Disruption of a Plasmodium falciparum patatin-like phospholipase delays male gametocyte exflagellation

An essential process in transmission of the malaria parasite to the Anopheles vector is the conversion of mature gametocytes into gametes within the mosquito gut, where they egress from the red blood cell (RBC). During egress, male gametocytes undergo exflagellation, leading to the formation of eight haploid motile microgametes, while female gametes retain their spherical shape. Gametocyte egress depends on sequential disruption of the parasitophorous vacuole membrane and the host cell membrane. In other life cycle stages of the malaria parasite, phospholipases have been implicated in membrane disruption processes during egress, however their importance for gametocyte egress is relatively unknown. Here, we performed comprehensive functional analyses of six putative phospholipases for their role during development and egress of Plasmodium falciparum gametocytes. We localize two of them, the prodrug activation and resistance esterase (PF3D7_0709700) and the lysophospholipase 1 (PF3D7_1476700), to the parasite plasma membrane. Subsequently, we show that disruption of most of the studied phospholipase genes does neither affect gametocyte development nor egress. The exception is the putative patatin-like phospholipase PF3D7_0924000, whose gene deletion leads to a delay in male gametocyte exflagellation, indicating an important, albeit not essential, role of this enzyme in male gametogenesis.


INTRODUCTION 35
Egress from gametocytes represents an essential process for the transmission of the 36 human malaria parasite Plasmodium falciparum to its mosquito vector. It is induced by 37 ingestion of mature gametocytes by female Anopheles mosquitoes during a blood 38 meal. Within minutes after ingestion, gametocytes become activated in the mosquito 39 midgut by a drop in temperature, a rise in pH, and the presence of the mosquito-derived 40 molecule xanthurenic acid (Bennink et al., 2016). Following their activation, they 41 undergo gametogenesis, whereby each female gametocyte produces a single immotile 42 macrogamete, whereas a male gametocyte produces eight flagella-like microgametes 43 in a process called exflagellation. After their release from host red blood cells (RBCs), 44 macrogemetes and microgametes fuse to form the zygote, which later develops into 45 the motile ookinete (Kuehn and Pradel, 2010). 46 During their intracellular development, gametocytes are surrounded by a 47 parasitophorous vacuole membrane (PVM), which needs to be ruptured before and in 48 addition to the RBC membrane for efficient release of gametes. Membrane rupture is 49 facilitated by the exocytosis of specialized secretory vesicles of the parasites. These 50 include the osmiophilic bodies that release a variety of egress-related proteins into the 51 parasitophorous vacuole lumen (Flieger et al., 2018). In addition, other vesicles are 52 released during egress than contain the perforin-like protein PPLP2, which is 53 necessary for erythrocyte lysis (Deligianni et al., 2013;Wirth et al., 2014). 54 observed for LPL1-mNeonGreen in schizonts as well as free merozoites ( Figure 4B). 144 This suggests that LPL1 localizes to the PPM and that its apparently lower expression 145 levels in asexual blood stages prevented its detection in the LPL1-mScarlet parasite 146 line. 147 148

Loss of individual phospholipases does not impair gametocyte development 149
To probe into the physiological function of the six putative phospholipases for 150 gametocyte development and egress, we next performed targeted gene disruption 151 using the SLI system (Birnbaum et al., 2017) in the NF54/iGP2 background (Boltryk et 152 al., 2021). Correct integration of the respective targeting constructs was confirmed by 153 PCR ( Figure S4). Previously, five of the six phospholipases (PARE,LPL4,PNPLA3,154 PL39,and PL38) were demonstrated to be non-essential for asexual blood stage 155 development using the same targeting constructs (Burda et al., 2021). We thus only 156 analyzed asexual blood stage proliferation of LPL1-knockout (KO) parasites. This, 157 however, did not reveal an impaired growth in the presence or absence of 2 mM choline 158 as compared to wildtype (WT) parasites, suggesting that LPL1 plays a non-essential 159 or redundant function in asexual blood stage parasites ( Figure S5). 160 To investigate the function of the selected phospholipases for gametocyte 161 development, we induced gametocyte commitment as previously described (Boltryk et 162 rates, calculated by the ratio of gametocytemia on day 11 and day 5 of gametocyte 164 development, were not significantly different between WT and KO-parasite lines 165 ( Figure S6). In line with this, quantification of gametocyte stages over the 11-day 166 course of gametocytogenesis of all phospholipase-KO lines revealed no major defect 167 in gametocyte maturation ( Figure 5). Taken together, these data suggest that the six 168 analyzed putative phospholipases are dispensable for gametocyte development. Next, we analyzed gamete egress from mature stage V gametocytes. During this 172 process, male gametocytes undergo a striking transformation resulting in the formation 173 of eight haploid motile microgametes, while female gametes maintain their spherical 174 shape after egress from the RBC (Kuehn and Pradel, 2010). To study whether 175 gametocyte egress is affected in the KO parasite lines, we used an established egress 176 assay based on live-cell fluorescence microscopy ( Figure 6A,B) (Suareź-Cortés et al., 177 2014). For this, the RBC membrane of RBCs infected with mature stage V 178 gametocytes was stained using the live-cell dye iFluor555-wheat germ agglutinin 179 (WGA) before activation. Subsequently, gametocytes were activated for 20 minutes 180 and parasite morphology (falciform shape versus spherical shape) as well as the WGA 181 staining pattern (WGA-positive versus WGA-negative) were investigated (Suareź-182 Cortés et al., 2014). As expected, the vast majority of non-activated gametocytes 183 showed a falciform shape with strong WGA staining of the erythrocyte membrane, 184 while after activation most gametocytes displayed a round morphology and were WGA-185 negative, indicating successful egress from the host RBC ( Figure 6C-H). Hereby, no 186 major differences between WT and KO parasite lines were observed, indicating that 187 all of the six tested phospholipases are dispensable for gamete egress. To further 188 analyze egress, we also separately quantified exflagellation rates of male gametocytes 189 on four subsequent days from day 11 until day 14 of gametocyte development. To that 190 end, we activated gametocytes for 12 minutes and subsequently determined the 191 percentage of exflagellating gametocytes within the next 5 minutes.  LPL1-KO gametocytes showed similar exflagellation rates to WT parasites, while 193 LPL4-KO, PL39-KO and PL38-KO showed significantly higher exflagellation rates 194 compared to WT parasites (Figure 7). In line with the WGA staining-based egress 195 assay, we can thus conclude that none of these five putative phospholipases plays an 196 essential role in gametocyte exflagellation. Remarkably, PNPLA3-KO gametocytes 197 showed a significant reduction in exflagellation within the initial five minutes of Phospholipases were shown to have important functions for asexual intraerythrocytic 204 development and egress of malaria blood and liver stage parasites (Burda et al., 2015;205 Burda et al., 2021;Ramaprasad et al., 2023). We here aimed to expand the functional 206 characterization of these enzymes and investigated the role of six putative 207 phospholipases for gametocytogenesis and gametocyte egress. 208 Two putative phospholipases analyzed in this study were PARE and LPL1. PARE was 209 previously reported to have esterase activity and to activate esterified pepstatin, a 210 peptidyl inhibitor of malarial aspartyl proteases, although its localization in the parasite 211 had not yet been determined (Istvan et al., 2017). By endogenous tagging, we now 212 localized PARE and LPL1 to the PPM in asexual and sexual blood stages of the 213 parasite. In addition to this, we showed that disruption of PARE and LPL1 does not 214 affect asexual blood stage development (this study and (Burda et al., 2021)), arguing 215 and functional data regarding LPL1 differ to a previous study that localized LPL1 to a 217 multi-vesicular neutral lipid-rich body next to the food vacuole in asexual blood stages 218 and that proposed an essential role of LPL1 in neutral lipid synthesis contributing to 219 hemozoin formation in the parasite (Asad et al., 2021). While the reasons for the 220 observed differences in LPL1 localization remain unclear, the absence of a growth 221 defect in asexual blood stages due to LPL1 disruption could be related to the different 222 Two other putative phospholipases that we analyzed in this study were PL39 and 228 PL38, which both contain predicted signal peptides and phospholipase C/P1 nuclease 229 domains. Enzymes containing these domains are typically involved in phosphate ester 230 hydrolysis of lipids or nucleic acids (Coleman, 1992;Desai and Shankar, 2003). The 231 genomic proximity of pl39 and pl38 and the existence of only one single orthologue in 232 the rodent malaria parasite P. berghei at this genomic location (according to 233 PlasmoDB.org) indicates that these loci might have arisen from gene duplication. 234 However, as PL39 and PL38 display only 46 % sequence identity (according to 235 ClustalW), they may have diverged to serve distinct functions over the course of 236 evolution. In line with this, PL39 localized to the food vacuole, while PL38 displayed a 237 possible apical localization. Interestingly, the Toxoplasma gondii patatin-like 238 phospholipase TgPL3 (TGME49_305140) has been localized to the apical pole where 239 its phospholipase activity is necessary for rhoptry secretion (Wilson et al., 2020). We 240 speculate that the putative phospholipase activity of PL38 might similarly contribute to 241 invasion. However, gene deletion did not affect asexual blood stage proliferation, 242 suggesting an unimpaired invasion process of PL38-deficient parasites (Burda et al., 243 2021). 244 To explore the role of the six selected putative phospholipase candidates for 245 gametocytogenesis and gametocyte egress, we performed targeted gene disruption of 246 the individual genes. We then analyzed sexual blood stage development of the 247 resulting mutant parasite lines. This revealed that gametocyte survival and maturation 248 were not affected by the individual disruption of the six selected putative 249 phospholipases, suggesting that, similar to our results obtained in asexual blood  Since activation of gametocytes for the WGA assay was done for 20 minutes, while 12 261 minutes of gametocyte activation were used for the exflagellation assay, the extended 262 time for gametocyte activation could explain why the PNPLA3-associated phenotype 263 in exflagellation was not observed in the WGA-based egress assay. What specific role 264 PNPLA3 might play in the exflagellation process remains to be elucidated. interfering with the release of egress-associated vesicles (Singh et al., 2019). It is thus 270 reasonable to hypothesize that a similar function might also be performed by PNPLA3. 271 Given that conditional inactivation of PNPLA1 only resulted in a partial egress 272 phenotype, both enzymes might even work together during egress, a scenario that 273 should be explored in future studies by generating mutants that lack both enzymes. Gametocytes of NF54/iGP2-derived parasites were basically induced as previously 335 described (Boltryk et al., 2021). For this, synchronous ring stage cultures (2-3 % 336 parasitemia) were washed in medium to remove remaining GlcN and plated at 2.5-5% 337 hematocrit in culture medium without GlcN (= day -1 of gametocyte development). After 338 reinvasion of committed rings (= day 1 of gametocyte development), asexual parasites 339 were depleted using 50 mM N-acetyl-D-glucosamine (GlcNAc) in the culture medium. 340 GlcNAc was added to the cell culture medium until day 6. From day 7 until day 14, 341 gametocyte cultures were fed with culture medium containing 0.25 % Albumax + 5 % 342 human serum. Gametocytes were cultured in the presence of GlcN from day 1 onwards 343 and gametocyte cultures were fed daily. 344 Induction of 3D7-derived gametocytes was performed as previously described 345 (Filarsky et al., 2018). To this aim, synchronous ring stage cultures were washed twice 346 and then cultivated in medium without choline (= day -1 of gametocyte development). 347 From day 1 of gametocyte development until day 6, asexual parasites were depleted 348 using 50 mM GlcNAc in the culture medium (now again containing choline). From day 349 7 until day 14, gametocyte cultures were fed with culture medium containing 0.25 % 350 Gametocyte stages and gametocytemia were monitored using Giemsa-stained thin 352 blood smears. 353 354

Fluorescence microscopy 355
For staining of nuclei, parasites were incubated with 0.45 µg/ml Hoechst in culture 356 medium for 30 min at 37°C. Images were acquired on a Leica D6B fluorescence 357 microscope, equipped with a Leica DFC9000 GT camera and a Leica Plan Apochromat 358 100x/1.4 oil objective. Image processing was performed using ImageJ and 359 representative images were adjusted for brightness and contrast. 360 361

Growth analysis of mutant parasites 362
For asexual blood stage growth analysis of parasites, schizont stage parasites were 363 isolated by Percoll enrichment and incubated with uninfected RBCs (5% hematocrit) 364 for 3 hours to allow rupture and invasion. Parasites were then treated with 5% sorbitol 365 to remove residual unruptured schizonts, leading to a synchronous ring stage culture 366 with a 3-hour window. These were allowed to mature to trophozoites for one day and 367 parasitemia was determined by flow cytometry and adjusted to exactly 0.1% starting 368 parasitemia in a 2 ml dish. Medium was changed daily and growth of the parasite lines 369 was assessed by flow cytometry over three intraerythrocytic cycles when parasites 370 were in the trophozoite stage. As a reference, WT parasites were included in each 371 assay. 372

Flow cytometry 374
Flow cytometry-based analysis of parasite lines was performed essentially as 375 described previously (Malleret et al., 2011). In brief, 20 μl resuspended parasite culture 376 were incubated with dihydroethidium (5 μg/ml, Cayman) and SYBR Green I dye (0.25 377 protected from light. Samples were analyzed on a ACEA NovoCyte flow cytometer. 379 RBCs were gated based on their forward and side scatter parameters. For every 380 sample, 100,000 events were recorded and parasitemia was determined based on 381 SYBR Green I fluorescence. 382 383

Imaging-based gametocyte egress assay 384
On day 14 of gametocyte development, the egress of mature male and female 385 gametocytes was tested as described previously (Suareź-Cortés et al., 2014) with 386 minor modifications. In brief, parasite-infected RBC were stained with iFluor555-WGA 387 (Biomol, stock 2 mg/ml, final concentration 5 µg/ml) and 0.45 µg/ml Hoechst at 37°C. 388 After 30 min, the samples were washed in prewarmed Ringer solution. 5 µL were 389 placed on a slide and immediately covered with a cover slip. Imaging of this non-390 activated control was performed for 20 min. To study egress, gametocytes were 391 activated by incubation in ookinete medium for 20 min at 26 °C before imaging for 20 392 min at room temperature. Ookinete medium was prepared by supplementing culture 393 medium (without serum or albumax) with 100 µM Xanthurenic acid and adjusting the 394 pH to 8.0, followed by addition of 20 % human serum. 395

Gametocyte exflagellation assay 397
From day 11 until day 14 of gametocytogenesis, exflagellation was determined daily in 398 technical duplicates. 250 µl of resuspended gametocyte culture were spun down in an 399 Eppendorf tube at 800 x g for 1 min. The supernatant was discarded and the pellet 400               iFluor555-WGA (red) and Hoechst (blue). DIC, differential interference contrast. Scale 579      background. The SLI targeting strategy and the primer localizations are depicted in Figure 8A. GFP was replaced by mScarlet or mNeonGreen-glmS (mNeon), respectively. Instead of the GFP rev primer the primer Neo rev was used. Primers are listed in Table 8 Figure 8A. Primers are listed in Table 8. WL, wildtype locus; 5'int, 5' integration; 3'int, 3' integration, TGD, targeted gene deletion. (B) Growth assay of LPL1-KO parasites over three cycles. Parasites were diluted 1:10 after the second cycle to prevent overgrowth. Mean ± SD. n=3. Unpaired students t-test (not significant).   Figure 8A. Primers are listed in Table 8. WL, wildtype locus; 5'int, 5' integration; 3'int, 3' integration, TGD, targeted gene deletion. (B) Growth assay of LPL1-KO parasites over three cycles. Parasites were diluted 1:10 after the second cycle to prevent overgrowth. Mean ± SD. n=3. Unpaired students t-test (not significant).