Genetic disruption of lsa3 and validation in P. falciparum blood-stages.

A. Schematic of the 176 kDa LSA3 protein conserved across human-infective malaria species. Regions between residues 45-64 and 1433-1456 are predicted to be a hydrophobic signal anchor (SA) and transmembrane domain (TM) (orange), respectively; residues 89-93 contain a predicted PEXEL motif (red) and residues 908-1154 have homology to the substrate binding site from the chaperone DnaK (blue). LSA3-C antibody (Ab) binds the C-terminal region (yellow). B. Schematic of LSA3 genetic disruption strategy in P. falciparum NF54. The endogenous coding sequence of LSA3 was disrupted by replacement with the DHFR selectable marker cassette flanked by homology targets (HT). C. Immunoblot of uninfected human red blood cells (Ui RBCs) or parasite blood-stage lysates of NF54 (parental control) and LSA3 KO clones E2 and D8, probed with LSA3-C antibodies. D. Immunofluorescence microscopy (IFA) of NF54 parental parasites following methanol/acetone fixation and staining with LSA3-C or anti-EXP2 (PVM) antibodies. IFA of ΔLSA3 parasites using LSA3-C and anti-EXP2 antibodies confirms loss of exported LSA3 in infected erythrocytes and non-specific binding to one or more proteins inside the parasites.

LSA3 is trafficked into the PVM and the erythrocyte during blood-stage development.

A. Immunofluorescence microscopy of NF54 parental parasites following fixation with methanol/acetone (top) and paraformaldehyde/glutaraldehyde (bottom) staining with anti-LSA3 (LSA3-C) and anti MAHRP2 (Maurer’s cleft tethers) antibodies (top). IFA of NF54 parasites following PFA fixation and staining with either anti-LSA3 (LSA3-C), anti-EXP2 or anti-REX3 antibodies (below). B. Proteinase K (PK) protection assays to localize LSA3. Erythrocytes infected with NF54 parental parasites were enriched by Percoll centrifugation and permeabilized with either equinatoxin II (EQT; selectively lyses the erythrocyte membrane), 0.03% saponin (selectively lyses the erythrocyte membrane and PVM) or 0.1% Triton-X-100 (lyses all membranes). The pellet was separated from the supernatant (supe) and both fractions were incubated with PBS containing Proteinase K (PK) at 37 °C. Protease inhibitor cocktail and PMSF was added to all samples followed by reducing sample buffer before immunoblotting with LSA3-C, anti-REX3 (exported control) or anti-aldolase (aldo; internal protein control for parasite membrane integrity) antibodies.

The PEXEL of LSA3 is processed by plasmepsin V.

A. Schematic of LSA3-GFP mini gene expressed from the Bip gene promoter following integration into the redundant p230 locus. Two minigenes fused to green fluorescent protein (GFP) were generated containing endogenous LSA3 residues 1-112 encoding the N-terminal signal anchor (SA) and either a native PEXEL motif (red) or with key PEXEL residues mutated to alanine (RLE>A). Predicted molecular weights following possible putative cleavage events are depicted. B. Immunoblots of blood-stage lysates from parental NF54 control and transgenic NF54 expressing LSA3 minigenes probed with anti-GFP antibodies. Left blow shows accumulation of uncleaved mLSA3 RLE>A-GFP compared to mLSA3-GFP and NF54 parental controls (lanes 1-3). Right blot shows accumulation of uncleaved mLSA3-GFP in the presence of plasmepsin V inhibitor WEHI-600 for 5 hr compared to DMSO vehicle control. The uncleaved band was a similar size as uncleaved mLSA3 RLE>A-GFP in the left blot. Predicted sizes: full length minigene: 39 kDa; PEXEL processing: 29 kDa; GFP core in the food vacuole following secretion of reporter: 27 kDa. C. Live fluorescence imaging of transgenic NF54 blood-stage parasites expressing either mLSA3-GFP (secreted to PV; top) or mLSA3 RLE>A-GFP (retained within parasite, likely the ER; bottom). Scale bars, 5 µm. D. Schematic of the strategy used to endogenously tag LSA3 with GFP at the C-terminus in P. falciparum NF54 (top). Bottom panel shows live fluorescence imaging of LSA3-GFP expressed from the endogenous locus in transgenic NF54 parasites with the protein localized at the PV and PVM (see E). Scale bars, 5 µm. E. Proteinase K (PK) protection assays to localize endogenous LSA3-GFP. Erythrocytes infected with NF54 LSA3-GFP were enriched by Percoll centrifugation and permeabilized with either equinatoxin II (EQT; selectively lyses the erythrocyte and Maurer’s cleft membranes), 0.03% saponin (selectively lyses the erythrocyte, Maurer’s cleft, and PVM membranes) or 0.1% Triton-X-100 (lyses all membranes). The pellet was separated from the supernatant (supe) and both fractions were incubated with PBS containing Proteinase K (PK) at 37 °C. Protease inhibitor cocktail and PMSF was added to all samples followed by reducing sample buffer before immunoblotting with anti-GFP (LSA3-GFP), anti-REX3 (exported control) or anti-aldolase (aldo; internal protein control for parasite membrane integrity) antibodies. LSA3-GFP is indicated in green while GFP core in the food vacuole is indicated.

LSA3-deficiency does not inhibit gametocytogenesis or oocyst or sporozoite development nor sporozoite infectivity.

A. Percentage of stage V gametocytes of NF54 parental and ΔLSA3 clones D8 and E2. Red bars indicate the mean. Data was compared by ANOVA using the Kruskal Wallis test. B. Oocyst counts of NF54 and ΔLSA3 clones D8 and E2 per mosquito midgut 7 days post bloodmeal. Data include the mean (number shown and red bar) from three independent transmission experiments. Data comparing ΔLSA3 mutant clones to NF54 parent performed by ANOVA using the Kruskal Wallis test, except experiment 3 (Mann-Whitney test). Prevalence of mosquito infection is shown below each graph, comparing NF54 to ΔLSA3 clones by chi-square analysis. C. Salivary gland sporozoite (SG SPZ) counts per mosquito at 17 days postbloodmeal. Data is mean compared ΔLSA3 mutant clones to NF54 parents by ANOVA using the Kruskal Wallis test. D. Cell traversal enumerated as percentage of FITC-dextran+ HC-04 hepatocytes after incubation with freshy dissected NF54 and ΔLSA3 clone E2 sporozoites from mosquito salivary glands. E. Percentage of HC-04 cells with CSP+ intracellular parasites at 5 and 18 hours of incubation with NF54 parent and ΔLSA3 clone E2 sporozoites. Data was analyzed by ANOVA using the Kruskal Wallis test. All data in the figure is mean ± SEM from n=3 (A to D) or n=6 (E) independent experiments; n.s., not significant (P>0.05).

LSA3-C localizes to the PVM in P. falciparum-infected hepatocytes.

A. Liver sections from humanized mice infected with P. falciparum NF54 sporozoites on day 5 post infection. Each panel shown is of the same liver-stage parasite at different cross-sections of the Z-stack. Sections were stained with anti-EXP1 antibodies that label the PVM and LSA3-C antibodies. LSA3-C localized in puncta on PVM extensions or beyond the PVM are indicated (yellow arrows). B. Liver section from humanized mice infected with P. falciparum NF54 sporozoites on day 6 post infection stained with anti-EXP2 antibodies that label the PVM and LSA3-C antibodies. For all images, parasite and host DNA was labelled with DAPI nuclear stain; scale bars, 10 µm.

LSA3 is required for P. falciparum liver-stage development in humanized mice.

A. Quantification of P. falciparum genomes per million human hepatocytes on day 5 postinfection from four co-infected humanized mice with chimeric human livers, each receiving 5×105 ΔLSA3 clone E2 and 5×105 parental NF54 sporozoites mixed for a total of 1×106 sporozoites via a single IV injection per mouse, by qRT-PCR. Values from each sample were normalized to a series of pretested standard curves of each amplicon and are shown as mean ± SD, statistically analysed by paired t-test. Each symbol corresponds to the same coinfected mouse. B. Percent of total parasite liver load on day 5 postinfection in each humanized mouse co-infected with equal numbers of ΔLSA3 E2 and parental NF54 sporozoites from (A), shows reduced fitness of LSA3-deficient parasites (i.e. less than 50%). Percent differences of the mean in each mouse is shown on the right ± SD. Each symbol corresponds to the same coinfected mouse. Means were compared by paired t-test. C. Human chimerism in each humanized mouse (21%, 22%, 19% and 18%) quantified by qRT-PCR using oligonucleotides specific to human and mouse PTGER2.

Genotyping P. falciparum NF54 and ΔLSA3 clones.

Diagnostic PCRs (MO545/aw560) confirm correct double cross-over integration of the ΔLSA3 construct in P. falciparum NF54 clones. Mutants used in this study are shown in blue and molecular weight sizes in base-pairs (bp) are shown.

The LSA3-C binding domain is not localized on the infected erythrocyte surface.

Erythrocytes infected with ΔLSA3 and NF54 parental parasites were enriched by Percoll centrifugation, with the former being included as a specificity control for the LSA3-C antibody. NF54-infected erythrocytes were incubated with PBS to maintain cells as intact. The pellet was then separated from the supernatant (supe) fraction by centrifugation, and both were incubated at 37 °C with Proteinase K (PK) in PBS. Protease inhibitor cocktail and PMSF were added to all samples followed by reducing sample buffer to stop the reactions. Samples were separated by SDS-PAGE and probed with LSA3-C or anti-REX3 (exported protein control) antibodies. The C-terminal portion of LSA3 recognized by LSA3-C antibodies was not present on the erythrocyte surface as it remained resistant to Proteinase K, like the exported REX3 control as expected. The cross-reactive internal protein recognized by LSA3-C is visible in all lanes.