant (PfΔcsp) was generated by disrupting the pfcsp gene by CRISPR/Cas9 gene editing using a construct (pL0083) that contained an mCherry reporter cassette (under control of the constitutive gapdh promoter; see Additional file 7A) flanked by pfcsp targeting sequences. This construct is designed to replace the pfcsp gene with the mCherry-expression cassette and contains a hdhfr-yfcu drug-selectable marker cassette. Transfection of P. falciparum NF54 parasites was performed by spontaneous plasmid uptake from plasmid-loaded red blood cells cultured under static conditions. Uninfected RBC were mixed with 50 µg of two sgRNA/Cas9 constructs (pLf0071 and pLf0072) and 50 µg of the donor-DNA construct (pLf0083). After electroporation, these uninfected cells were mixed with iRBC containing P. falciparum NF54 parasites and selection of transformed parasites was performed with the drug WR99210 during a period of 6 days. Subsequently the drug was removed and parasites were harvested at 0.6–0.8% parasitaemia for mCherry fluorescence microscopy analysis to determine the ratio of WT and mutant parasites expressing mCherry. Parasites were collected from cultures that contained > 80% mCherry-positive parasites (at a 4–10% parasitaemia). Genotyping of selected and cloned mCherry-positive parasites by diagnostic PCR and Southern blot analysis showed integration of the mCherry-expression cassette into the pfcsp gene (see Additional file 7B, C).

The growth of asexual blood stages in cultures and gametocyte production of the PfΔcsp line was comparable to that of WT P. falciparum NF54 parasites (see Additional file 4; Table 1). Gametocyte cultures of the PfΔcsp lines produced WT-like numbers of mature, stage V, gametocytes of both sexes. Mature male gametocytes of both lines underwent exflagellation upon activation and were able to form male gametes (Table 1). Anopheles stephensi mosquitoes were fed with WT gametocytes and PfΔcsp gametocytes using standard membrane feeding. The number of oocysts was determined at day 10 post feeding and the presence of sporozoites in salivary glands analysed at day 14 and 21 after feeding. PfΔcsp parasites were able to produce oocysts in the mosquitoes but no sporozoites were detected in salivary glands after feeding (Table 1). Moreover, sporozoite formation was not detected in oocysts (3 exp.; 30 mosquitoes per experiment). All oocysts degenerated before full maturation as shown both by the absence of sporozoite formation and the presence of large vacuoles in the cytoplasm of the oocysts (see Fig. 2; Additional file 5). This oocyst phenotype is comparable to the phenotype observed in P. berghei and Plasmodium knowlesi mutants lacking CSP expression [42, 43]. The absence of sporozoite formation in PfΔcsp oocysts is in support of partial complementation of PfCSP by PvCSP in the pf-pvcsp lines where sporozoites are formed in a fraction of the oocysts.

Heterologous CSP replacement and expression in different Plasmodium species